The Oxford Handbook of Comparative Health Law 9780190846756, 9780190846770, 9780190846763, 9780190846787, 0190846755

Citation preview

T h e Ox f o r d H a n d b o o k o f

C OM PA R AT I V E H E A LT H L AW

The Oxford Handbook of

COMPARATIVE HEALTH LAW Edited by

DAVID ORENTLICHER and

TAMARA K. HERVEY

1

1 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America. © Oxford University Press 2022 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. Library of Congress Cataloging-­in-­Publication Data Names: Orentlicher, David, 1955– editor. | Hervey, Tamara K., editor. Title: The Oxford handbook of comparative health law / edited by David Orentlicher and Tamara K. Hervey. Description: New York: Oxford University Press, 2022 Identifiers: LCCN 2021010387 (print) | ISBN 9780190846756 (hardback) | LCCN 2021010388 (ebook) | ISBN 9780190846770 (epub) | ISBN 9780190846763 (ebook) | ISBN 9780190846787 (ebook other) Subjects: LCSH: Public health laws—European Union countries. | Public health laws—Great Britain. | Public health laws—United States. | Medical care—Law and legislation—European Union countries. | Medical care—Law and legislation—Great Britain. | Medical care—Law and legislation—United States. | Comparative law. | Public policy (Law) Classification: LCC K3570 .O94 2021 (print) | LCC K3570 (ebook) | DDC 344.03/21—dc23 LC record available at https://lccn.loc.gov/2021010387 LC ebook record available at https://lccn.loc.gov/2021010388 1 3 5 7 9 8 6 4 2 Printed by Integrated Books International, United States of America

Contents

About the Editorsxi About the Contributorsxiii

1. Editors’ Introduction to The Oxford Handbook of Comparative Health Law1 Tamara K. Hervey and David Orentlicher

PA RT I   PA R A DIG M S OF H E A LT HC A R E SYST E M S , L AW, A N D R E G U L AT ION 2. Paradigms of Healthcare Systems, Law, and Regulation: A Transatlantic Conversation Alceste Santuari and William Sage

19

PA RT I I   P R E V E N T I N G I L L H E A LT H 3. Introduction to Public Health Law Wendy E. Parmet, Markus Frischhut, Amandine Garde, and Brigit Toebes 4. Communicable and Other Infectious Diseases: The EU Perspective Markus Frischhut

69

77

5. Communicable Disease Law in the United States Wendy E. Parmet

97

6. US Law Relating to Noncommunicable Diseases Wendy E. Parmet

119

7. The Lack of Coherence in the European Union’s Approach to Noncommunicable Disease Prevention Amandine Garde

143

vi   contents

8. Socioeconomic Health Inequalities in Europe: The Role of Law and Human Rights Brigit Toebes 9. Social Determinants in the United States Wendy E. Parmet

171 189

PA RT I I I   T R E AT I N G I L L H E A LT H Access to the Healthcare System 10. Introduction to Access to Healthcare André den Exter, Mark A. Hall, Allison K. Hoffman, and Keith Syrett

209

11. The American Pathology of Inequitable Access to Medical Care Allison K. Hoffman and Mark A. Hall

213

12. Access to Healthcare in Europe André den Exter and Keith Syrett

243

Regulation of Health Care Services, Facilities, and Transactions 13. Introduction to Regulation of Professionals and Facilities Deirdre Madden and Isaac D. Buck

275

14. Regulation of Professionals and Facilities in Europe Deirdre Madden

277

15. Regulation of Professionals and Facilities in the United States Isaac D. Buck

293

16. Introduction to Antitrust and the Provision of Healthcare in the United States and European Union: Common Challenges Thomas Greaney and Okeoghene Odudu 17. Health Care Services and EU Competition Law Okeoghene Odudu 18. Healthcare Antitrust in the United States: The Intersection of Regulation and Competition Thomas Greaney 19. Introduction to Fraud and Abuse Law Tracey A. Elliott and Joan H. Krause

311 317

349 365

contents   vii

20. Fraud and Abuse Law in the United States Joan H. Krause

375

21. Healthcare Fraud and Abuse in Europe Tracey A. Elliott

393

The Treatment Relationship: Confidentiality, Consent, and Conflicts of Interest 22. Privacy and Integrity of Medical Information Sharona Hoffman and Jean Herveg

417

23. Patient Autonomy, Capacity, and Consent: Children Jessica Berg and Emma Cave

459

24. Patient Autonomy, Capacity, and Consent: Vulnerable Adults Mary Donnelly and Jessica Berg

477

25. Introduction to Provider Conflicts of Interest in Health Care  Sunčana Roksandić and Richard S. Saver

505

26. Provider Conflicts of Interest in US Healthcare: The Intractable Regulatory Challenge Richard S. Saver

513

27. Provider Conflicts of Interest in European Healthcare: Trustworthy Physicians Sunčana Roksandić

537

Medical Malpractice 28. Medical Liability: Comparing Compensation Systems Karl Harald Søvig and Barry Furrow

559

PA RT I V   R E G U L AT I N G T H E DE V E L OP M E N T A N D U SE OF M E DIC A L T R E AT M E N T S Human Experimentation and Research 29. Introduction to Research with Human Participants Carl H. Coleman 30. Research with Human Participants in the European Union David Shaw and David Townend

607 613

viii   contents

31. Research with Human Participants in the United States Carl H. Coleman

633

32. Biobanks Graeme T. Laurie and Carl H. Coleman

657

Pharmaceutical and Medical Devices Law 33. Introduction to Medical Products Law Erika Lietzan, Aurélie Mahalatchimy, and Patricia J. Zettler

685

34. Regulating Medicines in the United States Erika Lietzan and Patricia J. Zettler

691

35. Regulating Medicines in the European Union Aurélie Mahalatchimy

721

36. Regulating Medical Devices in the United States Patricia J. Zettler and Erika Lietzan

755

37. Regulating Medical Devices in the European Union Aurélie Mahalatchimy

779

Control, Use, and Allocation of Body Parts: Organs, Human Tissue, Blood 38. Property in Human Body Parts: An Old Legal Question for a New Technological Age Justine Pila 39. Organ Transplantation Joaquin Cayon-­De Las Cuevas and David Orentlicher 40. Introduction to Incomplete Commodification and Its Creeping Counterpart Natalie Ram and Stéphanie Hennette Vauchez

809 841

875

41. Incomplete Commodification in American Law Natalie Ram

879

42. Of Markets and Principles: The European Perspective Stéphanie Hennette Vauchez

897

contents   ix

Ethical and Legal Implications of Advances in Genetics 43. Introduction to Genetics and the Law Maxwell Mehlman, Mette Hartlev, and Sonia Suter

919

44. Genetics and the Law: United States Maxwell Mehlman and Sonia Suter

923

45. Genetics and the Law: Europe Mette Hartlev

949

46. Genetics and the Law: Conclusion Maxwell Mehlman, Mette Hartlev, and Sonia Suter

981

PA RT V   H E A LT HC A R E AT T H E B E G I N N I N G A N D E N D OF L I F E Reproductive Rights 47. Introduction to the Right to Procreate and Assisted Reproductive Technologies I. Glenn Cohen and Emily Jackson 48. The Right to Procreate and Assisted Reproductive Technologies in Europe Emily Jackson

989

993

49. The Right(s) to Procreate and Assisted Reproductive Technologies in the United States I. Glenn Cohen

1009

50. Introduction to the Right to Avoid Procreation and the Regulation of Pregnancy Janne Rothmar Herrmann and Elizabeth Sepper

1027

51. The Right to Avoid Procreation and the Regulation of Pregnancy: A European Perspective Janne Rothmar Herrmann

1033

52. The Right to Avoid Procreation and the Regulation of Pregnancy: A US Perspective Elizabeth Sepper

1051

x   contents

The Law of Death and Dying 53. Decisions at the End of Life David Orentlicher and Judit Sándor

1073

Index

1109

About the Editors

David Orentlicher  is the Judge Jack and Lulu Lehman Professor at University of Nevada, Las Vegas (UNLV) William S. Boyd School of Law and director of the UNLV Health Law  Program. Widely recognized for his expertise in health law and constitutional law, Dr. O. has testified before Congress, had his scholarship cited by the US Supreme Court, and has served on many national, state, and local commissions. A  graduate of Harvard Medical School and Harvard Law School, Dr. O.  is author of Matters of Life and Death (Princeton University Press, 2001) and co-­author of Health Care Law and Ethics (Wolters Kluwer, 2018) now in its ninth edition. He has published numerous articles and essays on a wide range of topics, including healthcare reform, physician aid in dying, reproductive decisions, affirmative action, and presidential power, and his work has  appeared in leading professional journals such as the  New England Journal of Medicine and the Journal of the American Medical Association (JAMA), as well as in the New York Times, Time, USA Today, CNN Opinion, the Chicago Tribune, and other major newspapers. In addition to his academic background, Dr. O. brings important hands-­ on experience. He previously directed the American Medical Association’s Division of Medical Ethics, and he has worked on healthcare reform and other issues as a member of the Nevada legislature and the Indiana legislature. Tamara K. Hervey, LLB (Hons) (Glasgow), PhD (Sheffield), Fellow of the Academy of Social Sciences, Principal Fellow of the Higher Education Academy, and Honorary Fellow of the Faculty of Public Health, is Jean Monnet Professor of EU Law at City Law School, City University of London. Hervey researches, teaches, and writes on European Union health law, comparative health law and policy, equality law, interfaces between biosciences and (European) law, social rights, legal research methodologies, and legal pedagogy. She is author of 18 books and more than 100 other publications. One of the first to the field, Hervey has been writing on EU health law since the 1990s. She works with a large network of academics across Europe and in North America. Hervey served as Specialist Adviser to the House of Commons Health and Social Care Committee’s Brexit Inquiries. She also served the House of Lords EU Home Affairs Sub-­Committee Inquiry into Brexit and cross-­border healthcare. She has given keynote presentations about EU law, trade, health, and Brexit, to organizations including the Royal Society of Medicine, British Medical Association, Health Services Research UK, and the Scottish Lawyers European Group. She also gave Expert evidence to the Joint Oireachtas Committee on Health, Republic of Ireland, on the legal implications of Brexit in the health sector.

About the Contributors

Jessica Berg  is Professor and Dean at Case Western Reserve University. She has numerous publications in the areas of bioethics, law, and public health, and is first author of Informed Consent: Legal Theory and Clinical Practice, 2nd edition (Oxford University Press, 2001). Isaac D. Buck is Associate Professor of Law at the University of Tennessee in Knoxville. He is the author of a number of articles that critically examine American health law and policy, with a particular focus on finance and insurance, pharmaceutical drug pricing, and fraud and abuse. Emma Cave  is a Professor of Healthcare Law at Durham University, UK. Cave is a member of the UK fertility regulator, the Human Fertilisation and Embryology Authority, and publishes widely in healthcare law with recent articles on informed consent, making treatment decisions for young children, and adolescent decision-­ making. She is co-­author of Medicine, Patients and the Law (Manchester University Press) with Margaret Brazier. Joaquin Cayon-­De Las Cuevas is Associate Professor of Health Law and Director of the Research Group on Health Law and Bioethics at IDIVAL/University of Cantabria. His recent publications include (with Katherine Fierlbeck) Health Law and Policy from East to West: Analytical Perspectives and Comparative Case Studies (Thomson Reuters, 2020). I. Glenn Cohen is the author or co-­author of more than 150 articles in journals like Science and JAMA and the author, co-­author, editor, or co-­editor of 15 books. Cohen is one of the world’s leading scholars in bioethics and health law. He is deputy Dean and Professor at Harvard Law School. Carl H. Coleman is Professor of Law at Seton Hall University School of Law. He is the lead author of The Ethics and Regulation of Research with Human Subjects, 2nd edition (Carolina Academic Press, 2015) and a past member of the Department of Health and Human Services (DHHS) Secretary’s Advisory Committee on Human Research Protections. André den Exter is Associate Professor of health law at Erasmus University Rotterdam, the Netherlands. André holds several visiting (associate) professorships in International and European health law (Danylo Medical University Lviv, Ukraine, and the National School of Public Health, Portugal). He is the editor of European Health Law (Maklu Press, 2017).

xiv   about the contributors Mary Donnelly is Professor of Law at University College Cork, Ireland. She is author of Healthcare Decision-­ Making and the Law: Autonomy, Capacity and the Limits of Liberalism (Cambridge University Press, 2010) and joint editor of Ethical and Legal Debates in Irish Healthcare (Manchester University Press, 2016) and has published extensively in capacity law. Tracey A. Elliott is lecturer in Health Care Law at the University of Leicester. Formerly a practising barrister, she is co-­author of Clarkson & Keating: Criminal Law, 10th edition (Sweet & Maxwell, 2020) and has published in journals such as the Medical Law Review and Clinical Ethics. Markus Frischhut is Jean Monnet Professor, Chair of EU Values & DIGitalization for Our CommuNITY (DIGNITY), at Management Center Innsbruck, teaching and researching in the fields of EU law, EU health law, EU values, as well as EU law and ethics. His recent publications include The Ethical Spirit of EU Law (Springer, 2019). Barry Furrow is Professor of Law and Director of the Health Law Program at the Kline School of Law at Drexel University. He is the lead author of Health Law: Cases, Materials and Problems, 8th edition (West Academic, 2018), and he has published articles in journals such as the University of Pennsylvania Law Review and others. Amandine Garde is Professor of Law and Founding Director of the Law and NCD Unit at the University of Liverpool, UK. Her latest edited collection (with Joshua Curtis and Olivier De Schutter) is Ending Childhood Obesity: A Challenge at the Crossroads of International Economic and Human Rights Law (Elgar, 2020). She is Editor of Elgar’s Health and the Law series. Thomas Greaney, JD, is Visiting Professor of Law at UC Hastings College of Law and Distinguished Senior Fellow with the UC Hastings/UCSF Consortium on Law, Science and Health Policy. He is also Chester A. Myers Emeritus Professor of Law at Saint Louis University School of Law. He has written more than 60 scholarly articles and chapters and is co-­author of the nation’s leading health law casebook, Health Law: Cases, Materials and Problems, 8th edition (West Academic, 2018) and a treatise on health law, Health Law, 3rd edition (West Academic, 2014). Mark A. Hall is Professor of Law and Public Health at Wake Forest University, where he focuses on healthcare finance and regulation. His publications include, along with David Orentlicher and other co-­authors, a half-­dozen editions of Health Care Law and Ethics (Wolters Kluwer), the original casebook in the field. Professors Hall and Orentlicher also are coauthors of Health Care Law and Ethics in a Nutshell, 4th edition (West Academic, 2018). Mette Hartlev is professor of Health Law at the Faculty of Law, University of Copenhagen. Her recent publications include contributions to Informed Consent and Health: A Global Analysis (Edward Elgar, 2020) and Research Handbook on Global Health Law (Edward Elgar, 2018), and she has published in journals such as European Journal of Health Law, Journal of Community Genetics, and European Journal of Public Health.

about the contributors   xv Stéphanie Hennette Vauchez is Professor of Law at Université Paris Nanterre and a senior member of the Institut universitaire de France. Recent publications in the field of (bio)medical law include “SPUC v. Grogan (ECJ, 1991): Rereading the Case and Retelling the Story of Reproductive Rights in Europe” in EU Law Stories: Contextual and Critical Stories of European Jurisprudence (Cambridge University Press, 2017) and “Gender Equality in the European Convention on Human Rights,” in Frontiers of Gender Equality (University of Pennsylvania Press, 2021). Janne Rothmar Herrmann  is Professor of Health Law and Technology at the University of Copenhagen. Her latest monograph is the co-­authored The Cryopolitics of Reproduction: A New Scandinavian Ice Age (Emerald, 2020). She has published in journals such as European Journal of Health Law, Medical Law International, and European Human Rights Law Review. Jean Herveg is Director of the LIS Department, CRIDS, University of Namur, Belgium. He is a member of the Bar of Brussels (registered expert in IT law and health law). He is the author of more than 180 publications, reports, and communications on privacy, data protection, and e-­health. Sharona Hoffman is a Professor of Law and Bioethics and Co-­Director of the Law-­ Medicine Center at Case Western Reserve University. She has authored numerous articles and book chapters and has published two books, including Electronic Health Records and Medical Big Data: Law & Policy (Cambridge University Press, 2016). Allison K. Hoffman is Professor of Law at the University of Pennsylvania Carey Law School and Senior Fellow at Penn’s Leonard Davis Institute of Health Economics. She is co-­editor of The Oxford Handbook of US Health Law and co-­author of the forthcoming third edition of The Politics of Medicare. Emily Jackson is a Professor of Law at the London School of Economics and Political Science. She sits on the BMA Medical Ethics Committee and is the author of Medical Law, 5th edition (Oxford University Press, 2019). Joan H. Krause is Dan K. Moore Distinguished Professor of Law at the University of North Carolina School of Law. Her recent publications include (with Leslie C. Griffin) Practicing Bioethics Law (Foundation Press, 2015) and numerous articles on healthcare fraud and abuse and women’s health issues. Graeme  T.  Laurie  is Professorial Fellow in the School of Law at the University of Edinburgh and holds a Wellcome Senior Investigator Award in Medical Humanities. He is the lead editor of the Cambridge Handbook on Health Research Regulation (Cambridge University Press, 2021). Erika Lietzan is William H. Pittman Professor of Law and Timothy J. Heinsz Professor of Law at the University of Missouri. Her recent publications include “Distorted Drug Patents” (with Kristina Acri) in the Washington Law Review and “Access Before Evidence and the Cost of FDA’s New Drug Authority” in the University of Richmond Law Review.

xvi   about the contributors Deirdre Madden is Professor of Law at University College Cork, Ireland, and author of Medicine, Ethics and the Law (Bloomsbury, 2016). She is Deputy Chair of the Board of the Health Service Executive and has served on professional regulatory bodies in Ireland for many years. Aurélie Mahalatchimy is a legal scholar, permanent researcher at the National Centre for Scientific Research (CNRS), UMR 7318 International, Comparative and European laws, Aix-­Marseille University, France. She has published in journals such as Journal of Law and Society, Medical Law Review, International Journal of Bioethics, Law & European Affairs, and Nature Biotechnology. Maxwell Mehlman is Distinguished University Professor, Arthur E. Petersilge Professor of Law, Director of the Law-­ Medicine Center, Case Western Reserve University School of Law, and Professor of Biomedical Ethics, CWRU School of Medicine. He is a­ co-­author of Genetics: Ethics, Law and Policy, now in its fifth edition (West Academic). Okeoghene Odudu is Herchel Smith University Senior Lecturer and Deputy Director of the Centre for European Legal Studies in Competition Law at the Faculty of Law in the University of Cambridge. He serves on the editorial boards of the European Competition Journal, the Cambridge Yearbook of European Legal Studies, and Concurrences. Wendy E. Parmet is Matthews University Distinguished Professor of Law and Professor of Public Policy and Urban Affairs at Northeastern University. Her most recent book is The Health of Newcomers, Immigration, Health Policy and the Case for Global Solidarity, co-­authored with Patricia Illingworth (New York University Press, 2017). Justine Pila  is Professor of Law and co-­Chair of the Health, Law and Emerging Technologies (HeLEX) Research Group at the University of Oxford. Her books include European Intellectual Property Law (Oxford University Press, 2016, 2019, with P.  L.  C.  Torremans) and The Oxford Handbook of Intellectual Property Law (Oxford University Press, 2018, edited with R. C. Dreyfuss). Natalie Ram is Professor of Law at the University of Maryland Carey School of Law. She is a 2021 Greenwall Faculty Scholar in Bioethics and has published in journals such as Science, Nature Biotechnology, Harvard Law Review, and Virginia Law Review. Sunčana Roksandić is Assistant Professor of Criminal Law, Medical Law, and Bioethics at the Faculty of Law, University of Zagreb, and a head of the Croatian UNESCO Chair Unit (Haifa). Her recent publications include Prosecuting Serious Economic Crimes as International Crimes: A New Mandate for the ICC (Duncker & Humblot, 2017) and the chapter “Deprivation of Access to Essential Medicine and Medical Care: A Crime Against Humanity?” in Bio Law and International Criminal Law (Brill Nijhoff, 2020). William Sage is Professor of Law and Medicine at the University of Texas at Austin. A member of the US National Academy of Medicine, he has written more than 200 publications in law, medicine, and health policy, and he is a co-­editor of The Oxford Handbook of US Health Law (2017).

about the contributors   xvii Judit Sándor is Professor at the Faculty of Political Science, Legal Studies and Gender Studies of the Central European University, in Budapest and in Vienna. In 2004–2005, she was the Chief of the Bioethics Section at UNESCO. Since 2005, she chairs the Center for Ethics and Law in Biomedicine. Alceste Santuari is Professor of Law and Economics at the University of Bologna. His recent publications include Manuale di diritto dell’economia degli enti non profit (Bologna, 2020) and “Social Enterprises in the European Union and the Italian 2017 Social Enterprises Reform Act: A Legal Framework for Other Jurisdictions?,” in Third Sector Review (2020). Richard  S.  Saver  is Arch  T.  Allen Distinguished Professor at the University of North Carolina School of Law and Professor (Secondary Appointment) at UNC School of Medicine. He has numerous law and medical articles, and he edited (with S. Johnson, J. Krause, and R. Fretwell Wilson) Health Law & Bioethics: Cases in Context (Aspen, 2009). Elizabeth Sepper is Professor of Law at the University of Texas at Austin. Her recent works have appeared in the Columbia Law Review, Northwestern University Law Review, and Harvard Journal of Gender & Law, among others. With Holly Fernandez Lynch and I. Glenn Cohen, she co-­edited Law, Religion, and Health in the United States (Cambridge University Press, 2017). David Shaw is Assistant Professor of Health Law and Ethics at Maastricht University and Senior Research Fellow in Biomedical Ethics at the University of Basel. His background is in philosophy and medical law, and he is an expert on organ donation, research ethics, and research integrity. Karl Harald Søvig  is professor of law and currently Dean of the Faculty of Law, University of Bergen, Norway. He is also Chair of the European Association of Health Law. His research has focused on different forms of coercive measures within the welfare state, as well as on the rights of the children. Sonia Suter is Professor of Law and the Kahan Family Research Professor at The George Washington University Law School. She is the author of numerous articles published in law reviews and peer-­reviewed journals and is co-­author of the leading textbooks on genetics and the law and assisted reproductive technologies. Keith Syrett is Professor of Health Law and Policy in the School of Law, University of Bristol, UK. He has written extensively on access to healthcare, resource allocation and public health, including Law, Legitimacy and the Rationing of Healthcare (Cambridge University Press, 2007) and Public Health Law: Ethics, Governance, and Regulation (Routledge, 2017). Brigit Toebes is Chair of Health Law in a Global Context at the Faculty of Law of the University of Groningen, the Netherlands. She is the editor and author of numerous books and articles in the field, including Health and Human Rights in Europe (Intersentia, 2012/2021) and The Research Handbook on Global Health Law (Edward Elgar, 2018).

xviii   about the contributors David Townend is Professor of Health and Life Sciences Jurisprudence at Maastricht University. His work focuses on the changing nature of autonomy and solidarity; on concepts of privacy, property, and public interest; on politeness, manners, ethics, and law; and on personal data sharing, particularly in medical research and data science. Patricia J. Zettler is Associate Professor at the Ohio State University Moritz College of Law and a member of Ohio State’s Drug Enforcement and Policy Center and its Comprehensive Cancer Center. Her work has appeared in numerous journals including Science, JAMA, Indiana Law Journal, and Food and Drug Law Journal.

chapter 1

Editors’ I n troduction to The Ox for d H a n dbook of Com pa r ati v e Hea lth   L aw Tamara K. Hervey and David Orentlicher

1  Why Compare (Health) Laws? All human beings need healthcare at some time in their lives. Countries and societies organize the provision of healthcare and the protection of public health through different blends of individual, family, governmental, “third sector,” and private actor responsibilities. Situated at the interfaces between human rights, dignity, autonomy, and solidarity, and between multinational business activity, state regulation, and the most private of individual decision-­making, health law is perhaps unique in its capacity to provoke a range of responses to how law should constrain and facilitate human activity. Scholars, legislators, and judges grappling with critical questions—such as how to finance healthcare, how to regulate the pharmaceutical industry, how to protect against threats to public health that can quickly cross national borders—can learn from the successes and failures of the different approaches taken in different countries. Where there are “common challenges” in the law, in biomedical ethics and in the practices and provision of healthcare across the world, it may be useful to deploy experiences from other places to improve the law. In health law, common problems are frequent. In addition to ones already mentioned, there are questions about how to enhance primary care and overall population health while accommodating effective specialty/secondary care, and, in general, how to ­refocus

2   Tamara K. Hervey and David Orentlicher healthcare from the individual to the population. Other challenges include how best to coordinate care for the chronically ill across different professional and institutional providers, the state, the “third sector,” the market, and the family, and to incentivize efficient behavior in that context. Relatedly, common health law problems include how to design payment or other resourcing systems to reward high-­quality performance and “joined up thinking” between different actors. Furthermore, health law across jurisdictions that recognize fundamental human rights to non-­discrimination could be said to share the problem of how to address inequality in healthcare and health outcomes along a number of “forbidden grounds” (race, sex, gender, sexuality, social class, age, disability, and so on). Furthermore, health lawyers must consider how to balance an abstract notion of a “patient” with the embodied reality of individual experiences. In its “functionalist” or “instrumental” role,1 first and foremost, comparative analysis can support a search for a “better” solution to domestic legal problems, either as legislative reform or as judicial reasoning/advocacy and argumentation. For instance, the judgment in the English case A and Others v National Blood Authority,2 in justifying its conclusion, refers to legal literature from five different legal systems, both common law and civil law, and surveys of legal practice in eight European countries plus the United States. It also considers different-­language versions of the relevant legal text (an EU Directive). A “better” solution may be understood here endogenously, in terms of the coherence and structural “neatness” of the legal solution proposed, but is more often understood exogenously, in particular to improve the economic (or other standard of) efficiency of law or law’s wealth-­producing capacity. Second, instrumental comparative law can support the aim of understanding one’s domestic system better.3 Comparison helps us see our own system differently, to enhance domestic discourses, especially if applied with a healthy dose of humility. Better understanding makes us better lawyers: chauvinism and associated isolation has no place in legal science. Third, functionalist comparative health law could improve domestic legal education.4 To be sure, there are limits to the ability of one country to learn from another. All of the above implies some level of generalizability. It implies that we can recognize “health 1   Usually attributed to K Zweigert and H Kötz, An Introduction to Comparative Law (Oxford University Press 1998), trans T Weir, this problem-solving approach has been described as “one of the best known working tools in comparative law.” It has enjoyed a long-standing reputation as the basic methodology in comparative law literature. It flows from an assumption or belief that similar problems can be solved in similar ways within different legal systems and that the countries or societies under comparison experience problems at a sufficiently shared level to allow legal solutions to be “transplanted” from one system to another. For an early rebuttal see, e.g., B Markesinis “Comparative Law: A Subject in Search of an Audience” (1990) Modern L Rev 53, 1–30. 2   [2001] 3 All ER 289. 3   U Kischel, trans. A Hammel, Comparative Law (Oxford University Press 2019), 53; HP Glenn, “The Aims of Comparative Law” in J Smits, ed, Elgar Encyclopedia of Comparative Law (Edward Elgar 2012), 66–68, at 69; J Husa, A New Introduction to Comparative Law (Hart 2015), 69–70, including better “systemisation” of one’s own legal system, Husa, 32. 4   Which is “slowly overcoming its hegemonic biases,” Glenn supra n 3, 72; Husa, supra n 3, 91–95.

Introduction   3 law,” in whatever context we find it, by reference to either a “commonsense” notion of the interface between law and health or a set of (legal) principles and a legal logic and systemic coherence that characterizes the field, or, as in our case, a combination of the two. As we’ve laid things out, legal reasoning about, let us say, patient autonomy, health professional liability, or state responsibility for healthcare should be recognizable regardless of the jurisdiction in which it takes place. The benefit of such a position is that it allows us to embark on a comparative health law project like this one. However, we also recognize that health law is sensitive to its local environment. For us, legal scholarship at its best cannot operate outside of the contexts in which law is made, interpreted, and applied. Cultural, historical, and political considerations are important drivers of public policy, and one cannot assume that a health law framework that works in one country will also work in other countries. While we can identify fundamental, universal principles, such as dignity, autonomy, and the sanctity of life, their application to specific questions of health law cannot be determined solely in the abstract. As John Harrington has observed, “legal reasoning establishes only provisional or contingent truths.”5 Inevitably, conflicts will arise among the different basic principles—as when principles of self-­determination clash with the interest in preserving life—and different persons and societies will resolve these conflicts differently. A review of comparative law literature would seem to support the claim that health law is not a “natural” or frequently occurring topic for comparative legal methods.6 But we can learn from our differences as much as from our shared values. In this regard, we can extend the application of comparative analysis to health law by also placing ourselves within the notion of comparative law as an explanatory and critical enterprise.7 Explanatory/critical comparative law studies seek to explain or understand the history of a legal phenomenon in its contemporary contexts. In overtly critical

5   J Harrington, “Time as a Dimension of Medical Law” (2012) Med L Rev (2012) 491–515, at 495, citing AC Hutchinson, It’s All in the Game: A Non-Foundationalist Account of Law and Adjudication (Durham, NC: Duke University Press 2000). 6   In the past 12 years, the American Journal of Comparative Law, the International and Comparative Law Quarterly, and the Journal of Comparative Law have carried almost no publications on any topic of medical law or health law. There is one article on abortion, one on medical mediation, and two on surrogacy: W Cao, “‘Glorious Mothers’ and ‘Rational Women’: A Comparative Analysis of the Chinese and English Regulatory Models of Abortion” (2019) Am J Comp L 67(4), 745–774; C Ding, “A Dose of Medicine to Cure Medical Chaos: Medical Mediation in China” (2015) J Comp L 10, 158 ; D Gruenbaum, “Foreign Surrogate Motherhood: Mater Semper Certa Erat” (2012) Am J Comp L 60(2), 475–505 and C Spivack, “The Law of Surrogate Motherhood in the United States” (2010) Am J Comp L 58(suppl 1), 97–114. There is also nothing on medical or health law in M Reimann and R Zimmermann, eds, Oxford Handbook of Comparative Law (Oxford University Press 2006 or 2019). There are no health law examples in G Samuel, Introduction to Comparative Law Theory and Method (Hart 2014) and very few in Husa, supra n 3, or in Kischel, supra n 3. 7   See, P Legrand “Jameses at Play: A Tractation on the Comparison of Laws” (2017) Am J Comp Law 65, 1–132; Samuel, supra n 6, 13–14. As Husa, supra n 3, 60, puts it: “The fundamental object of comparative law is to acquire knowledge on what separates the legal systems or cultures studied and what connects them and to explain or assess what has caused the differences and/or similarities.”

4   Tamara K. Hervey and David Orentlicher mode, they may also seek to uncover the underlying power relations that determine a legal phenomenon. The explanatory and critical modes of comparative law are perforce attentive to the cultural, historical, and political factors that distinguish countries. “The law emerges out of the cultural domain just like a piece of architecture or a song.”8 It would be poor comparative legal scholarship to ignore the social, economic, cultural, or historical aspects of the written legal texts.9 Rather, comparative lawyers make explicit and shine a light on the specificities of the place and time that they study in terms of cultural, economic, social, or political arrangements that have become bundled in with the ideas that or­gan­ ize the field and thereby obfuscated or even entirely hidden from a conceptual/principlist view of the field. Where we seek to learn from this approach, an attention to local context means that we will avoid making crass assumptions about what it is feasible to “learn” about another system or our own system, or to “transplant” from one system to another. For instance, New Zealand and Sweden’s “no fault” liability systems are embedded in a range of other legal, institutional, economic, and cultural contexts. We cannot naïvely assume that they could be simply transplanted to other contexts. But it also would be wrong to exaggerate the influence of local context. As Timothy Jost puts it, “Judicial decisions on health law topics, for example, access to abortion or the right to die, are primarily grounded in national law.” However, they are “also based in understandings about the nature and meaning of human life, autonomy and community, that are, in turn, grounded in philosophical and religious traditions that transcend national boundaries.”10 Thus, as Emily Jackson and I. Glenn Cohen write, procreation receives recognition as a fundamental right on both sides of the Atlantic Ocean. Similarly, as Mette Hartlev, Maxwell Mehlman, and Sonia Suter discuss, principles of patient autonomy and informed consent loom large in genetics and the law in both Europe and the United States. So, the task of the comparative health lawyer is to first surface those background ontological assumptions. Only then can we consider the extent to which they are, or more likely are not, actually traditions that may “transcend national boundaries.” And once we understand that, we can learn about our own health law by juxtaposing it with lines of judicial reasoning or legislative language from other systems. This is where we locate our project in the existing literature. Ultimately, there is an important convergence of the functional and explanatory approaches. Comparative law endeavors are often ultimately aimed to argue for legal

  Samuel, supra n 6, 119. Or, as S Munshi, “Comparative Law and Decolonizing Critique” (2017) Am J Comp L 65, 207–235, at 217, citing Legrand, supra n 7, 47, puts it: “The fabric of any legal text consists of historical configurations enmeshed with traces of political rationalities intertwined with traces of social logics interwoven with traces of philosophical postulates plaited with traces of linguistic orders darned with traces of economic prescriptions interlaced with traces of epistemic assumptions.” 9   Samuel, supra n 6, 119. 10   T Jost, “Comparative and International Health Law” (2004) Health Matrix 14, 141–147, at 143. 8

Introduction   5 change, and the same is true for much health law scholarship. Effective use of law to argue for change requires attention to specific times and places. As a reform project, comparative law methods can help us to see just how contingent our own system—and its laws—is. Deployed sympathetically, comparative (health) law can even stretch our imaginations to understand and articulate the possibilities to “make things better.”11 Moreover, in our attention to non-­legal contexts, as comparative (health) lawyers we do not want to lose entirely the disciplinary contributions of positivist law, legal methodologies and epistemologies, and legal hermeneutics.12 Scholarship on legal language and reasoning has encouraged legal scholars to pay attention to the way the law and legal text shape social realities, just as much as vice versa. “Legal technicalities” exert influences on the way the world is understood.13 So comparative health law in a project like this one is also an exercise in understanding the significance of legal texts and the ways in which their meanings in the sense of legal definitions have implications for patients, providers of healthcare, payers, and states. For example, both Richard  S.  Saver and Sunčana Roksandić explore the legal meaning of “conflict of interest” in the context of physicians’ duties to their patients, the health system, and society in general, and they draw out how definitional matters make a difference in terms of what is caught by conflict of interest rules. This aspect of our research design is an example of a blended methodological approach, one that pays attention both to the contextual nature of law and legal reasoning in the economic, political, social, cultural, and biomedical world and to the effects of legal language and text and how they are deployed, including as if legal text were acontextual and principles-­driven and determined. Putting all of this together, we have taken the view that comparative health law analysis can identify important ways in which countries can learn from each other, as well as the ways in which local context may impede cross-­fertilization. As to opportunities for adoption of models from other countries, David Orentlicher and Judit Sándor consider whether states in the United States might find the European and Canadian experiences with physician-­administration of lethal medication at the end of life sufficiently reassuring to relax the current requirement of patient self-­administration. In terms of local barriers to cross-­fertilization, Amandine Garde and Wendy E. Parmet discuss the important steps taken by governments on both sides of the Atlantic Ocean to address the public health threat from tobacco use while also identifying the constraint on the ability of federal and state legislatures or agencies in the United States to adopt tobacco marketing regulations from abroad because of the Constitution’s robust protection of commercial speech under the First Amendment. Similarly, while government funding of healthcare services is substantial in the United States as well as in European countries, Allison K Hoffman and Mark A. Hall remind us that, in the United States, “health insurance

11   F Miller, “Review of Jost’s Readings in Comparative Health Law and Bioethics (Durham, NC: Carolina Academic Press 2007)” (2008) J L Med Ethics 36, 851–855, at 852. 12   Samuel, supra n 6, 86–88.    13. Id., 24.

6   Tamara K. Hervey and David Orentlicher developed as an employment benefit throughout the twentieth century and has become deeply entrenched as such.” Even when we cannot fully export health law doctrines across national boundaries, increasing globalization means that health policy challenges bring us together. Obviously, diseases, pathogens, and environmental factors do not respect state boundaries. Novel and life-­threatening diseases, such as Zika or COVID-­19, discussed by Markus Frischhut and Wendy E. Parmet, cross the globe with ease. Patients and their families, such as the parents of Alfie Evans and Charlie Gard, discussed in Jessica Berg and Emma Cave’s chapter, seek medical treatment in other jurisdictions, often in tragic circumstances. Multinational companies seek to enter healthcare markets.14 Economic globalization and nationalistic responses to it make it increasingly important for students, practitioners, and scholars to understand how health laws vary across national boundaries, how legal rules may be better harmonized from country to country, and what are the limits to such alignment. Since 1948, the World Health Organization has been providing ethical and evidence-­ based information on health policy options to help inform states’ law-­making. The World Association for Medical Law, established in 1967, brings together scholars from more than 100 countries to promote understanding of “health law, legal medicine and ethics, for the benefit of society and the advancement of human rights.”15 When faced with difficult questions concerning health or biomedicine (such as, e.g., end-­of-­life decision-­making), courts regularly consider approaches in other jurisdictions.16 Equally, when proposing law reform, other jurisdictions offer inspiration to legislatures.17 As global interconnections continue to increase, even as some countries seek to respond by (partial) retreats to national exceptionalism, this perhaps is an especially fitting time to publish a comparative view of any field, and health law is no exception.

14   See, for instance, Amazon’s announcement in January 2018 that it is entering a partnership with Berkshire Hathaway and JP Morgan Chase to tackle the “ballooning costs” of healthcare which “act as a hungry tapeworm on the American economy” https://www.reuters.com/article/us-amazon-healthcare/ amazon-berkshire-jpmorgan-partner-to-cut-healthcare-costs-idUSKBN1FJ1NF; https://www.ft.com/ content/f4694aa6-05c1-11e8-9650-9c0ad2d7c5b5 15  http://wafml.memberlodge.org/ 16   For example, the US Supreme Court in Washington v Glucksberg 521 U.S. 702, 718 (1997); the European Court of Human Rights is required to review European jurisdictions when considering its margin of appreciation doctrine, see, e.g., Sanles Sanles v Spain No 48335/99, ECHR 2000-XI; Pretty v the United Kingdom No 2346/02, ECHR 2002-III; Haas v Switzerland, No 31322/07, ECHR 2011; Koch v Germany judgment No 497/09, 19 July 2012. 17   For instance, the Law Commission of England and Wales, Injuries to Unborn Children Working Paper No 47, 1973; Mentally Incapacitated Adults and Decision Making: An Overview Consultation Paper No 119 1991; Law Commission of E&W, Scottish Law Commission, Northern Ireland Law Commission Regulation of Healthcare Professionals Cm 8839, 2014; Law Commission of E&W review of surrogacy law announced May 2018 https://www.lawcom.gov.uk/surrogacy-laws-set-for-reform-as-lawcommissions-get-government-backing/; the UK’s (unsuccessful) “Assisted Dying Bill 2015” was based on Oregon’s Death with Dignity Act; see Hansard 11 September 2015 https://publications.parliament.uk/ pa/cm201516/cmhansrd/cm150911/debtext/150911-0001.htm#15091126000003

Introduction   7

2  Determining the Scope of the Project Plenty of ink has been spilled on the question of what “counts” as health law or medical law.18 This kind of reflection has been around for at least 30 years in the English-­language literature, and this project does not seek to add to it. However, in practice, we did have to set some boundaries around “health law” as a category as we sought to determine its scope for the purposes of a project like this one. One way to think about health law is to adopt a pragmatic focus on the substantive content of the legal puzzles at issue: health law is about laws that affect the healthcare system. These would include laws regarding health insurance, professional liability, pharmaceuticals, procreation, and end-­of-­life decisions. Such an approach would pick up most of the important topics, but it also would conflate health with healthcare. And some of the most critical ways to improve people’s health are by preventing disease and making healthcare unnecessary. Accordingly, we thought more generally about laws that promote the health of the citizenry, such as environmental rules and gun regulations. The key strengths of this approach are its relative comprehensiveness and conceptual simplicity: its key weakness is that it is not obvious where the boundaries of “health law” lie. As Brigit Toebes observes in her chapter on the socioeconomic determinants of health, there is a danger that everything falls within the category “health law”: law on poverty, education, housing, taxation, criminal process . . . the list is endless. Recognizing this limitation, therefore, we also sought to work from the guiding principles or concepts that are assumed to underpin health law and to give it a certain coherence.19 Sometimes these are seen as endogenous legal principles, such as human rights, individual autonomy, or reasonableness or proportionality of state action. Sometimes they are exogenous precepts from other disciplines, principally from (biomedical) ethics, such as the principle that the human body ought not to be “property,” discussed by Justine Pila generally and more specifically by Natalie Ram and Stéphanie Hennette Vauchez in the context of payments for blood, sperm, and eggs. Such legal

  The literature here is voluminous, with some examples in the bibliography.   Such “principlism” is implicit, for instance, in Keown’s critique of Kennedy and Grubb’s 1998 Principles of Medical Law, which includes the indictment that it was “not an analysis of principles” the implied expectation being that it should have been. See J Keown, “Review of Principles of Medical Law” (2000) Cambridge L J 412, 413. Gostin’s foreword in honor of the special issue marking Brazier’s retirement as Editor in Chief from the Medical Law Review begins with the “First principles of medical law,” see LO Gostin, “Foreword in Honour of a Pioneer of Medical Law: Professor Margaret Brazier OBE QC FMEDSCI” (2012) Med L Rev 20, 1. See also M Hall, “Law, Medicine and Trust” (2002) Stanford L Rev 55, 463, describing a quest for “an organising principle” of health law; The source of such principles is often stated or assumed to be bioethics, and the classic source is T Beauchamp and J Childress, Principles of Biomedical Ethics (Oxford University Press 1979, now 2013), and the work of R Gillon, Philosophical Medical Ethics (Wiley 1986), taken up by legal theorists such as R Dworkin Life’s Dominion (Knopf 1993). 18

19

8   Tamara K. Hervey and David Orentlicher principles can be deduced through syllogism, inspired by the “civil law” method,20 or, equally, can be determined inductively, through analogy inspired by the “common law” method.21 So our approach as editors to determining the scope of the project combines pragmatism and principlism. Our next step was to apply that principled pragmatism in a dialogic and discursive process. This yielded some surprising insights. For instance, Hervey was taken aback that gun control, or fraud,22 should be counted as sufficiently central topics for a health law book: in European contexts those subjects are not usually associated with health law scholarship. We noticed that Orentlicher’s focus seemed to be more litigation-­based than Hervey’s, reflecting perhaps the common law idea that the law is the cases; whereas study of both common law and civilian systems within the framework of the European Union and Council of Europe attunes European legal scholars more to other sites of law-­ making and definitions of what is interesting for legal scholarship, particularly legislation. After several iterations, the process resulted eventually in a sketch of the field which was recognizably “health law” to both editors, but does not constitute what either of us would have come up with alone. After this introduction and a broad take from Alceste Santuari and William M. Sage in conversation on paradigms of healthcare systems, law and regulation in the United States and Europe (Part I), the book is organized into four main parts. We felt it especially important to begin with laws designed to prevent ill health (Part II). Even before

  See, for instance, A Laufs, B-R Kern, M Rehborn, Handbuch des Arztrechts (CH Beck 2019); T Spranger, Medical Law in Germany (Kluwer Law International 2012); H Nys, La Médicine et Le Droit (Kluwer 1995); L Westerhäll, Medical Law: An Introduction (Fritzes 1994). As Fleming has observed, medical law, being largely contractual, has emerged in civil law countries as a “vast gloss overlaying a few exiguous Code articles,” see JG Fleming, “Comparative Law of Torts” (1984) Oxford J Legal Studies 4, 235, at 241. 21   See, for instance, in addition to the common law texts already cited, S Pattinson, Medical Law and Ethics (Sweet and Maxwell 2020); E Jackson, Medical Law: Text, Cases and Materials (Oxford University Press 2019); Mark Hall et al., Health Care Law and Ethics, 9th ed. (Wolters Kluwer 2018); Barry Furrow et al., Health Law: Cases, Materials and Problems, 8th ed. (West Academic 2018); J Laing and J McHale, eds, Principles of Medical Law (Oxford University Press 2017); D Madden, Medicine, Ethics and the Law (Bloomsbury 2016); D Tomkin and P Hanafin Irish Medical Law (Round Hall Press 1995). 22   And yet, of course, the phenomenon of fraud and corruption in healthcare settings is just as much of an issue (in different ways) in Europe as it is in the United States. It is a bit surprising therefore that this is not in the mainstream of European medical/health law as it is well-established that health systems are particularly prone to corruption; see, e.g., WD Savedoff, K Hussman, “Why Are Health Systems Prone to Corruption?” in J Kotalik, D Rodriguez, eds, Global Corruption Report (Pluto Press 2006), 4–16. There are, of course, exceptions; e.g., see Emily Jackson on confusing signals sent by ethical guidance given in doctors’ professional bodies’ codes in response to patient fraud where the patient is not “ordinarily resident” in the United Kingdom: E Jackson, “The Relationship Between Medical Law and Medical Ethics” (2015) J Med Ethics 41(1), 95–98. But, in general, fraud and corruption doesn’t usually make it to the list of what “counts” as health/medical law in Europe, simply because we have different notions of where law and legal scholarship should be interested—presumably because of different histories with the United States, as well as different litigation patterns, regulatory and governance approaches, and possibilities. 20

Introduction   9 the spread of COVID-­19, it was clear that policy pays insufficient attention to public health programs that can address the causes rather than the symptoms of illness and injury. An ounce (or gram) of prevention really is worth a pound (or kilogram) of cure. Next, we consider key aspects of treating disease—access to the healthcare system; regulation of physicians, hospitals, and other providers of care; and regulation of the relationships between care providers and their patients (Part III). How do we use law to make sure patients can receive the care they need and that the care is delivered properly? Doctors and other health professionals cannot treat disease effectively without technological advances, so we continue with chapters on oversight of the research enterprise, drug and device development, genetics, and the use of organs and other body parts (Part IV). We conclude with the health law questions that arise regarding decisions to bring a new person into existence or decisions to hasten the end of someone’s life (Part V). When, for example, must control over one’s destiny yield to state interests in the sanctity of life?

3  Choice of Contributors and Invitation to a Collaborative Method Taking all of the preceding into account, the aim of our Handbook is to bring together the existing and emerging body of research in health law, with a key focus on the US– European comparative perspective. Maintaining this core focus contributes significantly to the coherence of the Handbook and also its claim to disciplinary credibility. This is one strength of our approach. As long-­standing scholars of health law in our respective regions, we each have extensive networks of scholarly collaborators. But those networks are largely focused in our respective geographical spaces. Our decision that, as editors, we could meaningfully offer a US–European comparative health law handbook, but not one that is global in its scope, was founded in an understanding that only those who work within a particular legal system can offer a sufficiently nuanced understanding of its specificities to contribute to a meaningful comparison in a critical-­ analytical explanatory mode. Only “insiders” can offer this kind of insight because of their educational and professional backgrounds informing their legal epistemological frames as well as their understandings of how the legal interacts with non-­legal contexts. The jurisdictional and geographical scope of the book thus covers law in one region (Europe) and one federal system (the United States) and its constituent states. As a federal system, the United States divides government power between the national government and the state governments. Matters of physician licensing and discipline and standards of practice typically reside at the state level, as Isaac D. Buck explains, while, as shown by Patricia Zettler and Erika Lietzan, pharmaceutical regulation is handled at the national level. For some matters, including the financing of healthcare (discussed by Alceste Santuari and William M. Sage, and Allison K. Hoffman and Mark A. Hall), antitrust

10   Tamara K. Hervey and David Orentlicher enforcement (discussed by Thomas Greaney), and protection of public health, important activity is seen at both levels, sometimes jointly, other times independently. Over time, as the federal government has assumed greater responsibility for financing healthcare, as with the Medicare and Medicaid programs, the federal role in healthcare has increased substantially. Or sometimes it becomes apparent that a national response is needed to address concerns previously handled by states, as with the data privacy protections in the Health Insurance Portability and Accountability Act (HIPAA), discussed by Sharona Hoffman in her chapter with Jean Herveg, or the mandate in the Emergency Medical Treatment and Active Labor Act (EMTALA) for hospitals to screen and stabilize anyone coming in with a potentially emergent need for care. The relevant jurisdictions in Europe are less straightforward to conceptualize. Painting with a broad brush, at the European level, with 47 member countries, the Council of Europe is the source of several human rights instruments which are important for health law: in particular, the European Convention on Human Rights and the Oviedo Convention on Biomedicine, discussed by several of our contributors, including Brigit Toebes (public health), David Shaw and David Townend (medical research), Joaquin Cayon-­De Las Cuevas and David Orentlicher (organ transplantation), and Graeme T. Laurie and Carl H. Coleman (biobanks). Twenty-­seven of those countries are also Member States of the European Union, which is the main site for regulation of some aspects of health law, particularly pharmaceuticals and medical devices regulation, explained in detail by Aurélie Mahalatchimy in her two chapters. Both European regional systems leave significant discretion in practice to domestic jurisdictions, which determine most of the key elements of health law, especially the design of healthcare systems, discussed by Alceste Santuari in his chapter with William M. Sage, and the regulation of delivery of healthcare by institutions and healthcare professionals, covered by Deirdre Madden. National discretion is also important in practice even where EU-­level rules purport to harmonize: data protection (see Sharona Hoffman and Jean Herveg’s chapter); antitrust (see Okeoghene Odudu’s chapter); and regulation of research (see David Shaw and David Townend’s chapter) being three cases in point. In some European countries, such as Spain or Italy, important matters of health law are devolved to a subnational regional level. Also, while the national government in the United States can enforce its laws and rules, the Council of Europe and the European Union rely more heavily for enforcement of their Conventions and Directives on the courts and administrative authorities of each different country. Of course, it should go without saying that important voices in health law exist outside of the United States and Europe. As we gathered our authors, we asked them, where possible, to draw on pertinent examples from outside those regions as part of their comparative analysis. In some contexts, such as the regulation of clinical trials and access to medicines, it is particularly important to be mindful of how approaches to health law differ between developed and developing countries and how health law in developed countries may be problematic for developing countries. In other contexts, like tobacco control, or capacity and autonomy of children and people with disabilities, our authors were able to draw on global legal texts such as the World Health Organization’s

Introduction   11 Framework Convention for Tobacco Control, relevant for Amandine Garde’s and Wendy Parmet’s public health chapters on unhealthy products; or the UN’s Convention on the Rights of the Child and Convention on the Rights of Persons with Disabilities, relevant for Jessica Berg, Emma Cave, and Mary Donnelly’s chapters on patient autonomy (see Berg and Cave; and Donnelly and Berg). But we acknowledge that a weakness of this Handbook is that it does not claim to be truly global in its scope: that would have involved a different team, both of editors and of authors. Our authors were of course carefully selected, not only for their knowledge of and contributions to the literature on their own particular state jurisdiction, but also for their understandings of both US federal law or the law of the Council of Europe and European Union and of other states in the United States or other European countries on the particular topic on which we asked each author to write. The section on healthcare financing by André den Exter and Keith Syrett, for instance, needed contributions that cover the range of approaches usually encapsulated by “Bismarck” and “Beveridge.” A comparative analysis of law on end-­of-­life decision-­making (David Orentlicher and Judit Sándor) needs attention to “outlier” states such as Switzerland and the Netherlands on the European side, and Oregon and California in the United States, as well as to the dominant approaches found elsewhere in Europe and the United States. Comparing approaches to physician and institutional liability needs knowledge of systems focused more and less on “fault,” as shown by Karl Harald Søvig and Barry Furrow. Studying organ donation, as Joaquin Cayon-­De Las Cuevas and David Orentlicher do, requires noting the very different approach adopted in Iran, where organs may lawfully be bought from living donors. And, as demonstrated by Janne Rothmar Herrmann and Elizabeth Sepper, an analysis of the right not to procreate entails attention to countries or states, such as Poland or Texas, that place greater priority on the fetus’ potential for life, as well as those jurisdictions that place greater priority on the pregnant person’s reproductive autonomy. Our overall aim in the Handbook is to offer an authoritative set of essays covering health law as we understand it and seeking to encapsulate the full range of discussion, opinions, debates, and controversies in that field, both defining and advancing knowledge and understanding. Each essay provides a critical-­analytical piece on its topic, surveying the essential issues and questions, and offering a contemporary perspective. Most of our animating reasons for the project are firmly within the evaluative/critical method of comparative law. The ultimate aim of the project is to learn from the diverse contexts in which health law is played out. Here we explicitly requested our authors to attend to spatial contexts (state, nation state, federation, “sui generis” polity, regional human rights organizations). Our authors also consider temporal contexts in explicating the various economic, social, cultural, and political backgrounds which influence the legal rules that they explore. As Tracey A. Elliott and Joan H. Krause observe, for example, these different influences lead to different definitions of healthcare fraud from country to country. Although our authors are interested in reflecting on how the balance of different principles in shaping rules for a particular jurisdiction might be improved, their analyses spring from the understanding that legal rules are never entirely distinct from the contexts in which we find them.

12   Tamara K. Hervey and David Orentlicher To achieve meaningful comparative analysis, we asked our authors if they would be prepared to adopt a collaborative method. We believe this approach is one of the greatest strengths of the Handbook. On a practical level, it involved the pairing or grouping of contributors, one or more from the United States and one or more from Europe. We asked each pair or group to work together on their contributions on a particular topic of health law. We suggested that the contributions be framed by an initial review of the topic at issue, to be co-­written by the two or more authors. As inspiration for that initial review, as well as asking our authors to think of common principles, we also suggested specifically contextual ideas: a real-­life case or cases or a fictitious scenario from a book, film, or play. The idea behind these requests to our authors was to engender conversations between them—sometimes facilitated by us as editors—before or as they began their individual work. These conversations had the aim of surfacing further the necessary contextual, spatial, and temporal awareness of the contingency of the legal phenomena which our authors were asked to analyze. We did not envisage that our authors would have much difficulty with the approach per se, and that seems to have been the case. After all, these days, few legal professionals or law professors adopt a nineteenth-­century legal positivist approach. It is vanishingly rare to find a contemporary scholar of health law who believes that settling a legal question requires only textual exegesis of legally authoritative texts in a given jurisdiction. In some instances, differences of style and approach to writing initially seemed hugely disparate in the pairs or groups, but these differences narrowed—or at least became less important—as the conversations continued through successive drafts shared within the pairings and with us as editors. Furthermore, our “insider/outsider approach”23 is designed to secure sufficient understanding of each other’s systems, and our own systems in a comparative context, to render communications mutually intelligible. This is a crucial feature of a research handbook that is intended to be read across many jurisdictions. The approach is based on a profound notion of equal and mutual respect of the other.24 It is designed to minimize the phenomena of legal hegemony, where the “first to the field” or the “loudest voices” get to determine what matters, what counts, and how we describe, and hence analyze and understand, health law. James Whitman suggests that the comparative method can be pursued through what he calls Einfühlung (loosely translated as “empathy”). An “understanding of a foreign time or place requires a practice of Einfühlung, of empathetic understanding, that must inevitably be conducted by the individual scholar.”25 Or, we might prefer Russell Miller’s notion of “hospitality”: indeed the “duty of hospitality owed to the ‘other.’ ”26   Samuel, supra n 6, 60–64.   E.g., “recognition and respect of an otherness that . . . enriches one’s appraisal of what there exists in the world as law,” Legrand, supra n 7, 128; noted with approval by Peer Zumbansen, “Les Jeux Sont Faits: Comparative Law – As It Really Was Meant To Be?” (2017) Am J Comp L 64, 237–255 at 253: “ ‘Yes! Yes!’ We almost want to shout out these words.” 25   JQ Whitman, “The Hunt for Truth in Comparative Law” (2017) Am J Comp L 65, 181–190, at 189. 26   R Miller, “On Hostility and Hospitality: Othering Pierre Legrand” (2017) Am J Comp L 65, 191–206, at 206. 23

24

Introduction   13 We also hoped to engender a certain humility through this approach. This is an important strength of the approach for a number of reasons, including the global he­gem­ony of the United States and of the English language and the fact that sometimes our author pairings involved people at different career stages. As best we could, we wanted to guard against the possibilities of legal imperialism, or scholarly domination, creeping unwittingly into our project. This mutual respect and self-­reflective humility embody core values of comparative law approaches which seek to secure an approach that does not settle for centering of the self, or nationalizing or universalizing the national, but tries to reposition legal scholarship in the world more generally, and in history, by respecting difference and seeking as much as possible to faithfully represent the legal position in other jurisdictions. In designing the common introductions, we encouraged our authors to see if they could think critically about how knowledge is conventionally structured and produced in their topic areas and try to move beyond asking only the questions obvious to each of them, beyond those invisible intellectual habits that define what we think “count(s)” as our particular (sub)discipline(s) or topics within them. Thus, behind the critical-­analytical aim of the Handbook sits a reflexive aim: to learn from how the diverse contexts in which health law is played out affect how we, as scholars and teachers of health law, define, analyze, and deploy the field that we study. What is it to be a health lawyer in the context of the United States, in a European context, and to what extent and in what respects do those two contexts differ? This reflexive mindset is true for the authors as well as encapsulating our own approach as editors as we defined and outlined the project as a whole.

4  Conclusions and Key Lessons Learned In the end, we are not surprised to find a field of health laws rather than anything approaching a universally understood health law. While we can readily point to shared principles, common approaches, and areas of convergence, there also are important and persistent variations across jurisdictions. And, as discussed, this is a feature rather than a bug of the US and European legal systems. This feature arises because US and European legal systems value both state sovereignty and federal (in the case of the United States) and/or regional (in the case of the Council of Europe and European Union) unity while at the same time permitting subnation state variance (or in the case of those European states where health law differs in subnational devolved regions, within the state). Similarly, the variations on which our authors reflect also are a feature of health law as a body—or rather bodies—of law. Even widely shared principles come into conflict with each other, as, for example, when individual autonomy clashes with pursuit of overall social welfare. This question has come sharply into focus for all jurisdictions as COVID-­19 vaccines have become available. Inevitably, the balance drawn will vary depending on a

14   Tamara K. Hervey and David Orentlicher jurisdiction’s cultural, historical, political, social, and economic contexts, and that is what our collective comparative law endeavors show in this Handbook, across the full range of health law topics. Still, our exploration of health law across the Atlantic Ocean confirms what we’ve suggested earlier in this introduction. Healthcare students, scholars, and policymakers can learn much from approaches beyond their national borders. While there will be practices that just won’t transfer to a different societal context, there also are very good ideas that can be brought back home. The comparative perspective is one of the most important paths to legal understanding and reform, and thus, ultimately, to better health outcomes.

Bibliography Why Compare (Health) Laws? D Cowan and D Wincott, eds, Exploring the Legal in Socio-Legal Studies (Palgrave Macmillan 2016). N Demleitner, “Comparative Law and Legal Education,” in M Reimann and R Zimmermann (eds) Oxford Handbook of Comparative Law (Oxford University Press 2019). A den Exter, ed, European Health Law (Maklu 2017). C Dietz, M Travis, M Thomson, eds, A Jurisprudence of the Body (Palgrave Macmillan 2020). H P Glenn, “The Aims of Comparative Law” in J Smits, ed, Elgar Encyclopedia of Comparative Law (Edward Elgar 2012), 66–68 M Graziadei, “Comparative Law, Legal Transplants, and Receptions,” in M Reimann and R Zimmermann, eds, Oxford Handbook of Comparative Law, Second Edition (Oxford University Press 2019), 442–473. D Gurnham, ed, Law’s Metaphors: Interrogating Languages of Law, Justice, and Legitimacy. Special edition (2016) J L Soc. 43(1), 1–165. M Hanne and R Weisberg, eds, Narrative and Metaphor in the Law (Cambridge University Press 2018). D Hoffman, “Comparative Health Law and Policy: What, If Anything, Can We Learn from Other Countries?” (2009) J L Med Ethics 37, 790–791. J Husa, A New Introduction to Comparative Law (Hart 2015). J Husa, “Comparative Law in Legal Education: Building a Legal Mind for a Transnational World” (2018) Law Teacher 52(3), 201–215. T Jost, “Comparative and International Health Law” (2004) Health Matrix 14, 141–147. E D Kinney, The Affordable Care Act and Medicare in Comparative Context (Cambridge University Press 2015). U Kischel, trans. A Hammel, Comparative Law (Oxford University Press 2019). P Legrand “European Legal Systems are Not Converging” (1996) Int’l Comp L Q 45(1), 52–81 P Legrand, “How to Compare Now” (1996) Legal Studies 16, 232–242 H Muir Watt, “Globalization and Comparative Law” in M Reimann and R Zimmermann, eds, Oxford Handbook of Comparative Law, Second Edition (Oxford University Press 2019), 599–623. E Örücü, “Methodology of Comparative Law” in JM Smits, ed, Elgar Encyclopedia of Comparative Law (Edward Elgar 2012), 442–454.

Introduction   15 A Riles, “A New Agenda for the Cultural Study of Law: Taking on the Technicalities” (2005) Buffalo L Rev 63(4), 973. A Riles, Collateral Knowledge: Legal Reasoning in the Global Financial Markets (University of Chicago Press 2011). G Samuel, Introduction to Comparative Law Theory and Method (Hart 2014). K Syrett, “On the Value of Lesson-Drawing in Comparative Health law and Policy” (2011) Med L Int’l 11, 45–52. A Watson, Legal Transplants (Scottish Academic Press 1974). JQ Whitman, “The Hunt for Truth in Comparative Law” (2017) Am J Comp L 65, 181–190. HE Yntema, “Comparative Law and Humanism” (1958) Am J Comp L 7, 493. P Zumbansen, “Les Jeux Sont Faits: Comparative Law – As It Really Was Meant To Be?” (2017) Am J Comp L65, 237–255.

Determining The Scope Of The Project GJ Annas, “Health Law at the Turn of the Century: From White Dwarf to Red Giant” (1989) Conn L Rev 21, 551. M Brazier and N Glover, “Does Medical Law Have a Future?” in D Hayton, ed, Law’s Futures (Hart 2000), 371–388. J Coggon, “Would Responsible Medical Lawyers Lose their Patients?” (2012) Med L Rev 20, 130. AM Farrell, C Stanton, A Mullock, and S Devaney, “Pioneering Healthcare Law,” in Farrell et al, eds, Pioneering Healthcare Law: Essays in Honour of Margaret Brazier (Routledge 2016), 1–15. M Hall, “The Legal and Historical Foundations of Patients as Medical Consumers” 9(2008) Georgetown L J 96, 583. T Hervey “Telling Stories About European Union Health Law: The Emergence of a New Field of Law” (2016) Comp Euro Polit 15(3), 352–369. J Montgomery, “Time for a Paradigm Shift? Medical Law in Transition” (2000) Curr Legal Problems 53, 363. T Ruger, “Health Law’s Coherence Anxiety” (2008) Georgetown L Rev 96, 625.

Choice of Contributors, and Invitation to a Collaborative Method J Gordley, “Comparison, Law and Culture: A Response to Pierre Legrand” (2007) Am J Comp L 65, 133–180 R Miller, “On Hostility and Hospitality: Othering Pierre Legrand” (2017) Am J Comp L 65, 191–206. S Munshi, “Comparative Law and Decolonizing Critique” (2017) Am J Comp Law 65, 207–235.

pa rt I

PA R A DIGM S OF H E A LT HC A R E SYS T E M S , L AW, AND R E GU L AT ION

chapter 2

Pa r a digms of Hea lthca r e Systems, L aw, a n d R egu l ation A Transatlantic Conversation Alceste Santuari and William Sage

1 Introduction How does the law create and sustain healthcare systems? There is widespread agreement among patients, politicians, and health professionals that healthcare is not an ordinary market service subject only to ordinary consumer laws such as caveat emptor. But beyond that broad-­brush agreement lie very different legal and regulatory approaches to the provision of healthcare in different jurisdictions. A simplistic but pervasive caricature suggests that European healthcare systems are based on paradigms of state-­led provision, whereas the United States relies more heavily on market paradigms. Yet even a superficial investigation suggests a more complex relationship among law, regulation, markets, and health systems. The connection between the public and private spheres in healthcare regulation remains important, but differences between European and American approaches should not be overstated. As the United States has moved toward debating national health insurance (often styled ‘Medicare-­for-­All’), European nations are more frequently supplementing national commitments to provide healthcare with private mechanisms. While both Europe and the United States rely on a mix of private and public mechanisms, there remain profound differences in the way each balances individualism with solidarity and in the extent to which the profit motive drives the provision of healthcare services. Europeans can count on access to very good healthcare regardless of their economic circumstances, race, or other status, and the contours of their healthcare systems are little affected by the desire of healthcare providers to maximize income.

20   Alceste Santuari and William Sage While the Affordable Care Act (ACA or Obamacare) has done much to eliminate barriers to access in the United States, and most Americans can count on access to the best healthcare that money can buy, tens of millions of Americans still lack health coverage. Especially for residents of Florida, Texas, and other US states that declined federal support to expand their Medicaid programs, one must either ‘earn’ one’s coverage—whether by securing a job with healthcare benefits, accumulating enough wealth to purchase health insurance in the private market, paying social insurance contributions while working, or serving in the military—or one must be thought sufficiently ‘deserving’ of government assistance (e.g., children, seniors, or persons who are pregnant or disabled). And whether one is viewed as deserving can depend on attitudes that disfavor racial or ethnic minorities. In addition, doctors and hospitals are most interested in providing care that generates substantial revenues, neglecting preventive care and early interventions that can do much to promote patient health but are not very remunerative. This chapter takes the form of a conversation between a European scholar of health law and policy and his American counterpart. It considers six paradigm concepts which are important in the structures of healthcare systems and how the law seeks to support and encourage the effective provision of healthcare to relevant populations. These concepts are quality/accessibility, health citizenship and a ‘right to health’, individual liberty, market competition, economic productivity, and ‘health justice’ and integrated social services. The chapter then turns to discuss five structures for health system regulation: public provision of health services, publicly funded but privately supplied health serv­ ices, self-­regulating health professions, public–private partnerships, and nonprofit and charitable organizations. Finally, the chapter compares European and American approaches to the geographical dimensions of health system law and regulation, discussing paradigms of federalism and localism. A theme or ‘red thread’ throughout the chapter is the extent to which different approaches to the planning of healthcare are seen as legitimate and necessary. If the goal is an effective healthcare system that protects the health of both patients and populations, with health viewed from varying perspectives (as an attribute of community, a key factor in a successful economy, a question of equity and justice, or even a human right), how is financing healthcare best approached? Options might include aggregate budgeting, uniform payment incentives, approval of capital investment, and geographically decentralized or professionally driven transactions.

2  Paradigms of Healthcare Systems Law 2.1  Quality Versus Accessibility Regulation of healthcare delivery is often framed as the challenge of maintaining quality of care while assuring access to services within budgetary constraints. National healthcare

Paradigms of Healthcare Systems, Law, and Regulation   21 systems approach this problem in various ways. European health systems have been described as prioritizing egalitarian access over the generation of new technologies and specialised services, at least within the publicly financed sector. The United States, by contrast, has been thought to prioritise supply over access. But improving a healthcare system’s fairness and quality are not necessarily in tension, as illustrated by a visual image currently popular in US and European health policy presentations. In the drawing, a child, a teenager, and an adult are trying to watch a sports game over a fence.1 Giving each spectator one wooden crate on which to stand is portrayed in the graphic as foolish ‘equality.’ Giving two crates to the child, one to the teen, and none to the adult— portrayed as sensible ‘equity’ in the graphic—is in fact also efficient.

2.1.1  Europe A traditional analysis of a European approach to regulating a healthcare system might emphasise two related elements: a state-­led responsibility to provide through social insurance or taxation necessary healthcare and a right of citizens and those lawfully present on the territory to access such healthcare. While the precise modalities of how such obligations and rights are fulfilled differ, these were typically seen, from a legal perspective, as public activities arising from notions of social solidarity and equality. The role of the market, and individual patient choice, was minimal. Cost, access, and quality were centrally determined through political decisions concerning the proportion of gross domestic product (GDP) or national borrowing to be spent on healthcare. Over time, as health technologies improved and became more costly, and as European populations enjoyed longer average lifespans, healthcare systems in Europe struggled more to afford universal access, with some parts of Europe lagging behind in terms of the quality of care they were able to offer their populations. This was exacerbated by the 2007–2008 global financial crisis, which caused financial strain sufficient to challenge the sustainability of healthcare services.2 Effectiveness and efficiency of services, of organizational models, and of processes as well as of the procedures with which healthcare providers must comply are seen throughout Europe as essential and strategic parts of enforcing the right to health. Effectiveness and efficiency are inevitably linked to specific organizational models through which integrated care is to be implemented. A lightly regulated market is not seen as an efficient option. Rather, local and national health authorities focus on structural or procedural approaches to achieving a balance between quality and accessibility. Even with an efficient health system, tradeoffs between quality and accessibility may be needed. New healthcare technologies can strain the budgets of even well-­designed systems, and inefficiencies in other parts of the economy may limit the resources available for healthcare. European systems therefore may be subject to healthcare rationing.

 http://interactioninstitute.org/illustrating-equality-vs-equity/.   See K. Syrett, Healthcare Rationing and the Law, in European Health Law (A. den Exter, ed., Maklu 2017) ch. 8. 1

2

22   Alceste Santuari and William Sage The experience of Italy is illustrative. As in other European states, the debate over rationing is not an outcome of the past decade’s financial crisis. It dates back to the early 1990s, when the Italian government started to connect a ‘health basket’ to the necessary economic resources. The 1992 Health Reform Act attempted both to ensure protection of the right to health and to cap the resources used to implement that right. This legal approach did not make it clear whether conservation of resources should be regarded merely as an assurance of access or as a separate objective to be accomplished. Such a doubt was the reason why many regional governments appealed to the Supreme Court questioning the Reform Act.3 Subsequent rulings of the Supreme Court and the 2001 Constitutional Reform Act, which defined a duty on the part of the central government to ensure all citizens an equal and universal access to health, confirmed that financial restrictions have not gained the upper hand over the right to health. In other words, the Italian legal and healthcare systems continue to recognize that the right to health consists of an irreducible core. Still, the right to health has been heavily challenged by an even more recent Constitutional provision, adopted in 2012, that places stringent financial restraints on both central and local governments. Under Section 81, the Italian Constitution, which was founded on a clear conception of the welfare state, no longer allows public authorities to go into debt. Such a prohibition has triggered a fierce debate concerning the possibility of disregarding this provision as applied to healthcare services because of citizens’ right to health. Indeed, the requirement of financial balance that Section 81 introduced in the Italian legal and healthcare system seems inconsistent with a constitutional obligation to grant citizens, especially those who are most vulnerable, social and health rights. Limiting those rights unless economic or financial circumstances allow for them would subject the healthcare provided to financial and budget restraints. Such a risk is all the greater given the contemporary context in which both European and national economic consequences seem to prevail over community solidarity and social cohesion. COVID-­19 has highlighted tensions between conventional measures of healthcare quality and population-­ wide accessibility and effectiveness. Although the Italian National Health Service (NHS) has shown resilience overall, some areas of the country have been coping better than others with the coronavirus. An excellent regional health system in Lombardy came near to collapse under the pressure of the pandemic, partly as a consequence of its managerial and competitive healthcare vision. Public health there has been largely construed as a productive sector engaged in developing specialized excellence through private, for-­profit hospitals and clinics that have been progressively absorbing a significant share of public finances. The Lombardy regional system has also attempted to reduce the role and function of general practitioners (GPs) who were to be replaced by specialist care programmes entrusted to private organizations or public health agencies, an attempt that met resistance from both GPs and the families they serve.

  See ruling No. 355 of 1993.

3

Paradigms of Healthcare Systems, Law, and Regulation   23 This approach to care was neither capable of coping with unforeseen circumstances nor equipped with forecasting systems to systematically monitor citizens’ health and respond in a timely way. Lacking alternatives, patients affected by the coronavirus turned to hospitals (often referred to emergency departments by marginalized GPs), which quickly reached capacity. The dramatic situation was broadcast around the world but gave rise to only a general litany of costly system improvements to increase capacity and preparedness: increasing the number of medical personnel, increasing medical equipment, and improving protocols and guidelines. In view of the current wave of COVID-­19, it seems clear that public authorities have not paid enough attention to implement a territorial response to the pandemic. Local health authorities should not ask all health facilities to convert into ‘COVID hospitals’ to increase the number of beds available for COVID-­infected people. Such an obligation risks compromise of the capacity of these facilities to cope with diseases other than COVID-­19. These will, in their turn, be postponed once again with the consequence of lengthening waiting lists and, accordingly, of swelling inefficiencies.

2.2.2  United States In the public’s mind, parameters of cost, access, and quality define US healthcare serv­ices as an optimised system in equipoise. Any substantial effort to universalise access to healthcare through a structured, affordable public insurance entitlement—which the United States alone among developed countries lacks—seems to risk ‘levelling down’ the quality of care. From this perspective, efforts to improve social equity in the US healthcare system would almost certainly reduce market efficiency. In the extreme view, government rationing of potentially life-­saving services would be widely implemented—the terrifying, if apocryphal, ‘death panels’ that the ACA was rumoured to impose. Accordingly, traditional regulation of healthcare system emphasises three functions designed to protect quality: (1) the selection, training, and self-­regulatory oversight of an expert, ethical cadre of physicians; (2) the creation and maintenance of healthcare facilities that meet high standards for performance articulated by those physicians; and (3) the development and dissemination of innovative diagnostic and therapeutic modalities that those physicians will deploy. These are typically described as private activities that government will support generously but police only at the fringes so as not to interfere with the presumptively efficient processes by which patients’ needs are met by their chosen physicians or with the competitive marketplaces that purport to make that possible. Health insurance has been more often a facilitating than a limiting force with respect to physicians, hospitals, and technology. Americans may receive private health coverage as a fringe benefit of employment, typically with large government subsidies administered through the income tax system, or they may be eligible for public insurance, notably Medicare (for the elderly) and Medicaid (for the poor, subject to conditions that vary by state). Both private and public insurers enforce rules and restrictions, but the essential character of medical services regardless of payer is determined by the norms and preferences of healthcare providers, particularly physicians.

24   Alceste Santuari and William Sage The problem is that extensive research has revealed that the supposed tradeoff between quality and access is incomplete and misleading. According to a 2012 report of the Institute of Medicine, the US system wastes US$750,000,000,000 annually as a result of excessive prices, unnecessary or poorly delivered services, redundant administration, and fraud. Historical path dependence, avoidance of political and fiscal controversy, and catering to the financial self-­interest of the ‘medical-­industrial complex’ emerge as more persuasive explanations for cost pressures than technology-­driven scientific necessity. Seen from this angle, affordable access need not reduce quality because the current healthcare system is far from optimal. This untapped potential underlies the ‘Triple Aim’ that has become the mantra of many US health policy experts: simultaneously improving the patient experience of care, improving the health of populations, and reducing the per capita cost of care. Guild-­ protective regulation of physician practice offers a simple example. Occupational licensure has proliferated in the United States over the past several decades, doubly disadvantaging poorer people by both reducing access to affordable services and curtailing educational and economic opportunity. Economist Milton Friedman pointed out, in 1962, how many benefits could flow from liberalizing medical licensing, including increasing average quality of care when one took account of those unable to afford physician services today. And decades of research has demonstrated that basic medical services can safely and effectively be delivered by others (e.g., advanced practice nurses). Yet many scientifically unjustifiable restrictions remain in place because they benefit established interests. Moreover, entrusting the healthcare system to physicians caring for their individual patients has medicalised many social problems while leaving others unsolved—a problem that also afflicts healthcare systems in Europe and elsewhere. In the United States, because of structural biases in the nominally private medical system and neglect of ‘social determinants’ in the residual public arena, there are marked disparities in both medical care and underlying health based on factors such as race, poverty, and lack of education. Rather than brokering a compromise between efficiency and fairness, regulatory deference to the norms of the medical profession has simultaneously retarded both marketization and human rights. An obstacle to improvement is that ‘quality of care’ is generally associated not with ease of access to routine and preventive services but with new, physician-­mediated technologies that can be layered on top of existing, often hard to access, processes of care. Drugs, medical devices, and biotechnology tend to complement diagnostic or therapeutic modalities delivered in hospitals or other facilities with specialised workforces under physician control or supervision. These treasured but costly discoveries, moreover, become powerful deterrents to a collective commitment to healthcare or health in the United States. Any suggestion, for example, that new drugs be priced closer to their marginal than their average costs of production is resisted as incompatible with future research and development. The unstated problem is that the billion-­dollar cost of bringing a drug to market cannot be recouped in bedside purchase-­and-­sale transactions

Paradigms of Healthcare Systems, Law, and Regulation   25 but instead must reflect a more reasoned combination of patient-­level payments and broader social investments. The COVID-­19 pandemic has cast a harsh light on two aspects of the supposed ­quality–access tradeoff. First, lack of community preparedness has forced healthcare providers into exactly the sort of rationing discussions—with respect to ventilators, medications, and vaccines—that the American devotion to technology was supposed to avoid. And some conservative commentators, forgetting their supposed outrage at the ACA’s wholly apocryphal ‘death panels’ for the elderly, have suddenly embraced the medical abandonment of those same older generations as the necessary price of ­economic recovery. Second, the concentration of COVID-­19 deaths in racially and economically disadvantaged communities has confirmed the urgent need to confront health inequities, as well as revealing that the conventional approach to rationing by prognosis stacks the deck even more against communities with poor underlying health.

3  Individual Liberty, ‘Health Citizenship,’ and the Right to Health Europe and the United States take very different approaches to healthcare regulation as an attribute of humanity, membership in the polity, and indicator of cultural affinity. How might a ‘right’ to health or medical care be formulated and enforced? Does a collective commitment to funding healthcare imply a right? What is the relationship between healthcare regulation, human rights, and patriotism or globalism? Beyond the general relationship between quality and accessibility in the regulation of services, national healthcare systems value the freedom and autonomy of the individual, often expressed in terms of choice among healthcare providers and treatment. Depending on country and context, however, individual liberty may suggest either a greater or lesser role for government in the provision of care. Does the empowerment of a healthcare entitlement free from potential financial catastrophe also further liberty? Does regulatory reliance on the ethics of the health professions support autonomy more reliably than direct government control? Does consumer choice of provider or treatment in a private marketplace enhance self-­determination?

3.1 Europe European constitutions protect individual liberty. But European constitutions also recognize social rights—obligations of the state to provide health, housing, etc.—even if, as in the United States, an individual’s abstract ‘right’ to a specific desired healthcare serv­ ice is not usually enforceable. Public regulation in the healthcare sector is aimed to

26   Alceste Santuari and William Sage accomplish two main goals. The first relates to the legal and organizational arrangements of national healthcare systems within the European Union (EU). Member States are responsible for the frameworks within which health authorities, health professionals, and private organizations are allowed to operate and through which citizens are empowered to access the services supplied. The second goal of regulation is concerned with benefits to citizens-­patients-­users who are not insured with the national system, including the potential inflow of new applicants. Possibly, this is the reason why in Europe regulation is largely accepted as essential to the delivery of healthcare. Across the European political spectrum, regulation in the healthcare sector is rarely criticized as an obstacle to individual freedoms. Rather, the public often supports regulation that can make national health systems adapt more effectively to modern challenges, such as population ageing, financial sustainability, information technology, and the coronavirus. In European contexts, regulation therefore strikes a balance between public authorities’ intervention and private freedom. In this way, individuals’ fundamental rights, including the right to health, are protected against arbitrary, unjustified, or useless interference.4 The right to health is also supported directly by regulation, which ensures that both public agencies and, if applicable, market and non-­market organizations deliver healthcare to citizens and local communities. The notion of individual liberty in these endeavours has a different feel in Europe, where collective structures and processes are understood as securing individuals’ rights. Liberty to depart from those structures is distinctly limited, for example in that obligations to be socially insured and contribute to social insurance or taxation-­funded healthcare systems are not optional. Support for this approach is not universal, however: certain political and economic actors consider social (and perhaps also health) policies as an obstacle to growth and competitiveness. All European states are responsible for implementing national health systems that are effective, accessible, and resilient to ensure European citizen’s right to health. This obligation on the part of national health systems makes public regulation all the more essential and strategic. In the EU legal framework, this mainly consists of the European Social Charter, the European Union Charter of Fundamental Rights, other European soft law measures, national legal and organizational settings, and the rulings of the European Court of Justice. This legal framework has progressively become the actual commitment of both Member States and European institutions. Within the broad notion of freedom of movement of people provided for by EU law,5 moreover, Directive 2011/24/EU positively recognized European citizens’ right to

  The intervention of governments in the economy very much depends on the concept of public good and on historical circumstances. See Ian Payne, The Acceptable Limits of State Interference in Private Property Rights During Periods of Economic Crisis (2010), Comparative Research in Law & Political Economy. Research Paper No. 29/2010. 5   According to a recent survey, the main reasons why European citizens are willing to go abroad to access healthcare services are identified with the lack of treatments at home (71%) and higher quality of services abroad (53%). See European Commission, Special Eurobarometer 425: Patients’ Rights in 4

Paradigms of Healthcare Systems, Law, and Regulation   27 cross-­border care. However, the (high) expectations that Directive triggered, especially as to the potential European ‘free market’ of patients, have been largely frustrated by the implementation process in individual Member States and by the financial crisis of 2007/2008.6 Member States have narrowed the right to health in the face of budget restraints and austerity policies. In some Member States, not only have national health authorities limited the right to cross-­border healthcare, but they have also reduced the actual provision of some healthcare services. Marketization, globalization,7 increasing demand for healthcare services, and the progressive ageing of the population are the underlying forces that undermine the financial sustainability of individuals’ right to health. Sometimes, it seems as though the obligation on the part of governments and public authorities to ensure citizens’ fundamental rights to healthcare is no longer an essential dimension of modern European welfare states.

3.2  United States The ACA’s patchwork approach to universal health insurance coverage created a rallying point for a vocal, retrogressively nationalistic set of conservative commentators and Republican politicians, which facilitated their rapid rise to control of both the federal government and the majority of states by 2017. Part of their opposition was imagined— that Obamacare would institute centralised government rationing of care—but another part was provoked by the ACA’s generous subsidies from rich to poor and from healthy to sick as a matter of national policy. In essence, opponents of the ACA objected to the European model of health as a fundamental right, believing instead that individual liberty and health solidarity are incompatible. This reasoning was evident in the principal legal challenge to the ACA that reached the US Supreme Court, NFIB v. Sebelius (2012),8 a constitutional case that did not directly involve a right to health. The Court’s holding that mandatory purchase of health insurance (but not payment of a tax on remaining uninsured) exceeded the constitutionally enumerated powers of Congress reflected its conservative majority’s commitment Cross-Border Healthcare in the European Union, 18 May 2015, p. 14. See also R. Baeten, Cross-Border Patient Mobility in the European Union: In Search of Benefits from the New Legal Framework, J. Health Serv. Res. Pol’cy 19 (2014). 6   The international economic and financial crisis has caused a worsening in the waiting lists, reduced the availability of healthcare facilities, and increased out-of-pocket payments of healthcare treatments. See EUROFOUND, Access to Healthcare in Times of Crisis (Luxembourg: Publications Office of the European Union 2014) 12 and 17. 7   Globalization ‘has significantly affected the law and economics approach, causing a re-thinking of the mechanisms of balance and un-balance between economic freedoms and individual rights. Accordingly, new systems of regulation are required to adequately and effectively match new social and economic needs.’ m. a. Stefanelli, Prefazione, in Dopo la globalizzazione: sfide alla società e al diritto (m. a. Stefanelli, ed., Torino: Giappichelli 2017) 2. 8   132 S. Ct. 2566 (2012).

28   Alceste Santuari and William Sage to protecting individual freedom by preventing any single sovereign—national or state—from acquiring plenary authority over people’s lives. The national government cannot compel commerce in order to regulate it; state government can only impose such obligations selectively (benefits offered by private employers, for example, are exclusively regulated by federal law). As a result, the closest that the Obama administration came to articulating a collective rationale for its law was consumerist rather than democratic. In remarks at the White House, for example President Obama used the term ‘consumer’ rather than ‘citizen’ to represent the ACA’s commitment to ordinary Americans.9 Collateral constitutional or statutory rights (e.g., reproductive freedom, religious liberty, or protection from discrimination) that might be affected by government pursuing policies approximating a positive right to health or medical care have become proxy battlegrounds for adjudicating the right that does not exist. For example, the ACA’s attempt to ensure contraceptive coverage for individuals who access health insurance through employment has provoked seemingly endless litigation, including several cases heard by the US Supreme Court, with employers asserting moral or religious objections to forced ‘complicity.’10 On the other hand, ‘health citizenship’ or a ‘right to healthcare’ has no clear meaning in the United States. Being an American citizen, or being lawfully resident or even merely present in the United States, does not include an expectation of medical entitlement except after one ages into the Medicare-­eligible senior population. Although selected communities may regard their local hospitals with affection, there is no civic pride associated with any organised set of healthcare providers at the national level or in any US state. As the United States’ halting response to COVID-­19 has shockingly demonstrated, the commitment and capacity of its healthcare system to protect the population against communicable diseases or environmental toxins is tacit, untested, and sometimes illusory. Ironically, the Trump administration elaborated a divisive version of ‘health citizenship’ as part of its harsh anti-­immigrant rhetoric, attributing disease risk to immigration and demanding of those seeking to enter the country proof that they could pay for any healthcare they require.11 Constitutional rights in the United States tend to represent domains in which individuals are free from government rather than obligations of government to those ­individuals. Consequently, the implicit individual ‘right’ to healthcare in the United States consists of decisional authority that is often impossible to exercise, either because the information and dispassion on which a medical choice should rest are absent or because the financial resources necessary to operationalize a choice are unavailable. Affirmative rights to healthcare services are not administratively enforceable except in the limited

9   President Barack Obama, Remarks in the Rose Garden of the White House on the Supreme Court’s Decision in King v. Burwell (June 25, 2015). 10   Little Sisters of the Poor Saints Peter and Paul Home v. Pennsylvania (consolidated with Trump v. Pennsylvania), Docket No. 19–431 (2019). 11   Kathleen R. Page, Maya Venkataramani, Chris Beyrer, & Sarah Polk, Undocumented US Immigrants and Covid-19, 382(22) N. Engl. J. Med. e61(1)–e61(3) (2020).

Paradigms of Healthcare Systems, Law, and Regulation   29 circumstance where a specific entitlement to a substantial government-­funded health insurance benefit exists (e.g., Medicare or Medicaid), or (under the Emergency Medical Treatment and Active Labor Act [EMTALA] statute) should an individual seek emergency care in a hospital. With respect to healthcare regulation, the closest the United States comes to population-­ based oversight through dedicated facilities is the Veterans Health Administration, a federal agency serving former members of the military, and the Indian Health Service, a chronically underfunded set of federal clinics for Native American tribes living on geographically isolated, self-­governing reservations. The national Medicare program insures the senior population and a few others through diffuse private providers of care that are subject to complex conditions of payment but little substantive federal regulation. The same is true for Medicaid, a jointly funded but state-­ administered program insuring those categories of poor Americans deemed deserving of public support. Neither coverage system enables beneficiaries to access care without extensive paperwork or the potential for residual out-­of-­pocket cost. No American state has a coordinated healthcare system for its entire population, though a few such as Hawaii (an island chain), Massachusetts, Oregon, and Vermont have taken steps in that direction. Hospitals and clinics in small towns and poor urban neighbourhoods often take functional responsibility for serving their communities, but they are not formally charged with that obligation. Some health maintenance organizations (HMOs) provide comprehensively for their members’ needs in focused geographic areas, but the beneficiaries of coverage offered through most health insurers and large employers (including public employees) are scattered across the country and are served in a fragmented fashion by private providers. Even the TRICARE program, which covers active duty US military personnel and their dependents, does so largely through private provider networks that have contracts with approved health plans, although at no cost to beneficiaries and with far less paperwork than a typical private insurer would impose. Paradoxically, the first public mobilization in response to the COVID-­19 pandemic undercut US health solidarity rather than reinforcing it. Protests—many funded by wealthy conservative political donors and fomented by right-­wing media (and, predictably, President Trump)—targeted health-­minded state governors as ‘authoritarians’ for restricting commercial activity and enforcing social distancing measures.12 Face coverings intended to signal cooperation and slow disease spread were resisted as incursions on fundamental liberties, with disagreements sometimes resulting in violence. Ethnic groups, notably Asian Americans, faced anger and discrimination. Immigration was further restricted. Although public opinion has shifted in favour of universal health coverage and greater investment in public health preparedness, the pandemic-­related economic contraction required massive stimulus spending by Congress. This may 12   See, e.g., Mary McCord, Trump’s ‘LIBERATE MICHIGAN!’ Tweets Incite Insurrection. That’s Illegal. Wash. Post. (Apr. 17, 2020). https://www.washingtonpost.com/outlook/2020/04/17/ liberate-michigan-trump-constitution/

30   Alceste Santuari and William Sage preclude longer term financial commitments to taxpayer-­funded healthcare and insurance coverage as the pandemic recedes. In sum, there are few signs of a right to health or healthcare on the horizon in the United States, although a counter-­movement favouring government-­sponsored insurance emerged on the political left after the 2016 election losses. The progressive rallying cry of ‘Medicare for all’ reinvigorated a long-­standing critique of commercial health insurance but was misguidedly nostalgic in many ways, such as relying excessively on physician altruism and underestimating the political obstacles to converting trillions of dollars of nominally private transactions into ones that appear explicitly on the government’s accounts as taxation and spending. Still, the Medicare-­for-­All movement might signal that statements of principle, whether framed as safeguarding rights or as redressing injustices, are once again relevant to a national health reform conversation that for too long consisted of only pragmatic, incremental expansions of coverage without any explicit sense of social justice or solidarity.

4  Market Competition and Economic Productivity Whether competitive outcomes are an explicit goal of healthcare regulation is open to discussion. Certainly, transactional freedom is a touchstone for liberty more generally, and opportunities for private commercial success help safeguard democratic values. Yet systems of subsidised health insurance distort market processes in many ways, as do information asymmetries, vulnerabilities, and often dependence inherent in the patient role. Professional practices can also be profoundly anti-­competitive. How, then, do healthcare regulators define and oversee market processes to optimize their functioning within larger system dynamics? Through competition and in other ways, health system regulation has implications for national economic growth. Most intuitively, a healthier population will tend to be more economically productive. Causality runs in the opposite direction also, perhaps more strongly: a wealthier society will be healthier. As a result, the economics of healthcare investment may be complex, as are the distributional effects. Healthcare needs are distributed according to population patterns, not economic wealth, whether that be in the US ‘Rust Belt’ or on Europe’s poorer Eastern and Southern periphery. In both Europe and the United States, the healthcare system can be a significant employer—of low-­paid general service workers and well-­educated, well-­heeled health professionals. Does national investment in health draw resources from other areas with potentially greater economic payoff? Are recipients of healthcare more or less likely to be economically productive? Does job creation in healthcare and the education necessary to achieve it represent a clear benefit to individuals and communities, or does it distort market processes and limit growth? Does the broad geographic distribution of healthcare employment within countries compensate for any efficiency loss?

Paradigms of Healthcare Systems, Law, and Regulation   31

4.1 Europe Economic regulation in the form of market competition assumes that regulated markets are capable of supplying all of the utilities that citizens ask for. Subsequently, private companies operate on the market according to their economic interest, thus exercising their freedom of economic initiative. Conversely, public authorities may decide to take action in case of market failures or when they believe that the market is not capable of adequately matching citizens/consumers’ demand for health. This is typically the case for most healthcare services in European contexts: only a few services, such as cosmetic surgery or some types of human reproduction treatments, are deemed appropriate to be governed only by the regulated economic market. But public action to provide services of general interest, such as healthcare services, does not necessarily imply that public authorities directly produce and deliver those serv­ ices. Rather, service delivery for which the state is ultimately responsible can be entrusted to a private company, a public–private partnership, or contracted out to a nonprofit organization. Private organizations will then operate to accomplish the commitments that they accept to fulfil within a specific relationship agreement with public entities. Whatever the relationship, under EU law, private organizations are often granted a special legal status that allows them to operate on behalf of the relevant National Health System.13 The inclusive attitude toward healthcare that overall EU policy promotes is based in part on the assumption that economic growth and health protection are linked together.14 In this respect, EU Member States are called upon to encourage innovation in the healthcare sector, enhance sustainability of healthcare systems, improve citizens’ health, and protect them from international health threats. The linking of economic prosperity and good health is also found in health policies and structures in other European states. Notions of solidaristic investment in European health systems and new technologies are sometimes justified by reference to the benefits associated with sustainable growth. The Italian experience of COVID-­19 has increased calls for a better equipped and more generously funded healthcare system that is more consistent with the new epidemiological patterns that the pandemic has revealed. Reorganization would require overcoming the current paradigm, which is centred on high-­ performing hospitals competing to deliver services to individuals. It would be replaced by a community-­ based approach that promotes public health at all levels and includes not only individuals as citizens or patients but local populations as a whole. Regardless of the exact strategy chosen, policy makers and experts believe that Italian healthcare organization should build stronger local networks to make it more effective, efficient, and equitable. Heightened awareness during COVID-­19 may make this the right moment to act lest political and organizational inertia worsen current shortages because such an approach

  See ECJ 2013/XX, Fédération des maisons de repos privies de Belgique (Fermabel) ASBL v. Commission communautaire commune de Bruxelles-Capitale, 11 July 2013 (C-57/12). 14   COM (2011) 709. 13

32   Alceste Santuari and William Sage would require a substantial redirection of public funds and overall health expenditure away from the hospital sector.

4.2  United States Despite the veneer of market competition, problems such as persistently rising costs, economically ungrounded prices, middling quality, and inaccessible service delivery strongly suggest that healthcare consumerism and economic efficiency are not well aligned even in the supposedly market-­based US system. The productive efficiency of US healthcare particularly seems questionable, although scholars differ regarding its amenability to improvement. Healthcare regulation often has done more to make things worse, both by stifling competition through professional protectionist laws and by failing to police provider mergers and other private anti-­competitive practices. To be sure, with increasing consolidation among hospitals, physician practices, and private health insurers, calls have intensified for stronger market oversight through antitrust (competition) law. But American antitrust law is directed at restoring competitive processes rather than dictating competitive outcomes. Hence, it is poorly equipped to revisit existing regulatory settlements. The regulation itself must change. COVID-­19 arguably has done more to reduce regulatory barriers to competition in a few months than had occurred in decades of political manoeuvring. Reversing long-­ standing restrictions put into place mainly by medical licensing boards, the governors of US states and state agencies acting under emergency public health authority have expanded telehealth services, eliminated documentation requirements, enabled out-­ of-­state (and even non-­US) physicians to treat patients, and removed additional burdens on nurse practitioners and other non-­physician health professionals.15 The federal government has augmented state reforms by expanding payment for these services under Medicare and Medicaid. Although this liberalization of provider regulation is denominated as temporary and organized interest groups are already pressuring states to return to the status quo ante, the benefits of enhancing access through greater competition are likely to alter the political dynamic going forward. For example, states that have joined interstate compacts or formed regional alliances to help overcome pandemic-­related inertia at the federal level may continue their newfound mutuality after the crisis eases. An important but understudied trend in the United States is the rapid, substantial entry into the healthcare sector of technology conglomerates, private equity, and other core capital market participants that manage or supply healthcare providers. Until recently, the investor community limited its interest in healthcare to new technologies and other ancillary products or services with familiar business characteristics—taking 15.   Many states, for example, have expanded practice authority for nurse practitioners and similarly qualified health professionals, albeit on an ‘emergency’ basis. https://www.aanp.org/advocacy/state/ covid-19-state-emergency-response-temporarily-suspended-and-waived-practice-agreement-requirements

Paradigms of Healthcare Systems, Law, and Regulation   33 advantage of the reliable revenues associated with healthcare but avoiding its professional and regulatory complexities. Occasional aberrations tended to prove the rule, including for-­profit hospital chains that invested heavily in California real estate in the 1960s, tax-­favoured Real Estate Investment Trusts in the 1980s, physician practice management companies in the 1990s, and health information technologies in the 2000s. This appears to have changed. Private equity firms now own significant stakes in specialty medical practices, urgent care centres, and other provider organizations. Unusual corporate combinations such as the CVS-­Aetna merger blend the delivery of medical services with broad retail operations. And global corporate leaders continue to explore disruptive interventions into medical markets, although the much-­touted Haven joint venture among Amazon, Berkshire Hathaway, and JP Morgan Chase is winding down its operations. The regulatory questions associated with these developments have yet to be asked, much less answered. As is the fundamental issue of whether such new ventures have as their principal goal to remake the US healthcare industry through competition or merely to siphon off large economic rents that had previously remained within the professional and nonprofit sectors. In combination with the previous European discussion, this example raises the useful construct of a ‘residual market’ in national health policy that necessarily varies by country in its size, characteristics, expectations, and constraints. In some countries, the publicly and privately funded healthcare sectors operate nearly independently of each other, with patients divided along socioeconomic lines. In other countries, the privately funded healthcare sector supplies only a small number of largely discretionary services. Competitive incentives have a place even within predominantly public systems, although the terms of competition may require adjustment over time as context and performance evolve. The COVID-­19 experience, moreover, makes it clear that an effective response to a systemic challenge such as a pandemic may require better coordination and greater engagement of economically disadvantaged communities than private competition among healthcare providers can achieve. The United Kingdom offers an interesting contrast to the United States. The transition in recent decades of England’s National Health Service (NHS)—though not replicated in Scotland or Northern Ireland—to more of a competitive footing is often described as the NHS’s internal market. The competitive conditions under which it operates and therefore the outcomes it generates are tethered to and circumscribed by the detailed regulatory scheme for the NHS as a whole, and are monitored closely by government. By contrast, unconstrained spending without clearly specified objectives has taken the supposed ‘free market’ in American healthcare in profoundly unnatural directions. Healthcare competition in the United States is heavily distorted by regulation, but with far less public coordination or oversight. The competitive structures, processes, and outcomes of the US residual market therefore are unpredictable—making US healthcare operate as an ‘emergent’ system. ‘Juiced’ by massive public subsidies, private competition routinely strays from expected patterns of purchase and sale and easily spins out of control. For fans of American cinema, if the NHS internal market is ‘The Truman Show,’ that in the United States is ‘Jurassic Park.’ It is too soon to tell

34   Alceste Santuari and William Sage whether the healthcare system that follows the COVID-­19 pandemic will behave less chaotically. Broadly considered, the relationship among medical care, health, and economic productivity in the United States is a challenging one. Healthcare accounts for approximately one-­sixth of the economy and roughly 1 in 10 American jobs, many in the several thousand hospitals and ancillary businesses that dot the US landscape. America’s nearly 4 million registered nurses and 700,000 active physicians each account for more aggregate salary dollars than any Bureau of Labor Statistics occupational classification except ‘managers’ and ‘chief executives’ (including small business owners). The earnings of US nurses, considered on a stand-­alone basis, exceed the total GDP of half of American states. More generally, patterns of medical investment and employment in the United States and the political interest groups that protect them tend to attenuate what might otherwise be a strong connection between health at the population level and economic productivity. The US healthcare system prioritizes acute care over both primary and secondary prevention, while investing public funds mainly in the elderly who have Medicare coverage. The Medicaid program is an exception, especially with its emphasis on preventive services and care for children and child-­bearing women. But Medicaid, although it pays for one-­third of births, does not seem equipped to address major racial disparities in maternal and infant mortality that make the United States a shameful negative outlier in member states of the Organisation for Economic Cooperation and Development (OECD). The US health system’s overall inability to prevent chronic disease in communities, particularly those suffering persistent, compound socioeconomic disadvantage, takes a significant toll on healthy, productive working lives.

5  ‘Health Justice’ and Integrated Social Services There are several determinants of health, including genetics, behaviours, social circumstances, and environment. Since the end of World War II, however, the principal focus of national social policy in developed countries has been on medical care, particularly expensive acute care services. Attention is once again being paid to population health and to the integration of medical and social services for individuals and communities. Can social services offset unjust disparities in access to medical care? How are healthcare systems in Europe and the United States responding to this challenge through revised regulation?

5.1 Europe It has been clear for some time that working on traditional healthcare patterns to meet rising demand is not the answer. National health services need to work differently by providing services that match community needs and breaking down barriers between

Paradigms of Healthcare Systems, Law, and Regulation   35 health and social services.16 Local governments and local health authorities therefore have begun to adopt policies and deploy tools that take a holistic approach to individual and collective needs. This philosophy inspires ‘place-­based’ integrated care systems— these are expected to overcome the traditional dichotomy between health and social care action. Although there seems to be ‘no universally accepted definition of integrated care, no one model of care that can be replicated locally, and little evidence to tell us that it works,’17 integrated care is largely understood as a public system that combines both healthcare provisions and other services such as vocational training, social work, and social housing. The ultimate aim of integrated care is to offer a comprehensive or holistic approach to patients’ demand for health. To make integrated care systems effective, accountable, and sustainable, they could consist of both public entities and private organizations, especially nonprofit ones, that might enforce the right to health services while also promoting public health. Although local and health authorities would maintain all their responsibility as to the organization and arrangements of healthcare services, new private actors can be incorporated and important changes in the organization of traditional players could be introduced.18 The integration of health and social care services relies on one of the main pillars of the programming function: a legal and organizational framework within which local health authorities and municipalities can cooperate. This is all the more important in the face of the ever-­increasing complexity of individuals’ needs and demand for health. In building up integrated care within communities, local and health authorities and nonprofit organizations work in two main phases: programming and delivery of serv­ices. In the first phase, the quantity of services to be produced and distributed, the relevant beneficiaries, and the needed resources are determined and reported. In the second phase, the legal and organizational requirements for citizens to access healthcare serv­ices are stated. Municipalities, local health authorities, and nonprofit organizations would seek to make decisions and implement them jointly through a programming action plan. These plans would replace a number of different planning documents that have been both ineffective and burdensome for patients and citizens. The local programming action plans depict an integrated care scheme in which public authorities are bound to cooperate with each other and with nonprofit organizations. Effective coordination enables the planners to predict the quantity of services and provisions to deliver and leads to a better selection of nonprofit organizations to entrust with doing so.19 Integration between   See C. Ham, Making sense of integrated care systems, integrated care partnerships and accountable care organisations in the NHS in England, The King’s Fund (20 February 2018). 17  N. Goodwin, Understanding Integrated Care: A Complex Process, a Fundamental Principle, Int. J. Integr. Care 13 (Jan-Mar 2013). 18   On this issue, see IBM Institute for Business Value, La sanità e l’assistenza sanitaria nel 2015. Evoluzione dei modelli di erogazione dei servizi sanitari, https://docplayer.it/63963-La-sanita-e-lassistenza-sanitaria-nel-2015.html 19   Although the voluntary, community, and social enterprise sector is increasingly valued for its contribution to the health and care of individuals and local communities, it has not yet realized its potential as an equal partner with the public sector. See Department of Health, NHS England, Public Health England. Joint Review of Partnerships and Investment in Voluntary, Community and Social Enterprise Organisations in the Health and Care Sector. (2016). http://www.gov.uk/government/ publications/review-of-partnerships-and-investment-in-the-voluntary-sector 16

36   Alceste Santuari and William Sage health and social care services and institutional cooperation between different public actors are pursued through funding strategies which seek to defeat the ‘fortress syndrome’ of the various services (‘siloes’ in American parlance), which are usually thought of as sectorial and self-­centred. The effectiveness of integrated care action plans depends on the capacity of public authorities to develop common projects and shared procedures. Public actors must agree on priorities. The higher the level of coordination and dialogue among the different public bodies involved, the more effectively projects can be defined, regulatory frameworks strengthened, and programs implemented. Three levels of integration between health and social care services are essential. The first level relates to the institutional partnership that different public authorities are willing to undertake by stating shared goals and subscribing to partnership agreements and protocols. The second level is concerned with managerial integration between the different services, including methodologies for assessment by users-­patients. The third level implies a professional integration that involves the establishment of cross-­cutting teams, which are charged with governance as well as with ensuring the quality and accessibility of community services. An example of effective integrated care systems is represented by the so-­called ‘Health Houses’ in the Emilia-­Romagna Region in Italy. These facilities group together some healthcare functions, except for severe pathologies to be treated in hospitals, and some social care provisions. The facilities include the activities of GPs and a 24-­hour medical service, consisting of blood test laboratories, physiotherapy services, first aid, and pharmacies. The aim of the Health Houses is to provide a ‘citizen-­centred’ care system that can overcome the traditional hospital-­centred approach. Another example comes from Great Britain, where, since 2016, sustainability and transformation partnerships have been designed to improve health and care for patients. In some areas of the country, a partnership will evolve to form an integrated care system, a new type of even closer collaboration. In an integrated care system, NHS organizations, in partnership with local councils and others, take collective responsibility for managing resources, delivering NHS standards, and improving the health of the population they serve. Areas like Greater Manchester that had a head start in working as systems represent a successful context in which NHS providers and commissioners have established a track record of good performance. Conversely, areas where there has been a history of competitive behaviours or no previous experience of organizations working together are taking longer to realize the benefits of integration.20 In Italy, the COVID-­19 pandemic has revealed that different approaches to integrated health and social care systems have a dramatic effect on public health emergencies. An effective response to the pandemic requires a strong and sound relationship between hospitals, general practitioners, and territorial preventive medicine, as well as social care of the elderly. Where that strong relationship is in place, the healthcare system is  https://www.kingsfund.org.uk/blog/2018/03/progress-report-integrated-care-systems

20

Paradigms of Healthcare Systems, Law, and Regulation   37 more capable of monitoring and responding to the spread of the pandemic. By contrast, regional healthcare systems with excellent hospitals but weakly integrated systems have failed to effectively cope with the spread of the pandemic. Going forward, health authorities in Italy may decide to act along four main paths. The first relates to the programming of actions: public authorities at the local level and nonprofit organizations are summoned together to identify priorities and projects to be completed. This methodology is provided for by Italian law concerning nonprofit organizations whose strong rooting within local communities positions them to raise awareness about social needs. The second involves decisions concerning the contracting-­out of services and provisions, as opposed to producing them within public agencies. The third includes the definition of organizational models by which to realize integrated care services at the community level. The fourth, which is consistent with EU law, supports the active engagement of nonprofit organizations in the provision of integrated care services. Through non-­competitive procedures, such as direct partnership agreements or accreditation,21 nonprofit organizations are selected to deliver a wide range of health and social care services on behalf of the public welfare system.

5.2  United States Integration of medical and non-­medical social services has been slow to develop in the United States. Or, rather, to redevelop. As the largely uncontrolled spread of COVID-­19 in the United States demonstrates, over the past half-­century core public health functions have been eclipsed by the rapid expansion of high-­technology medical care. This shift gives the illusion of income progressivity because of high annual spending on hospital care for the poor under the Medicare and Medicaid programs. Yet basic indicators of population health, such as life expectancy and maternal or infant mortality, failed to improve in parallel with medical investment. As a result, health spending skews less toward poor individuals over the course of their lifetimes; more importantly, lack of social support condemns many poorer communities to higher rates of disease, reduced educational and economic opportunity, and shorter lifespans. High medical spending arguably draws public funding away from non-­medical social services, such as housing and education, that would have even greater health benefits. Research studies are inconclusive, with some experts suggesting that American investment skews more towards the medical than the social, while others finding that countries such as the United States that spend more on health also spend more in related areas.22 Still, the crowd-­out effects of high medical spending seem compelling at the level of state government in all but the wealthiest parts of the country. State governments cannot lawfully incur fiscal deficits, and poor economic conditions simultaneously   See ECJ 2013/XX.   E. H. Bradley & L. A. Taylor, The American Healthcare Paradox: Why Spending More Is Getting Us Less (New York: Public Affairs 2013). 21

22

38   Alceste Santuari and William Sage reduce tax receipts and expand the number of residents who are eligible for medical assistance. This sets up a pernicious competition between publicly funded serv­ices such as education and publicly funded medical care in most state budgets, pitting young against old and sacrificing economic welfare over the longer term. Moreover, well-­ educated, prosperous individuals and communities are markedly healthier than their less-­educated, poorer counterparts. The US healthcare system’s ostensibly private character, its reliance on decentralized private charitable organizations delivering social services, and its devolution of many governmental functions to local subdivisions such as independent school districts all tend to discourage coordination. In the aftermath of the COVID-­19 pandemic, a serious commitment may emerge to address social determinants of health through partnerships between medical and non-­medical entities. At the national level, the US Department of Health and Human Services has adapted a population-­based, primary care–oriented service delivery model for Medicare beneficiaries (accountable care organizations, or ACOs) into a geographically distributed set of pilot programs to address health-­related social needs for both Medicare and Medicaid beneficiaries; these are known as accountable care communities. Several states are experimenting with comprehensive primary care service models that encompass social care, while federal health programs increasingly permit payment for non-­medical ‘enabling services’ such as transportation, housing, and legal representation. Building on ACA requirements for community needs assessment, acute care hospitals are investing in the provision of housing, transportation, nutrition, and other services that might reduce utilization of high-­cost emergency and in-­patient care. A variety of philanthropic organizations, both private foundations and public charities, are launching similar programs. New workforce models (such as community health workers) and new information technologies (such as integrated patient portals) are being developed and deployed to facilitate these efforts. COVID-­related ‘emergency’ legal changes that liberalize workforce regulation and authorize insurance coverage and payment for transportation, telehealth, and other community-­based facilitation of care may accelerate the timeline for social service integration. ‘Place-­based initiatives’ are gaining adherents in the United States because of their potential to connect the delivery of medical services to the underlying health of both individuals and populations. Healthcare and public health are separately governed in the United States, with only a few points of connection until the recent shift of attention toward social determinants and population-­based outcomes—and then the tidal wave of interest in public health because of COVID-­19. Public health remains predominantly a state and local matter, although federal agencies including the Centers for Disease Control (CDC) and the Health Resources and Services Administration (HRSA) are charged with providing essential support for non-­federal subdivisions of government. Drawing from community advocacy and with funding from nonprofit organizations such as the Robert Wood Johnson Foundation and the California Endowment, local governments became active in tobacco control and obesity prevention more than a decade ago. Some of these initiatives built on or coordinated with existing organizations

Paradigms of Healthcare Systems, Law, and Regulation   39 such as Federally Qualified Health Centers (FQHCs) and ‘social HMOs.’ Localities are now developing more comprehensive approaches to place-­based health improvement, such as community-­centred health homes and Accountable Health Communities as well as multipronged efforts to promote ‘health equity’ by addressing non-­medical needs (e.g., affordable housing) and other social determinants (e.g., systemic racism) through neighbourhood and urban partnerships. FQHCs have taken a leading role in many of these initiatives, including through medical-­legal partnerships that help address the ‘health-­harming legal needs’ of poor families such as housing, employment, and legal status. The post-­2008 opioid epidemic in the United States added urgency to these efforts, with local law enforcement rethinking its criminal justice orientation and instead playing a public health role in many cities, towns, and rural areas. These needs will only intensify given the COVID-­associated economic stress on the United States, especially as the pandemic highlights massive inequities among communities based on race and social class. The trend toward service integration has both a practical and an ethical dimension. As a practical matter, rising medical costs and mediocre performance on metrics for provider performance have drawn attention to social services as a largely unexplored care and cost management tool. But the jarring recognition of profound injustice in a healthcare system that aspires to technical perfection arguably has been an even greater inspiration for medical-­social innovation. Widening income and wealth disparities in US society generally, perpetuated by implicit biases and structural racism, are strongly reflected in health at the individual and community level. COVID-­19 has made these disparities harsher and more visible, as disproportionate amounts of serious illness and death have been visited on African American communities. Younger, more diverse generations of Americans see these circumstances as the result of prejudicial policies and find them morally unacceptable. In only a few years, the language of reducing ‘health disparities’ has shifted to promoting ‘health equity’ and then to achieving ‘health ­justice’—more clearly conveying both a sense of urgency and a call to action.

6  Regulation and Provider Governance The paradigms discussed so far in this chapter, through which the legal structures underpinning healthcare systems are understood, perpetuated, and challenged, are intertwined with the organization and governance of healthcare suppliers, be they publicly operated, independent health professionals, small-­scale not-­for-­profit organizations, or large corporate entities. In both Europe and the United States, these governance structures vary by healthcare service type and setting. Acute care hospitals differ from office-­based physician practices, which in turn differ from suppliers of technical inputs such as pharmaceuticals, medical devices, diagnostic or treatment equipment, and (recently) information systems. Moreover, as already noted, whether medical care is conceptualized as independent of other social services or as part of an integrated approach to fostering and supporting health is increasingly a consideration in both Europe and the United States.

40   Alceste Santuari and William Sage

6.1  Attitudes Toward Public Provision of Services 6.1.1 Europe From a European perspective, the principal choice in health system oversight is between ‘direct’ and ‘indirect’ regulation. Generally, regulation implies any set of legal provisions that are to be complied with in order to pursue certain public goals and interests. According to the different purposes to be served, regulation systems may imply both indirect and direct public action. Indirect regulation presumes that the positive outcomes for the economy (and, indirectly, for health) are expected to derive from the effective functioning of markets. Accordingly, economic regulation aims to make economic operators’ behaviours consistent with market rules and more efficient. By contrast, direct action comes in when an activity falls under government responsibility. To ensure citizens’ right to health means that both central and regional governments have to come to terms with two different dimensions: namely, the programming, pla­ nning, and organization function and the financial sustainability of their system. The programming or ‘planification’ function implies that the public organization of healthcare services is regarded as a necessary part of modern welfare states. Governments are responsible for both the quality of the healthcare services delivered and the organizational models that supply those services. In the past, in Europe, healthcare services used to be mainly supplied directly by publicly owned and controlled agencies, with directly publicly employed staff. However, there were always also privately operating health professionals, except in the case of the former Soviet states with their variants of the Semashko system. Since the late 1980s, across a range of European contexts, healthcare services have been increasingly delivered by private organizations, especially nonprofit entities. Both governments and private entities are now regarded as entrusted with ensuring citizens’ fundamental health rights. Considering the Italian case may be helpful. The Italian NHS and social security system are mixed systems in which public authorities and private organizations cooperate to increase accessibility to services through high-­quality standards. Private organizations, both for-­profit and nonprofit, along with public agencies, are engaged to deliver both health and social care services. When private organizations are called upon, they must comply with a stringent range of requirements aimed at ensuring patients’ access to quality services. Legal arrangements described in greater detail later, including public procurement and public-­private partnerships, are deployed to achieve these objectives.

6.1.2  United States The US healthcare delivery system has always relied on private service provision, traditionally consisting of independent physicians in small practice groups and nonprofit community hospitals with self-­governing ‘medical staffs’ constituted from those physicians. State laws long protected this artisanal, guild-­based organizational form from corporate interference, including by banning the direct employment of physicians by business enterprises, while both private and government health insurance honoured it

Paradigms of Healthcare Systems, Law, and Regulation   41 in benefit design and reimbursement practices. Doing so was justified on the assumption that aggregated professional activity would achieve collective public goals. For the vast majority of care, the same organizations have provided healthcare serv­ ices subject to the same legal requirements—regardless of whether the individual being treated is covered by public insurance (Medicare or Medicaid), covered by private insurance (employer-­sponsored or individually purchased), or paying out of pocket. In recent decades, healthcare has selectively industrialized by accommodation with professionally protective regulation, much of which has served to facilitate consolidation among conventional hospitals and health insurers, walling off these incumbent firms from potential new entrants and disruptive innovators. Yet, as the COVID-­19 pandemic has revealed, the clinical transactions that are the lifeblood of the existing system do not reliably respond to, much less prevent, threats to the public’s health. Emergency needs in response to COVID-­19 therefore have caused a temporary rethinking of some regulatory assumptions, but it is an open question whether the system failings revealed by the pandemic will result in a broader rebalancing of public and private care provision. The category of ‘direct regulation’ has a different valence in US health policy than in Europe. Public employment is substantial in the United States, but mainly occurs at subnational levels, tends to be lower in percentage terms than in Europe, and is less frequent in healthcare than in education or law enforcement. By contrast, public funding of private activity is central to US health policy; it is subject to legal and constitutional constraints that would not apply to private funding decisions. Indirect regulation—rules set by government that apply to private actors—is also widespread and important to healthcare services. Some indirect regulation is ‘economic’ in the sense that it is justified by the existence of market failure, but indirect regulation can also serve ‘social’ goals regarding welfare, fairness, personal freedom, or political voice—again subject to constitutional limits. Military medicine—on-­base services for active duty personnel and the Veteran’s Health Administration’s network of hospitals and clinics for veterans—is the most vibrant subsector of publicly delivered healthcare in the United States. Well-­developed leadership skills, staffing and team structures less constrained by antiquated state scope of practice laws, and the potential for telehealth and information technologies to operate independently of insurance claims generation and reimbursement are among the system’s strengths. Even so, the political appeal of making high-­end services seem more accessible to military families while subsidizing private provider interests tends to diminish the visibility and reputation of America’s impressive military health infrastructure. Another area for potential public sector expansion is community health. As previously described, the rapprochement between direct patient services and population-­ oriented health improvement is a major conceptual and operational change in orientation for US healthcare. The core public health workforce remains largely governmental, and it may turn out to be more effective and less expensive to leverage that infrastructure than to convert the private hospital sector into public health organizations. Pre-­COVID, for example, it was still unclear whether very sensible efforts by private

42   Alceste Santuari and William Sage hospitals and physician practices to more broadly deploy lay or para-­professional community health workers would be financially sustainable. With the pandemic raging, more sweeping questions are being asked about the need for a broader public workforce in health and healthcare, about the competence of established hospitals and other private provider organizations to serve collective interests and about the future fiscal capacity of the United States to finance the necessary infrastructure and human capital.

6.2  Publicly Funded but Privately Supplied Services 6.2.1 Europe The social market model that defines the European Union has traditionally implied that services of general interest can be provided by private organizations, whether not-­for-­ profit or for-­profit, though remaining publicly funded. While the services may be or­gan­ ized and delivered through a wide range of private organizations, their funding is a specific obligation of public authorities. In some legal systems, such as Italy, local health authorities also may set up private companies to supply services to communities. These companies are private in terms of management and organization but their financing derives mostly from public funds. Under EU law, public authorities may decide to contract out the production and supply of services of general interest, such as healthcare, or to set up their own controlled (in house) companies. Directives 2014/23 and 2014/24/EU allow public authorities of Member States to choose among these alternatives, although, according to the internal market rule, EU law tends to prefer competition procedures. Public authorities decide to set up their own in house companies when they want to maintain full control over the capital and the company’s governance, make sure that the company’s activities are wholly consistent with public authorities’ strategies and programmes, and exercise political control over the company’s activities. In-­house providing companies are usually set up at the local level and provide services such as elderly homes, work integration programmes for the disabled, pharmacies, home care services, and social housing. City councils and regional/local health authorities entrust in-­house providing agencies to carry out services of general interest without any competition procedure.23 EU law allows for such a legal exception exactly because these companies are publicly run and governed. These public agencies are funded with public resources or/and by customers’ co-­payments. While in-­house providing organizations account for stringent public control, they may cause an increase in local public expenditure, thus bringing 23   The in-house providing model is also part of the statutory implementation in both the United Kingdom Public Contracts Regulations and the Public Contracts (Scotland) Regulations. 2006. The details and precise application in the United Kingdom, however, are best illustrated by case law; namely, the rulings in Portsmouth and Risk Management (R. v. Portsmouth City Council Ex p. Coles and Ex. p. George Austin (Builders) Ltd [1997] 95 LGR 494 (Court of Appeal).

Paradigms of Healthcare Systems, Law, and Regulation   43 their effectiveness into question. Still, in-­house providing agencies assure public powers over essential services of general interest, such as water supply.24 Alternatively, when public authorities decide to contract out management and supply, they may do so by reserving public tenders for nonprofit organizations and social enterprises according to Article 77 of Directive 2014/24/EU. Tenders are selected by public authorities at the local level, especially when they are willing to achieve public goals through the active cooperation of civil society’s organizations.25 The non-­ distribution constraint and the social purposes of nonprofit organizations are recognized as essential features to ensure the reputation of these organizations, which are then entrusted with the supply of health and social care services. This is a way to support and enhance specific projects such, as the integration of people with disabilities into the labour market. Moreover, public authorities engage nonprofit organizations and social enterprises in supplying social and healthcare services through licensing.26 For certain purposes, nonprofit organizations are authorized to act on behalf of the national or regional health system, thus allowing citizens and patients at large to interact with those licensed organizations as if they were public. This accreditation system recognizes that economic value is not the most important aspect of the services, as it would be for typical public tenders. In fact, private organizations must meet other requirements, such as quality of services, patient involvement, nonprofit nature, and pursuit of the public interest.

6.2.2  United States Public support for private organizations is the norm in US healthcare because of paradigmatic concerns (described earlier) regarding consumer choice, unrestrained technology, and the purported benefits of (and undoubted profit opportunities from) marketplace competition. In addition to payment through public insurers such as Medicare and Medicaid, massive tax preferences attach to health insurance that is offered through private employers, giving substantial public support even to ostensibly private coverage arrangements. Essentially all these forms of health coverage fund private provider organizations, within which licensed physicians retain the most powerful voice. Through their ordering, referral, and prescribing decisions, US physicians drive roughly two-­thirds of national expenditures on medical care. Hospitals in the United States were for many years reluctant businesses while playing a public role that is more implied than explicit. They have functioned variously as physicians’ workshops, social institutions, and centres for training and research. But despite

  It worth mentioning that in 2011, in Italy, a referendum vote decided that water property and supply would fall under the exclusive power of public authorities. Accordingly, since then a significant number of city councils (see Naples) decided to set up in-house providing agencies to manage that service. 25   This is the case of migration centres in Italy, which are mostly managed by social enterprises that public authorities engage to carry out a wide range of health and social care activities for the benefit of migrants. 26   Court of Justice of the European Union, C-57/12 of 11 July 2013. 24

44   Alceste Santuari and William Sage their ostensibly private character and capacity for generating wealth and employment among their stakeholders, they have seldom acted as market-­driven healthcare enterprises. The post-­World War II Hospital Construction Act (Hill-­Burton) provided funding for capital investments in healthcare facilities across the United States while imposing only modest obligations regarding charity and access. Medicare’s original inpatient payment policies—including ‘cost-­ plus’ reimbursement for patient care expense and a capital cost supplement—enhanced hospital solvency and liquidity and attracted substantial new investment, as did favourable tax treatment of debt issued by nonprofit hospitals. Subsequent changes in government policy, such as the 1980s shift to Medicare prospective payment (DRGs) and the repeal of state laws that prohibited selective contracting between payers and hospitals, slowed growth in hospital beds in an equally predictable fashion (though cost growth largely continued). Other dynamics have been less easily anticipated. Concern that hospitals were failing to fulfill their public role with respect to charity care led to the enactment of the EMTALA law in 1986, ensuring that anyone with an emergency medical condition or in active labor would be screened and stabilized regardless of ability to pay. Fears of ‘managed care’ rationing and profiteering in the 1990s weakened competition law as applied to hospital mergers, ultimately drawing many large hospital systems and large insurers into symbiotic if not openly collusive relationships that raised prices and harmed consumers. As in Europe, moreover, local communities strongly oppose efforts to shutter inpatient capacity. Hospitals are not only valued as medical resources, but also tend to be among the largest employers and a reliable source of economic inflow from Washington, DC. Still, rural hospital closures in the United States have been common as those facilities have less potential to generate revenue from specialized care, struggle to attract physicians, and cannot afford to upgrade their information and management systems. COVID-­19 has exposed a major flaw in the unusual admixture of public and private spending applied to hospitals in the United States. Hospitals have an essential role in caring for those most severely sickened by coronavirus, notably elderly Medicare beneficiaries and individuals from crowded, poorer neighbourhoods that suffer from a variety of health disparities. Yet most of the funds that support these hospitals are associated not with that type of care or those categories of patients but with largely elective services delivered to patients covered by more lucrative (for the hospital) private health insurance. Many hospitals have found themselves in dire financial straits as a result. This illogical revenue imbalance raises an important question for proponents of an expanded federal role in health insurance—so-­called Medicare-­for-­All. On one hand, subjecting a much larger percentage of medical transactions to a uniform set of regulatory requirements, payment methods, and public expectations could improve both efficiency and fairness while rendering the enormity of medical spending sufficiently visible that more resources could be redirected to non-­medical social needs. On the other hand, centralizing revenues could further empower a narrow set of medically seasoned special interests with established power at the federal level and therefore could make it harder not easier to improve the efficiency and fairness of the healthcare system.

Paradigms of Healthcare Systems, Law, and Regulation   45

6.3  Self-­Regulating Professions 6.3.1 Europe National or local institutions in European countries take on the responsibility of providing citizens with services which are effective, of a reasonable quality, non-­discriminatory and accessible. Whatever the legal form that the provider takes on to carry out its activities, public and health authorities expect the provider to meet these criteria. Under EU law, though the organization of healthcare services falls under the single Member States’ jurisdictions, these services are characterized by at least two common features. First, healthcare services need to be supplied by regulated professionals regardless of the way the services are organized or managed. Second, healthcare services may not be subject to the EU competition rules as any other kind of service; on the contrary, healthcare providers are entrusted with some specific powers, rights, and duties exactly because they are to accomplish a public aim.27 Self-­regulation of healthcare professionals is common throughout Europe. Self-­ regulation is mainly used for access to some medical professions and for disciplining doctors who fall short of appropriate standards of care. Healthcare professionals usually group under national associations. These both set out the criteria according to which professionals are entitled to work and draft their own national contract schemes, which needs to be agreed upon by national or regional authorities. The peculiar characteristics of self-­regulated professionals, including doctors and pharmacists, lie behind their mutual recognition under EU law. Directive 2005/36/EC, as amended in 2013, enables free movement of professionals, such as doctors, among others, within the European Union. It provides for automatic recognition for those professions based on harmonized minimum training requirements (sectoral professions), a general system for the recognition of evidence of training, and automatic recognition of professional experience.28 All these requirements are aimed to ‘offer a high level of public health and patient safety’.29 The mutual recognition system represents an effective way by which Member States may increase the free movement of health professionals, thus increasing the spreading out of a common, shared set of values and best practices.

6.3.2  United States In many ways, the fundamental design principle of the US healthcare system seems to be not to have a system but instead to extrapolate an idealized relationship between a 27   This is the question considered by Newdick in stating that ‘the provision of public services is not amenable to market forces. As we have seen, the [UK] NHS attaches importance to equality and distributive ethics so that services are allocated fairly and consistently between people. These are not especially prominent values. in the philosophy of markets’. C. Newdick, From Hippocrates to Commodities: Three Models of NHS Governance, 22 (2) Med. L. Rev. (2014). 162–179 28   See Peeters M., Free Movement of Medical Doctors: The New Directive 2005/36/EC on Recognition of Professional Qualifications, 12 Eur. J. Health L. 373–396 (2005). 29   Directive 2013/55/EU.

46   Alceste Santuari and William Sage single patient and his or her physician to the population level and support it financially to the extent practicable. ‘Free choice’ of patient by physician, as well as of physician by patient, has been an iconic objective vigorously defended by the organized medical profession. Paradigmatically, this reflects supposed US preferences for patient choice and customized fiduciary service. But it also serves the self-­interest of US physicians, who (as Starr observed decades ago in his seminal history) mustered sufficient political power to fight off corporate and government control simultaneously. The COVID-­19 pandemic has surfaced many examples of selfless, heroic behaviour, but it has also demonstrated the perils to both public health and professional well-­being of celebrating lack of coordination in health system design. In the United States, whether regulation is nominally at the federal or at the state level, a substantial portion is highly deferential to the medical profession. Some authority is formally delegated to physicians; in other instances, government relies on professional processes to approve services for funding (e.g., the Joint Commission). Licensing boards controlled by physicians determine who can practice and in what ways. Hospitals’ eligibility for Medicare and Medicaid payment depend on compliance with standards set by physician-­centric accrediting bodies. Powerful physician advisory committees influence how much Medicare pays for professional services. Professional power is exercised by both state medical societies and national specialty organizations as well as by the American Medical Association and other umbrella bodies. Professional self-­regulation has mainly been charged with quality, which is natural and desirable insofar as health professionals have an advantage in expertise. But as there are few centralized assurances of access or constraints on cost, professional processes have been charged also with these functions. This often has been counterproductive, vulnerable as it becomes to economic protectionism, self-­dealing, overspecialization, and general medicalization even though most US physicians are well-­trained, ethical, and hard-­working individuals. Moreover, most industry stakeholders depend on physician orders and referrals for revenue and therefore are loath to oppose the medical profession in the political arena. As a result, many market distortions in US healthcare are ultimately traceable to the fact that American law broadly empowers the organized medical profession to act on behalf of individual patients, limits competition from other professionals and non-­ professionals, discourages industrial management and control, and lavishly subsidizes the resulting output. Physicians seldom subject their work to market accountability through transparent pricing and quality. At the same time, physician-­directed public spending has fewer explicit safeguards than structured government contracting (public tenders), making it easier for private interests to profit by piggybacking on professional control. US healthcare regulation has assumed that medical professionals self-­govern effectively and, by delegating public authority to professional licensing boards, has created barriers to entry by competitors, discouraged price discounting, and retarded innovation in ways that would violate competition law if engaged in by private commercial entities. At the same time, consumer protection authorities that would police advertising and marketing

Paradigms of Healthcare Systems, Law, and Regulation   47 practices for fraudulent or misleading content in most areas of commerce often defer to professional bodies where healthcare is concerned. After the US Supreme Court decided the North Carolina State Board of Dental Examiners case in 2015,30 there has been an upswing in market innovation such as telemedicine, which the COVID crisis has accelerated. But the long-­term regulatory settlement between physician direction and either market discipline or public need remains uncertain. At the same time, the United States imposes few public obligations on physicians— particularly regarding access—simply expecting them to be expert, honest, and altruistic. There are some exceptions, such as complicated fraud-­and-­abuse laws that have the purpose—if seldom the effect—of walling physician decision-­making off from commercial influence. Or the federal government’s attempt through EMTALA to obligate hospital emergency departments (and, to some extent, physicians working in or ‘on call’ to those departments) to screen and stabilize patients without regard to payment.

6.4  Public-­Private Partnerships 6.4.1  Europe Over the past few decades, the ways in which public authorities, including local health authorities, organize and supply services of general interest such as healthcare services have significantly changed. Along with the traditional government-­owned agencies and public tenders, this has introduced new organizational and legal models that better match citizens’ needs by bringing the public sector and the market couple together to pursue a common goal. Public–private partnerships (PPPs) represent a widespread phenomenon through which private companies and social entrepreneurs co-­operate with public authorities to define the most adequate legal and organizational tool whereby to accomplish a collective purpose. According to EU law, PPPs may be incorporated under any legal form available in the single Member States. However, PPPs are to be defined by some common characteristics: private parties, whether nonprofit or for-­profit organizations, must bear the highest portion of economic risks; private parties must be selected according to clear criteria; and the agreement between private parties and public authorities needs to be long-­lasting. PPPs in the health sector seem to be an effective legal tool whereby public authorities, which are usually constrained by cuts in their health budgets, may reach out for private investors and management. On their part, private organizations are called upon to share public interest and not only the possibility of making some profits out of the activity to be carried out. This is the reason why PPPs differ from traditional legal frameworks, such as tenders. While the former are focused on the willingness of private and public 30   135 S. Ct. 1101 (2015) (holding that licensing boards controlled by the professions they license are not ‘state actors’ immune from federal antitrust law).

48   Alceste Santuari and William Sage parties to enter a steady, long-­lasting, and accountable project, the latter are typically used to identify providers that are capable of offering particularized services at the lowest price. In the healthcare sector, PPPs can then attract both private funds and skills, which would be more difficult to obtain by means of other legal tools such as public tenders. While these mainly account for a market-­driven approach, PPPs—though also defined by some necessary market decisions—imply an agreed-­upon vision between public authorities and private businesses or nonprofit organizations which replaces the top-­down scheme in which public authorities have long acted. Add to this that, thanks to these characteristics, PPPs may be particularly suitable in times of economic and financial constraints during which public funds are less available. For example, public authorities implementing integrated care action plans may set up cooperation and partnership agreements with nonprofit organizations. The role of these private organizations may be two-­fold: as co-­producers of health and care services and as social promoters. The ultimate goal of supporting these actions by nonprofit organizations is by no means the engagement of cheap private service providers. On the contrary, the active involvement of the voluntary, community, and social enterprise sector is aimed at creating a local network of integrated care services. Nonprofit organizations are well positioned to become ‘agents’ of social innovation whose action is consistent with the programming background of a given local community. Such a responsible engagement of nonprofit organizations in the delivery of health and care services seems to respect both public responsibilities and contributions by those organizations of their own financial resources for the benefit of the community. A successful example of PPPs in the health sector is represented by IRST of Meldola, a small town near Forlì, in the Emilia-­Romagna Region.31 IRST was established in 2007 under the legal form of a limited liability company as a cancer centre of excellence for the whole region, one that, over the years, has been recognized as one of the national health facilities for research and care on cancer-­related diseases. Another example of a successful PPP in healthcare is the West German Proton Therapy Centre. It was established in 2006 as a ‘design, construction, financing’ PPP to deliver healthcare services, and it represents ‘one of the biggest German PPP transactions in public real estate and the first German PPP project directly financed by the capital market.’32 PPPs in the health sector also have witnessed some downsides, especially when private organizations do not wholly share the values and aims of the public authorities with which they partner. If for-­profit companies are driven by a profit-­making approach, their activities risk failure if either the service itself or the market conditions do not allow for sufficiently high return on investments. Likewise, a PPP in healthcare could be endangered by a nonprofit organization seeking excessively to increase its own reputation to the detriment of the social goal to be pursued by the partnership. A PPP could also be unsuccessful when public authorities choose to partner with a private organization  https://www.irst.emr.it/it.   H. W. Alfen & D. Daube, First PPP Steps in German Healthcare, 2(4) Eur. Pub. Priv. Partnership L. Rev. 206 (2007). 31

32

Paradigms of Healthcare Systems, Law, and Regulation   49 that is not capable of any social or technical innovation in those local situations in which innovation is seen as the main driver for setting up PPPs.

6.4.2  United States The PPP is rhetorically appealing in the United States but less often has been reduced to practice. As a functional matter, the expansion of private health plans offering insurance coverage to publicly supported beneficiaries is the dominant partnership model in US healthcare. This includes Medicare Advantage (MA) plans for the elderly, Medicaid managed care plans for states operating under federal waivers to shift away from traditional fee-­for-­service Medicaid, and heavily subsidized coverage made available through the ACA ‘marketplaces’ to low-­income Americans. These delegations have various goals, such as expanding Medicare benefits beyond those available under the traditional program, boosting access to physicians who refuse to participate in Medicaid directly because of low fees, and enabling a more aggressive approach to contracting and cost management than government can do directly. More generally, the non-­taxability of private health insurance made available as a fringe benefit of private employment reflects public consensus to ‘partner’ with private organizations in pursuit of public aims. Other relatively common forms of PPP seek to improve health and healthcare within communities. These are often supported by tax-­favoured philanthropic dollars channelled through private foundations or charities rather than directly by government, although federal support for FQHCs is an important exception. Governmental entities not primarily concerned with healthcare, such as the federal Department of Housing and Urban Development (HUD) and the Federal Reserve banking system, also may partner with private organizations on substantial initiatives to improve social determinants of health through infrastructure improvement or social impact investing.

6.5  Nonprofit and Charitable Organizations 6.5.1 Europe In Europe, nonprofit organizations have long played an essential role in delivering health and social care services. Until the end of the eighteenth century, nonprofit organizations were engaged in charitable activities such as social work, health and social care, alms housing, and education, especially for the benefit of the needy.33 Under both common and civil law systems, charities prospered and operated as key civil society

33   At the end of the thirteenth century, in Milan, there existed 10 hospitals, among which the S. Stephan Hospital could supply 500 beds and welcome 350 babies and 1,000 adults. See A. Cova, La situazione italiana: una storia di non profit, in Il Non Profit Dimezzato (G. Vittadini, ed., Milan 1997) 31–32. The case of the ‘Cà Granda’ Hospital is paradigmatic: it was a huge ‘enterprise’, which every day provided 1,600 men (barbers, chemists, bookkeepers, tailors), beside the inmates, with food. See E. Verga, Storia della vita milanese (Milan 1931) 174.

50   Alceste Santuari and William Sage organizations, especially at the local level.34 With the establishing and subsequent strengthening of the welfare state, public authorities started to provide most of the serv­ ices that nonprofit organizations had been delivering in the past. Hence, charitable and nonprofit organizations began to lose their role as health and social care providers in favour of public bodies/agencies. Accordingly, the downsizing of their service provider role turned nonprofit organizations to devote more attention and efforts to advocacy activities.35 During the 1970s, this organizational pattern changed when the increasing inability of traditional welfare policies to respond to an increasing demand for health and social care services gave birth to a legitimacy crisis of the European welfare regimes. To face this crisis, policy makers believed that governments would have to reduce the impact of public services on budgets by decentralizing the organization of social and healthcare polices to local authorities and by partly contracting out some health services to nonprofit organizations competing on quasi-­markets. This stimulation on the supply-­side contributed to boost new projects in a sector that for-­profit enterprises regarded as of little interest to them and opened up new opportunities for the nonprofit sector. The increasing engagement of nonprofit organizations in the delivery of health and social care services implied some significant changes in their organizational models. While increasingly carrying out economic activities, nonprofit organizations started to find new financial resources for their mission. The steady and entrepreneur-­oriented provision of health and social care services progressively shaped the traditional nonprofit form into a new organizational and legal framework: namely, the social enterprise. This is the case of those nonprofit organizations (associations or foundations) that were established to take care of a specific group of beneficiaries. The progressive changes both in the groups’ demand for health and social care and in the nonprofit organizations’ public funding have triggered an organizational remodelling. The shift, which has also legal implications,36 is usually concerned with the actual processes whereby the entities carry out their activities, which become business-­oriented. The organization no longer revolves around volunteers only and charitable donations: it starts to employ paid personnel and sign contracts with private companies and especially with public 34   As for the United Kingdom, in the twentieth century, ‘foundations emphasized five key principles in their work: doing what the state did not do; pump-priming—small grants to attract further funding; innovation; unpopular causes and risk taking; emergency funding.’ D. Leat, United Kingdom, Section I: The Foundation Sector in Europe, in Foundations in Europe (A. Schlueter, V. Then, & P. Walkenhorst, eds., Bertelsmann Foundation 2001) 270. 35   The rationale for a nonprofit advocacy role is that it advances the ‘public interest’, defined as the collective or the indivisible interests of the general public. See J. C. Jenkins, Non-Profit Organizations and Political Advocacy, in The Non Profit Sector. A Research Handbook, 2nd ed. (Yale University Press 2006) 307; and A. Hudson, Making the Connection: Legitimacy Claims, Legitimacy Chains and Northern NGOs’ International Advocacy, in After the ‘New Policy Agenda’? Non-Governmental Organisations and the Search for Development Alternatives (d. Lewis & T. Wallace, eds., Kumarian Press 2000). 36   These usually include a higher degree of liability for boards of directors and the obligation of keeping balance sheets like any other company.

Paradigms of Healthcare Systems, Law, and Regulation   51 authorities. This new perspective brings with it the need for better skilled and equipped management and a new approach to the ‘market’ of services. To give a concrete example, an Italian nonprofit foundation decides to employ some of its own patients, namely fragile women coming from North Africa. The foundation sets up a social enterprise that is 100% controlled and directed by the foundation itself. This enterprise manages a restaurant in which the employees are trained and paid to work as either cooks or waitresses. The social enterprise is largely funded by the clients who pay for their meals and partly by the scholarships that are granted by local businesses, which are willing to increase their own social responsibility. When compared to traditional nonprofit organizations, social enterprises display the following main distinguishing characters: (1) a productive, entrepreneurial, and market-­ oriented behaviour directed to mandating prioritization of social goods; (2) a higher propensity to innovate the supply of integrated care services as to the types of services provided, the target groups (often the more marginalized) benefited, and the organization of the services provision (high attention to active policies and to the empowerment of users); (3) a particular attention to the creation of new jobs, especially for hard-­to-­ place people (long-­term unemployed youth); (4) a greater attention to the local dimension of their activity, the strong link with a well-­defined community, and with its needs; (5) a higher degree of financial autonomy; and (6) a more relaxed nonprofit distribution constraint. The aforementioned aspects definitely make social enterprises more apt than traditional third-­sector organizations to act as entrepreneurs, thus making them less de­pend­ ent on public funds.37

6.5.2  United States Private charitable enterprises have long played a dominant role in US healthcare. Charitable organizations, most with religious affiliations, were early entrants into caring for the aged and infirm and expanded that role as new immigrant groups arrived in nineteenth-­century America and cities grew to accommodate them. Governments played a role in hospitals caring for the indigent and mentally ill, but public health functions in the United States were far less bureaucratized than in Europe, and those unable to afford private service from physicians were often more willing to trust charitable organizations. Even today, most US hospitals (and nearly all famous ones) are chartered as nonprofit corporations under state law and are organized and operated as public charities under federal tax law. The powerful American medical profession has favoured this organizational model because, until the past three decades, it provided free support staff, technology, and infrastructure to physicians who comprised the ‘voluntary 37   j. m. Coston, A Model and Typology of Government-NGO Relationships, 27(3) Nonprofit Voluntary Sect. Q. 358–382 (Sep. 1998). Before such a turning point, the voluntary welfare sector, ‘when matched against the welfare state, was consistently viewed as the “junior partner in the welfare firm” in terms of both overall size and scale of service delivery’. M. Chesterman, Foundations of Charity Law in the New Welfare State, in Foundations of Charity (C. Mitchell & S. R. Moody, eds., Portland OR: Hart Publishing, Oxford 2000) ch. 9, 251.

52   Alceste Santuari and William Sage medical staffs’ of community hospitals without challenging physicians’ clinical or financial independence. Although hospitals receive substantial tax benefits associated with that status, neither federal nor state law demands specific charitable services in return. Instead, tax-­exempt private hospitals must provide ‘community benefit,’ including performing a community health needs assessment (CHNA) and making required reports under section 501(r) of the Internal Revenue Code, which was added by the ACA. Under the federal EMTALA statute, all private hospitals receiving federal funding (whether nonprofit or commercial) must screen patients seeking care for emergency medical conditions and must stabilize such conditions without regard to payment. No formal commitment to a service area or population is required of hospitals, however, nor is the amount of services typically measured. As a result, essentially all nonprofit hospital systems in the United States today are organized and act as major business enterprises, with imposing physical facilities, highly compensated executives, and sophisticated business and billing functions that prioritize revenues, earnings, and competitive advantage no less than any of their for-­profit rivals. Other US healthcare enterprises have also embraced the nonprofit form—notably many ‘Blue Cross’ or ‘Blue Shield’ insurance plans, certain large HMOs such as Kaiser-­ Permanente, and a subset of skilled nursing facilities. With some exceptions, however, these established organizations do not stand out as particularly innovative. Still, FQHCs and similar community clinics have been among the most active organizations in attempting to improve access to care, prevent and manage chronic disease, and address the social determinants of health, and new nonprofit entities are constantly emerging in pursuit of similar goals.

7  Federalism and Healthcare Regulation Cutting across the paradigms of health law and the structures for health system regulation discussed earlier are questions of geographical scale. In both the United States and Europe, some healthcare regulation is applied at the national level while other aspects of regulation are devolved to smaller geographic units. American federalism constitutionally protects state sovereignty but makes national law supreme in specified areas. By contrast, the power to determine the modalities of a healthcare system remains at state level within the European Union and in greater Europe, although certain aspects, such as human rights protection, are subject to some supranational oversight through the Council of Europe. In both European countries and the United States, healthcare systems lodge many health regulatory responsibilities in localities or other public subdivisions.

Paradigms of Healthcare Systems, Law, and Regulation   53

7.1 Europe As to the organization and provision of healthcare services, the European Union consists of Member States and European institutions, which have different levels of responsibility. Whilst EU Member States are solely responsible for the organization of their own healthcare systems,38 the European Parliament and the European Commission act as coordinators of national policies and take action whenever there are cross-­border issues at stake.39 Despite these different assignments, both Member States and the European institutions share a common approach to healthcare services. These are or­gan­ized and delivered to accomplish a fundamental mission—universal and equitable access to welfare provisions that protect European citizens’ individual rights, including the right to health. National and regional public authorities in most European countries, in turn, divide or share responsibilities in pursuit of legal and organizational solutions that offer the best allocation of financial resources. There are various reasons why powers in the healthcare sector have been devolved from central to regional/local governments, with identifiable effects on healthcare provisions. Important questions, sharpened by the COVID-­19 pandemic, include whether we are facing an era of re-­centralization of powers and whether national and regional systems can be made to work together more effectively. Devolution has been a key reform issue mainly in tax-­funded countries, where the public sector performs the roles of financing, purchasing, and providing care. There are two main reasons for the devolution of powers, in general and in healthcare. The first reason has to do with policy: devolving powers from the central to local levels is expected to strengthen local democracy. This is supposed to improve the capacity of welfare systems to respond more effectively to citizens’ needs. Public policies and finances are applied to enhance government capacity and service accountability at the local level. The second reason is fiscal accountability, that is, the financial risk is placed on local governments, which are held responsible for public services. The combination of these two reasons in healthcare has forced national welfare systems to be divided into as many health regional systems as exist within the national territories. Such a division may cause significant territorial inequalities (a ‘patchwork quilt’) since the actual enforcement of healthcare rights heavily depends on the organization of the regional welfare systems. The rich areas would then tend to offer better and more effective healthcare services than the poor ones. For example, in Italy, due to the 38   Member States have developed their own independent powers in the organization, management, and supply of healthcare services because the Treaty of Rome did not entrust the European Union with any specific responsibility in the healthcare sector. The European Union was exclusively to be in charge of developing the internal market. 39   Progress in medicine, IT, and biotechnologies, along with the recognition of citizens’ freedom to access cross-border healthcare, have made healthcare also a transnational and international issue.

54   Alceste Santuari and William Sage great economic and social divide between the North and South, there is a significant health tourism of patients moving from Sicily, Campania, and Calabria to be treated in hospitals and clinics up north. Such a movement is largely funded by the Southern local health authorities, which causes a rise in health expenditure. Except for a few cases in which the decisions and policies of regional governments can be appealed before national Supreme Courts, regional and county authorities are free to implement the actions and policies they consider more effective and sustainable for their own areas.40 Such an approach can be harmonized through coordination meetings at the national level, in which regional and local governments share with central governments their own views, projects, and prospects. In their turn, central governments may propose to provide regional and local governments with funds not according to their expenditure history but according to standard costs incurred in efficiently providing healthcare services. The Nordic countries have long experienced healthcare devolution, even though, unlike in Italy (and Spain, among others), the functions and powers of local governments are not listed or protected in the Constitution. In Sweden, healthcare is financed mainly through county council tax revenues and is publicly provided by hospitals and health centres owned and managed by the public county councils. Swedish healthcare is organized on three levels: national, regional, and local. The regional level connects the county councils to the central government and forms the backbone of the system. Overall responsibility for the healthcare sector rests at the national level, with the Ministry of Health and Social Affairs. As in Italy, the responsibilities of municipalities in Sweden include social welfare and healthcare services consisting of child care, school health services, elderly homes, and care for the disabled and long-­term psychiatric patients. Sweden, like Italy, has witnessed shortcomings in the organization and delivery of healthcare at the local level. Although devolution helps match welfare provisions to local communities’ needs, it depends on effective political organizations. City councils’ decisions are determined by voters’ choices; accordingly, local public policies and priorities may not represent rational, prudent, or sustainable options. Politically driven city councils may oppose the decisions taken by health authorities, which are usually regional, independent, and technical organizations. This was the case in Italy when small city councils and their populations fought the closure of country hospitals. A decision that had been taken at the national level to rationalize healthcare expenditures and improve effectiveness of service delivery was contested by local authorities who relied on rhetoric and political pressure rather than in-­depth analysis. The foregoing example suggests that local and national health authorities should develop ways to channel their independent government powers into constructing joint programmes and policies. The Italian experience demonstrates that the more the regional government is engaged in issuing guidelines for city councils and health authorities, the more effective is the resulting action. Although at times there are some 40   In this perspective, it is noteworthy that some regions belonging to neighbouring countries can also program and plan cross-border healthcare services.

Paradigms of Healthcare Systems, Law, and Regulation   55 attempts to reverse the current devolved healthcare systems,41 it would be difficult, if not impossible, to turn back the clock on this trend.42 In Italy, changing the focal point of health policy will revolve around four main preconditions. The first is the need for a reliable and accountable organization that can activate a wide coalition of interests to overcome inertia and challenge the status quo. Second, this organization must develop and reinforce an alternative narrative to restore glamour and appeal to public health and social care. Third, the system should measure and report the benefits of a community-­based approach to public health. Finally, a new healthcare governance structure needs to be designed, including a new role for leading GPs as ‘territorial head physicians’ in charge of supporting, directing, and evaluating the network of GPs and healthcare facilities in a given area. At present, unfortunately, the government and the existing public system do not appear capable of bringing about a serious healthcare reform programme. Tensions between quality and accessibility may arise with respect to regionalization. Individuals in Italy appear especially satisfied with health and social care services and provisions that are near their homes. In some cases, such as in the Autonomous County of Trento and in some small towns in Emilia-­Romagna, citizens and local governments have taken action against the decisions of regional political leaders to close down country hospitals, even those regarded as underperforming and potentially dangerous for public health. It would seem that access to local healthcare is an important attribute of health system design as it is understood in European contexts and that at least some European citizens are willing to sacrifice quality for accessibility in this sense.

7.2  United States Healthcare in the United States is subject to two sets of public decision-­makers: the federal (national) government, and the governments of each of 50 sovereign states. The US Constitution (1) limits federal legislation to specifically ‘enumerated’ powers (e.g., interstate commerce, taxation) in the exercise of which federal law overrides conflicting state law; (2) regards only a few areas (which are rarely at issue in healthcare) as exclusively national, such as treaties with foreign governments; (3) restricts the federal government’s ability to dictate what state governments do or how they do it (though with ample opportunities for financial incentives); and (4) leaves the states in charge of protecting public health and safety (the traditional ‘police powers’). There is considerable overlap in this governance structure, which some constitutional scholars regard as essential to the protection of individual liberty by denying any 41   Some scholars write that ‘regional differentiation is by no means synonymous of waste and inefficiencies’. See M. Bertolissi, Tutela della salute: esigenze di eguaglianza e modelli organizzativi differenziati (con spunti di comparazione), in P.I Tronconi (ed.), Unione europea e diritto alla tutela della salute: problematiche giuridiche comparate, Santarcangelo di Romagna, 2016 82. 42   On December 4, 2016, the majority of Italian citizens voted no to a referendum that, among others, intended to recentralise some powers, including healthcare services.

56   Alceste Santuari and William Sage level of government unchallenged, plenary authority over people’s lives. In general, state governments regulate healthcare professionals and health facilities. The federal government regulates commerce in pharmaceuticals through the FDA but primarily applies its enormous fiscal capacity to financing the healthcare services that state laws permit to be delivered. Health insurance regulation is shared, though not in a particularly rational fashion. Federal dollars support Medicare coverage for the elderly and the lion’s share of Medicaid coverage for the poor. States are the dominant regulator of all types of private insurance, but federal law (the Employee Retirement Income Security Act [ERISA]) governs corporate pensions and fringe benefits and therefore the majority of private coverage that is linked to employment. In combination with Obamacare’s restrictions on medical underwriting, pricing rules, and benefit requirements, this creates a significant federal presence in private health insurance oversight. In the United States, importantly, neither the national government nor state governments view their mission as one that should support a right to health or even guarantee universal or equitable access to all but a few medical services. As discussed at the beginning of this chapter, American constitutionalism is built around freedom from government interference—so-­called negative liberty—rather than a public obligation to confer positive benefits. Specific statutes may reflect legislative intent to create such entitlements, but those are limited to the specific terms of the enacted laws, are politically malleable, and in some cases are subject to budgetary limits (i.e., annual congressional appropriations). That different American states offer disparate levels of support for healthcare services is, as the saying goes, a feature, not a bug. Because the United States pours so much public and private money into healthcare, individuals with health insurance (or those requiring emergency care, for which they will be billed) find nearly all services available throughout the nation, although much less infrastructure exists in rural areas. For those depending on financial assistance, however, there is enormous variability in access and quality from state to state (which tends to track race and ethnicity as well, as those characteristics correlate with educational and economic opportunity). The Medicaid program is mainly supported by federal funds and operates under national rules for certain classes of indigent beneficiaries, but it is administered by states and grants them considerable discretion regarding eligibility. Obamacare’s expansion of Medicaid was designed to promote uniformity among the states’ programs—granting coverage at predominantly federal expense to everyone earning less than 138% of the ‘federal poverty level’—but the US Supreme Court held that states could be offered but not compelled to accept those changes.43 Under the Trump administration, moreover, states were given even greater leeway through ‘Section 1115 waivers’ to modify Medicaid eligibility and benefits. The result is a version, or perhaps perversion, of the two goals cited for devolution in Europe. ‘Local democracy’ is being strengthened by enabling conservative states to act on their provincial

  National Fed. of Independent Business v. Sebelius, 567 US 519 (2012).

43

Paradigms of Healthcare Systems, Law, and Regulation   57 impulses regarding the distinction between ‘deserving’ and ‘undeserving’ poor, including imposing work requirements on Medicaid recipients.44 ‘Fiscal accountability’ is being served by allowing those same states to spend less of both their own money and federal matching funds on health insurance for the poor (although all states find creative ways to access federal financial resources). These policy tensions and reversals raise the question of whether centralized ‘health justice’ is more or less effective than localized ‘health justice.’ Since the battles over racial desegregation in the 1950s and 1960s, which Medicare greatly accelerated for hospitals, it has been assumed that federal policymaking would attempt to overcome state-­level prejudices. This is no longer self-­evident. There are alternative paths to harmonization in the United States. Umbrella organizations of state healthcare regulators, notably the National Association of Insurance Commissioners (NAIC) but also the National Governors Association (NGA), National Association of State Attorneys General (NAG), and the National Association of State Health Policy (NASHP), offer opportunities for interstate dialogue and the drafting of consensus documents. In the case of NAIC, its technical recommendations are routinely incorporated into state regulatory practice on a uniform basis. There is a National Conference of Commissioners on Uniform State Laws, a nonprofit organization that recommends consistent changes on a non-­partisan basis. Recent years have also seen renewed interest in formal Interstate Compacts, which are explicitly permitted under the US Constitution. For example, well-­established compacts harmonize state licensing practices for physicians and nurses.45 The need to alleviate state shortages of personnel and equipment, provide safe modalities for patient care, and facilitate public health surveillance has intensified interest in regional health regulatory cooperation during the COVID-­19 pandemic.

8 Conclusion The transatlantic conversation in this chapter shows that different regulatory approaches to the planning and provision of healthcare in Europe and the United States often arise from different priorities and may lead to different outcomes. It has also shown that important differences between European and American health systems exist in theory and how they operate in practice, particularly under the recent challenges presented by fiscal austerity, population growth, technological change, and a global coronavirus pandemic for which there is, as we write, still no effective cure or widely available vaccine. In our conclusion, we reflect on the directions of change. 44   Federal courts have invalidated some of these requirements, typically because of flaws in the Trump administration’s review process. See, e.g., Stewart v. Azar, 366 F. Supp. 3d 125 (D.D.C. 2019). More broadly, they arguably are inconsistent with Congress’s purposes in the Medicaid statute to provide coverage and improve health. Gresham v. Azar, No. 19-5094 (D.C. Cir. 2020). 45   See National Center for Interstate Compacts. http://www.csg.org/NCIC/about.aspx

58   Alceste Santuari and William Sage What lessons regarding the regulation of healthcare will be learned in Europe, and what in the United States? Will the balance of direct public service and reliance on private organizations change? If so, will it be driven by the need for public health preparedness, by financial considerations, by concerns over health equity, or by something else? What has COVID-­19 taught Western democracies about the relationship between liberty and collective obligation? About preventing medical scarcity or rationing life-­ saving resources? About caring for the elderly? About protecting healthcare workers and promoting their well-­being?

8.1 Europe Public regulation of healthcare in Europe does not appear to be in its dying days. Given the complexity of current social and healthcare systems across Europe, it is still very much needed, though in different forms than in the past. In the traditionally defined Beveridge health systems, health provisions and services now revolve less around direct delivery on the part of public bodies than around regulated contractual agreements between national and local health authorities and private organizations, especially nonprofits. In the Bismarck health systems, by contrast, public regulation aims to bring back to the central or regional health authorities a kind of coordination role that enables the overall system to control different health suppliers. European healthcare regulation therefore remains primarily concerned with designing and implementing legal and organizational frameworks within which public authorities, private organizations, and health professionals may all perform their duties in pursuit of one common goal: to ensure citizens’ right to health. If devolved approaches are the future of healthcare systems in Europe, does that necessarily mean that regional and local health systems will remain ‘national-­proof ’ in terms of retaining power, with all that this implies? Do we have to raise our hands in defeat in the face of progressive and apparently unstoppable social and health inequalities within the same national legal systems? Do we have to get used to fragmentation among services and territorial areas? Alternatively, is there any possibility for European healthcare systems to combine national guidelines with regional and local implementation of healthcare services? Central governments retain some general regulatory and monitoring power as long as either national constitutions or statutes provide for responsibility on the part of public authorities to ensure services of general interest, including healthcare. Generally, Secretaries or Ministers for Healthcare exert their powers by involving regional and local governments in the decision-­making process concerning issues such as migration, health risks, setting of quality standards, and the like. Regional and county governments are then free to implement health policies according to their economic, social, and demographic dimensions. Within a legal framework in which regional and local authorities are entrusted with organizing healthcare services, the regional level is expected to promulgate guidelines

Paradigms of Healthcare Systems, Law, and Regulation   59 that support local welfare actions and projects. It is at the local level that coordination plays its own strategic role since the different public responsibilities are challenged and compared locally. In this respect, public tenders and PPPs may play a significant role in the development of local communities.46 Not only can public procurement foster the growth of small and medium enterprises, but it also enables loosely coordinated government to better meet local needs. Regulation is all the more crucial in these endeavours because public and local health authorities need to be supported in their policies and choices. National settlements in favour of activities performance by nonprofit organizations also are supported by EU law, which regards them as better equipped and oriented than commercial enterprises to pursue social goals. In sum, the regional programming and planning function is supposed to provide health authorities and local municipalities with integrated, efficient, and effective responses, including health rationing and organizational rearrangements. As shown by the COVID-­19 pandemic, it is not an easy task to deliver on these promises since the multilevel decision-­making processes require a high level of coordination among the different actors and stakeholders who may be more accustomed to confronting each other than to working together for strong, sustainable, and universal healthcare.

8.2  United States Healthcare in America is simultaneously in need of regulation and liberalization (deregulation). Its principal regulatory needs are to define and pursue collective goals similar to those in Europe—broader access to necessary medical care, better preventive health services including healthier personal behaviour, and greater investment in non-­medical social services that improve health, economic productivity, and overall well-­being. Its principal deregulatory needs are to assess and, where necessary, undo decades of legal restrictions empowering and sheltering the medical profession that, in the aggregate, unnecessarily increase the cost of healthcare while diminishing its quality and accessibility. The COVID pandemic has revealed major failures of coordination in US healthcare that are related to its regulatory underpinnings—particularly its reliance on physician authority to direct expenditures. US healthcare delivery is simultaneously too fragmented and too consolidated, with lack of coordinated regulation at least partially responsible. COVID testing failed in the United States not only because the CDC and FDA were unable to devise and approve reliable assays, but also because a public health model of surveillance and control was incompatible with the commercial expectations of a lavishly funded system that generates revenues by processing individual claims for covered services. COVID supply chains failed because provider enterprises that vary 46   WHO EUROPE, Economic and Social Impacts and Benefits of Health Systems, Report (2019), https://www.euro.who.int/en/publications/abstracts/economic-and-social-impacts-and-benefitsof-health-systems-2019

60   Alceste Santuari and William Sage widely in wealth and influence were placed in competition with one another in a strained global marketplace. With the federal government largely abdicating its coordinating role, state governments became responsible by default for their internal situations and for regional alignment, with varying degrees of success. The political success of the Republican party’s sustained legal and rhetorical campaign against Obamacare shows that US health policy remains surprisingly resistant to central control. Public fears of ‘socialized medicine’—however antiquated the term sounds—retain salience given the extreme partisan divisions in the country today. The national government will always supply the vast majority of public funding for healthcare, and federal laws and administrative regulations will exert significant influence on healthcare delivery. But professional standard-­setting and the state laws that facilitate it is unlikely to be eclipsed any time soon. Legal changes necessary to facilitate care during COVID may be retained even after the crisis ends, which may exert a long-­term beneficial effect on access to care. Expansion of the public workforce charged with community health and preparedness may garner political support. It remains unlikely, however, that a truly universal, national system of care will emerge from the pandemic in the United States.

Bibliography Quality Versus Accessibility American Academy of Nurse Practitioners, State Practice Environment. https://www.aanp. org/advocacy/state/state-practice-environment (visited Jan. 7, 2021). Peter Buerhaus, Nurse Practitioners: A Solution to America’s Primary Care Crisis (Washington, DC: American Enterprise Institute 2018). https://www.aei.org/wp-content/ uploads/2018/09/Nurse-practitioners.pdf M. Campedelli, Riconfigurare l’universalismo?, cit., p. 2. Ezekiel  J.  Emanuel, Govind Persad, Ross Upshur, et al., Fair Allocation of Scarce Medical Resources in the Time of Covid-19, 382 N. Engl. J. Med. 2049–2055 (2020). Milton Friedman, Capitalism And Freedom (1962) 149–159. E.  Furno, Pareggio di bilancio e diritti sociali: la ridefinizione dei confini nella recente ­giurisprudenza costituzionale in tema di diritto all’istruzione dei disabili. 1 Nomos 2 (2017). Angie Drobnic Holan, PolitiFact’s Lie of the Year: Death Panels. (Dec. 18, 2009). https://www. politifact.com/article/2009/dec/18/politifact-lie-year-death-panels/ Institute of Medicine, Best Care at Lower Cost: The Path to Continuously Learning Healthcare in America (Washington, DC: National Academy Press 2012). Lawrence R. Jacobs, Politics of America’s Supply State: Health Reform and Technology, 14(2) Health Aff. 143–157 (1995). Thomas  A.  LaVeist & Lydia  A.  Isaac, edS., Race, Ethnicity and Health: A Public Health Reader (New York: Wiley, 2nd ed. 2012). Dayna  B.  Matthew, Just Medicine: A Cure for Racial Inequality in American Healthcare (New York: NYU Press, 2015). National Academies of Science, Engineering, and Medicine, Making Medicines Affordable: A National Imperative (Washington, DC: National Academy Press, 2018).

Paradigms of Healthcare Systems, Law, and Regulation   61 Office of Economic Policy et al., Dep’t of the Treasury, Occupational Licensing: A Framework for Policymakers. (Jul. 13–14, 2015). https://obamawhitehouse.archives.gov/sites/default/ files/docs/licensing_report_final_nonembargo.pdf Rachel Rebouche & Scott Burris, The Social Determinants of Health, in Oxford Handbook of US Health Law 1097 (I. Glenn Cohen et al., eds. 2017). William M. Sage, Assembled Products: The Key to More Effective Competition and Antitrust Oversight in Healthcare, 101 Cornell L. Rev. 609 (2016). William M. Sage, Minding Ps and Qs: The Political and Policy Questions Framing Healthcare Spending, 44(4) J. L. Med. Ethics 559–568 (2016). Linda Villarosa, ‘A Terrible Price’: The Deadly Racial Disparities of COVID-19 in America, NY Times Magazine. (Apr. 29, 2020). https://www.nytimes.com/2020/04/29/magazine/ racial-disparities-covid-19.html

Individual Liberty, ‘Health Citizenship,’ and the Right to Health l. m. H. Bongers & D. M. R. Townend, The Implementation of the Directive on the Application of Patients’ Rights in Cross–Border Healthcare in the Netherlands, 2 Eur. J. Health L. 65 (2014). See R.  Cavallo Perin, Il welfare state nell’Unione Europea in tempo di crisi economica e l’inesatta contrapposizione tra Stato e mercato, 1(34) Fenomenologia e Società 51 (2013). a. Den Exter, A.  Santuari, & T.  Sokol, One Year After the EU Patient Mobility Directive: A Three-Country Analysis, 40(2) Eur. L. Rev. 281 (April 2015). European Commission, Evaluative Study on the Crossborder Healthcare Directive (2011/24/EU). Final report 21. (March 2015). S.  Giubboni, Cittadinanza europea, libertà di circolazione e solidarietà transnazionale: un riesame critico. 1 Politiche Sociali/Soc Pol. 169–184 (2016). Corte dei Conti, Sezioni Unite in sede di controllo, Rapporto 2016 sul coordinamento della finanza pubblica, 15 marzo 2016 (Del. N. 2/SSRRCO/RCFP/16), Presentazione, pp. 9–12 M. Karanikolos, P. H. Mladovsky, J. Cylus, S. Thomson, S. Basu, D. Stuckler, J. Mackenbach, & M. McKee, Financial Crisis, Austerity and Health in Europe, 381(9874) Lancet 1323–1331 (2013). M. Kattelus, Implementation of the Directive on the Application on Patient’s Rights in Crossborder Healthcare (2011/24/EU) in Finland, 2 Eur. J. Health L. 23 (2014). Paolo Lazzara, La funzione regolatoria: contenuto, natura e regime giuridico, in L’intervento pubblico nell’economia (M.  Cafagno, F.  Manganaro, eds. Firenze University Press, 2016), 117 ff. A.  Maresso et al. (eds.), Economic Crisis, Health Systems and Health in Europe: Country Experiences, Copenhagen (European Observatory on Health Systems and Policies, 2015). C. Newdick, From Hippocrates to Commodities: Three Models of NHS Governance, 22(2) Med. L. Rev. 161–179 (Spring 2014). S. Olsena, Implementation of the Patients’ Rights in Cross–Border Healthcare Directive in Latvia, 2 Eur. J. Health L. 46 (2014). l. Prudi, Implementation of the Directive 2011/24/EU in the Czech Republic, 2 eur. j. health l. 15 (2014). m. a. Requejo, Cross–Border Healthcare in Spain and the Implementation of the Directive 2011/24/EU on the Application of Patient’s Rights in Cross–Border Healthcare, 2 eur. j. health l. 79 (2014). Sara Rosenbaum, The Enduring Role of the Emergency Medical Treatment and Active Labor Act, 32(12) health aff (millwood) 2075–2081 (2013).

62   Alceste Santuari and William Sage m. schwebag, Implementation of the Cross–Border Care Directive in EU Member States: Luxembourg, 2 eur. j. health l. 56 (2014). B. Vanherke, S. Sabato, & D. Bouget (eds.), Conclusions. Social policy in the EU: High Hopes but Low Yields, Social Policy in the European Union: State of Play 2017, European Social Observatory (OSE), p. 201 t. Vidalis, & I. Kyriakaki, Cross–border Healthcare: Directive2011/24 and the Greek Law, 2 eur. j. health l. 33 (2014).

Market Competition and Economic Productivity Jennifer Cheeseman Day & Jeffrey Rosenthal, Detailed Occupations and Median Earnings: 2008. US Census Bureau, Working Paper. (2008). https://www.census.gov/content/dam/ Census/library/working-papers/2008/demo/acs08-detailedoccupations.pdf Atul Gawande, The Cost Conundrum, new yorker. (June 1, 2009). https://www.newyorker. com/magazine/2009/06/01/the-cost-conundrum Sherry Glied & Adam Sacarny, Is the US Healthcare System Wasteful and Inefficient? A Review of the Evidence. 43(5) j. health polit. pol. l. 739–765 (2018). chris ham, reforming the nhs from within beyond hierarchy, inspection and markets (Kings Fund 2016). Anne Kim, Healthcare’s Biggest Problem Is Getting Worse, wash. monthly (Oct. 12, 2019). David Orentlicher, Healthcare, Health, and Income, 46 j. l. med. ethics 567–572 (2018). Heather Perlberg, How Private Equity Is Ruining Healthcare, bloomberg news. (May 20, 2020). https://www.bloomberg.com/news/features/2020-05-20/private-equity-is-ruininghealth-care-covid-is-making-it-worse RAND Corporation, The Evolving Role of Retail Clinics. (2016). https://www.rand.org/ content/dam/rand/pubs/research_briefs/RB9400/RB9491-2/RAND_RB9491-2.pdf) Sara Rosenbaum & Elizabeth Taylor, The Irreplaceable Program in an Era of Uncertainty, 46(4) j. l. med. ethics 883–886 (2018).

‘Health Justice’ and Integrated Social Services Dawn E. Alley, Chisara N. Asomugha, Patrick H. Conway, & Darshak M. Sanghavi, Accountable Health Communities: Addressing Social Needs Through Medicare and Medicaid, 374 n. engl. j. med. 8–11 (2016). Centers for Medicare and Medicaid Services, CMS Roadmap: Strategy to Fight the Opioid Crisis (June 20207). CMS Opioid Roadmap: Strategy to Fight the Opioid crisis.m. Cinelli, L’ effettività delle tutele sociali tra utopia e prassi, 16(1) rivista del diritto della sicurezza sociale 21 (2016). richard cooper, Poverty and the Myths of Healthcare Reform (Baltimore: Johns Hopkins University Press 2016). G. Fiorentini, Dimensioni orizzontali e verticali dell’integrazione socio-sanitaria in I servizi sanitari in Italia. 2004 (G. Fiorentini, ed., il Mulino 2004) 226–227 C. Ham & H. Alderwick, Place-based Systems of Care. A Way Forward for the NHS in England (King’s Fund 2015). National Academies of Sciences, Engineering, and Medicine, Integrating Social Care Into the Delivery of Healthcare (Washington, DC: National Academies Press 2019).

Paradigms of Healthcare Systems, Law, and Regulation   63 National Ass’n of Community Health Centers, Highlighting the Role of Enabling Services at Community Health Centers. (2010). http://www.nachc.org/wp-content/uploads/2015/06/ EnablingServicesReport.pdf r. Pessi, Tornando su adeguatezza e solidarietà nel welfare, 4 Rivista del Diritto della Sicurezza Sociale 594 (Dec. 2016). l. Rampa, Paternalismo, autonomia e diritti sociali: una rilettura in termini di analisi economica, 47 Politica del Diritto 305–336 (2016). Joel Teitelbaum & Ellen Lawton, The Roots and Branches of the Medical-Legal Partnership Approach to Health: From Collegiality to Civil Rights to Health Equity, 17(2) Yale J. Health Polc’y. L. Ethics 343–377 (2017). Elizabeth Tobin-Tyler & Joel B. Teitelbaum, Essentials of Health Justice: A Primer (Burlington, MA: Jones and Bartlett Learning 2019). Louise  G.  Trubek, The Social HMO for Low-Income Families: Consumer Protection and Community Participation, 26 Seton Hall L. Rev. 1143–1163 (1995–1996). Bruce C. Vladeck, The Political Economy of Medicare, 18(1) Health Aff. 22–36 (1999).

Regulation and Provider Governance Attitudes Toward Public Provision of Services C.  Barbero, S.  Gilchrist, J.  F.  Chriqui, M.  Martin, A.  Wennerstrom, J.  VanderVeur, K. Prewitt, J. N. & Brownstein, Do State Community Health Worker Laws Align with Best Available Evidence?, 41(2) J. Comm. Health 315–325 (2016). William M. Sage & Jennifer E. Laurin, If You Would Not Criminalize Poverty, Do Not Medicalize It, 46(3) J. L. Med. Ethics 573–581 (2018).

Publicly Funded but Privately Supplied Services Andrea Park Chung, Martin Gaynor, & Seth Richards-Shubik, Subsidies and Structure: The Lasting Impact of the Hill-Burton Program on the Hospital Industry, 99(5) Rev. Econ. Stat. 926–943 (2017). Mario Comba & Steen Treumer (eds), The In-House Providing in European Law (Copenhagen: DJØF Publishing 2010). R.  Craig, Public Works Procurement: The Portsmouth Case, 15 Construction  L.  J. 88–110 (1999) at 107. Martin Gaynor & Robert Town, Robert Wood Johnson Found., The Impact Of Hospital Consolidation—Update 2 (June 1, 2012). The Impact of Hospital Consolidation - RWJF C. Iaione, The In-House Providing in European Law: When Nothing Gets Lost in Translation, 1 Ital. J. Pub. L. 192 (2010). G.  Brystana Kaufman, Sharita  R.  Thomas, Randy  K.  Randolph, Julie  R.  Perry, Kristie  W.  Thompson, George  M.  Holmes, & George  H.  Pink, The Rising Rate of Rural Hospital Closures, 32(1) J. Rural Health 35–43 (2016). William  M.  Sage, Adding Principle to Pragmatism: The Transformative Potential of ‘Medicare-for-All.’ Working Paper. Feb. 2020. https://papers.ssrn.com/sol3/papers. cfm?abstract_id=3387120 William B. Vogt & Robert Town, Robert Wood Johnson Found., How Has Hospital Consolidation Affected The Price And Quality Of Hospital Care? 11–12 (feb. 2006).untitled (iupui.edu)

64   Alceste Santuari and William Sage Self-­Regulating Professions Erin P. Fraher, Patricia Pittman, Bianca K. Frogner, Joanne Spetz, Jean Moore, Angela J. Beck, David Armstrong, & Peter I. Buerhaus, Ensuring and Sustaining a Pandemic Workforce, 382 N. Engl. J. Med. 2181–2183 (2020). Joan  K.  Krause, Integration, Fragmentation, and Human Nature: The Role of the Fraud and Abuse Laws in a Changing Healthcare System, in Oxford Handbook of US Healthcare Law (I. Glenn Cohen et al., eds. 2017) 852. Miriam  J.  Laugesen, Fixing Medical Prices: How Physicians Are Paid (Cambridge, MA: Harvard University Press 2016). James S. Roberts et al., A History of the Joint Commission on Accreditation of Hospitals, 258 JAMA 936 (1987). Paul Starr, The Social Transformation of American Medicine (New York: Basic Books 1982). Charles D. Weller, Free Choice as a Restraint of Trade in American Healthcare Delivery and Insurance, 69(5) iowa L. Rev. 1351–1392 (1984).

Public–Private Partnerships Allison K. Hoffman, What Health Reform Reveals About Health Law, in Oxford Handbook of US Healthcare Law (I. Glenn Cohen et al., eds. 2017) 50–69. Kaiser Family Foundation, State Medicaid Waiver Tracker. https://www.kff.org/medicaid/ issue-brief/medicaid-waiver-tracker-approved-and-pending-section-1115-waivers-bystate/ (visited Jan. 5, 2021). Theodore R. Marmor & Jonathan Oberlander, Medicare at Fifty, in Oxford Handbook of US Healthcare Law (I. Glenn Cohen et al., eds. 2017) 742–764. National Conference of State Legislators, Social Impact Bonds. https://www.ncsl.org/research/ labor-and-employment/social-impact-bonds.aspx (visited Jan. 7, 2021). Sara Rosenbaum, Medicaid at Fifty, in Oxford Handbook of US Healthcare Law (I. Glenn Cohen et al., eds. 2017) 765–786.

Nonprofit and Charitable Organizations a. Bacchiega & C. Borzaga, Social Enterprises as Incentive Structures: An Economic Analysis, in The Emergence of Social Enterprise (c. Borzaga, J. Defourny eds.) 2001, 273–295. t. Bovaird, Efficiency in Third Sector Partnerships for Delivering Local Government Services: The Role of Economies of Scale, Scope and Learning, 16(8) Pub. Mgmt. Rev. 1067–1090 (2014). D. Brakman Reiser & S. A. Dean, Social Enterprise Law. Trust, Public Benefit and Capital Markets (Oxford University Press 2018) 31. F.  Calò, S.  Teasdale, C.  Donaldson, M.  J.  Roy, & S.  Baglioni, Collaborator or Competitor: Assessing the Evidence Supporting the Role of Social Enterprises in Health and Social Care, 20(12) Pub. Mgmt. Rev. 1790–1814 (2018). g. Dees, Enterprising Non-Profits: What Do You Do When Traditional Sources of Funding Fall Short?, 76 Harv. Bus. Rev. 54–67 (Jan-Feb. 1998). a. Den Exter, & M. Guy, Market Competition in Healthcare Markets in the Netherlands: Some Lessons for England?, 22(2) Med. L. Rev. 255–273 (2014). European Center for Not-for-Profit Law, Legal Regulation of Economic Activities of Civil Society Organizations, Policy paper, Budapest. (February 2015).

Paradigms of Healthcare Systems, Law, and Regulation   65 A. Fici (ed.), Diritto dell’Economia Sociale, part III, Naples, 2016. h. b. Hansmann, The Role of Nonprofit Enterprise, 89 Yale L. J. 843 (n 32) (1980). Jill R. Horwitz, Nonprofit Healthcare Organizations and the Law, in Oxford Handbook of US Healthcare Law (I. Glenn Cohen et al., eds. 2017) 535–555. D. Jiménez & P. Smith, Decentralisation of Healthcare and Its Impact on Health Outcomes (York: Centre for Health Economics, University of York 2005). j. Le Grand, Quasi-Markets and Social Policy, 101(408) Econ. J. 1256–1267 (Sep. 1991). r. Levaggi & P. Smith, Decentralization in Healthcare: Lessons from Public Economics, Paper Prepared for Conference on Economics and Health Policy Centre for Health Economics, University of York (Dec. 16, 2003). h. Maarse, The Privatization of Healthcare in Europe: An Eight-Country Analysis, 31(5) J. Health Polit. Pol’cy L. 981–1014 (Oct. 31, 2006). g. Metallo, M. T. Cuomo, D. Tortora, & M. Galvin, Innovation Networks and Social Enterprises. A Social Network Analysis of the Third Sector in Italy, 3 Rivista Piccola Impresa/Small Business 47 (2016). B. Palier, A Long Good Bye to Bismarck? The Politics of Welfare Reforms in Continental Europe. Paper presented at the RC19 Conference on Social Policy in a Globalizing World: Developing a North-South Dialogue, Florence University (Sep. 6–8, 2007) 9. m. Powell & R. Miller, Privatizing the English National Health Service: An Irregular Verb? 38(5) J. Health Polit. Pol’cy L. 1051–1059 (Oct. 2013). m. Syinge, Public Benefit Post-Pemsel, in Research Handbook on Not-For-Profit Law (m. Harding, ed, Elgar 2019) 363–386. R. Ter Meulen, W. Arts, & R. Muffels (eds.), Solidarity in Health and Social Care in Europe (Springer, New York, 2001). A. Withey, Medicine and Charity in Eighteenth-Century Northumberland: The Early Years of the Bamburgh Castle Dispensary and Surgery, c. 1772–1802, 29(3) Soc. Hist. Med. 467–489 (Aug. 2016).

Federalism and Healthcare Regulation G. Fiorentini (ed.), I servizi sanitari in Italia (Bologna, Il Mulno, 2012) 10–11. C.  Hjortsberg & O.  Ghatnekar, Healthcare Systems in Transition. Sweden, European Observatory on Healthcare Systems, 2001, p. 17 Timothy Stoltzfus Jost, Access to Health Insurance and Health Benefits, in Oxford Handbook of US Healthcare Law (I.  Glenn Cohen et al., eds. 2017) 147–166 Srinivasa Vittal Katikireddi, Katherine  E.  Smith, David Stuckler, & Martin McKee, Devolution of Power, Revolution in Public Health? 39(2) J. Pub. Health 241–247 (June 1, 2017). Dayna B. Matthew, Next Steps in Health Reform: Hospitals, Medicaid Expansion, and Racial Equity, 46(4) J. L. Med. Ethics 906–912 (2018).

Conclusion C. Ham & H. Alderwick, Place-Based Systems of Care. A Way Forward for the NHS in England (King’s Fund November 2015).

Pa rt I I

PR E V E N T I NG I L L H E A LT H

chapter 3

I n troduction to Pu blic Hea lth L aw Wendy E. Parmet, Markus Frischhut, Amandine Garde, and Brigit Toebes

Public health laws are among the most ancient health laws. Long before states had complex healthcare financing systems or regulated pharmaceuticals and medical technologies, they enacted laws designed to stem fearsome epidemics and protect their populations from the more common diseases that plagued the pre-­modern era.1 Over the past 150 years, as medical knowledge has increased and healthcare systems have grown in terms of cost and complexity, public health law has often been overshadowed by other areas of health law. That may now be changing. In the face of emerging infectious diseases, such as SARS, Ebola, and COVID-­19, and a growing appreciation of the critical role that social, environmental, and economic forces play in shaping public health, policymakers and scholars have begun to pay increasing attention to the theoretical foundations of, ethical justifications for, and scope of public health laws. In the three paired chapters that follow, we focus on three critical areas of public health law: communicable disease control, the control of noncommunicable diseases (NCDs), and efforts to address the social determinants of health. In each set of paired chapters, we follow a common structure and consider the US perspective on the one side and the European perspective on the other. The analyses that emerge from these chapters, which were written before COVID-19 overtook much of the globe, reveal that legal systems on either side of the Atlantic face broadly similar questions. These include: What is the appropriate scope of public health law? How should responsibility be divided between centralized and more localized authorities? What circumstances justify the limitation of individual liberty to protect public health? How should ­governments regulate in the face of scientific uncertainty? And, what protections should be in place to prevent public health laws from targeting vulnerable individuals and populations? 1  An “epidemic” refers to a contagious, infectious, or viral illness that spreads to many people in a specific region, whereas a “pandemic” surpasses this region.

70   Wendy E. Parmet et al. Yet while the questions raised on either side of the Atlantic are common, the answers given often differ. In part, this is because the European Union is for the most part a supranational organization, while the United States is a more integrated, albeit federal, nation state. In addition, important distinctions between the legal traditions and the value given to individual liberty versus the public good provide a focus of our comparison between the US and the European approaches to public health. The following discussion offers a brief introduction to the key issues and themes presented in the paired chapters. We begin by shedding some light on the broad definition of public health law. We also explicate what we are referring to when we discuss “Europe” in the chapters that follow. We then provide a short preview of some of the key themes in each paired set of chapters.

1  What Is Public Health Law? There is no authoritative definition of public health law. In a recent report for the World Health Organization (WHO), Roger Magnusson explains that “the concept of public health, as understood in this report, is not restricted to laws that regulate the provision of health care services, but includes the legal powers that are necessary for the State to discharge its obligation to realize the right to health for all members of the population.”2 Taking a somewhat different tack, Richard Goodman and colleagues from the US Centers for Disease Control and Prevention (CDC) define public health laws as including “any laws that have important consequences for the health of a defined population.”3 Both of these definitions accept that public health laws are laws that affect health and are not limited to those that regulate the care, financing, or organization of clinical health services. Instead, public health laws encompass a wide array of laws that impact the health of populations, including traditional infectious disease control laws, such as quarantine or vaccination laws; environmental laws; and laws that regulate food safety, to give just a few examples. The focus on impact also holds true for Micah Berman, who writes, “[p]ublic health law is the study of the legal powers and duties of the state to identify, prevent, and ameliorate risks to the health of populations, as well as the study of legal structures that have a significant impact on the health of populations.”4 In their influential treatise, US scholars Lawrence O. Gostin and Lindsay F. Wiley accept the idea that public health laws relate to the health of populations and not just the provision of clinical care. They also observe that public health law recognizes other

2  Roger Magnusson, Advancing the Right to Health: The Vital Role of Law, 27 (2017), https://www. who.int/healthsystems/topics/health-­law/chapter2.pdf?ua=1. 3  Richard Goodman et al., “Law and Public Health at CDC,” 55 Morb Mort Wk Rep Sup. 02 23–33 (Dec. 22, 2006), https://www.cdc.gov/mmwr/preview/mmwrhtml/su5502a11.htm. 4  Micah L Berman, “Defining the Field of Public Health Law” (2013) 15(2) DePaul J Health Care L 45 80.

Introduction to Public Health Law   71 important interests, including individual liberty and social justice.5 Likewise, in Europe, Tamara Hervey and Jean McHale adopt a broad definition in which public health is seen as “an almost impossibly wide concept,” which encompasses environmental law (e.g., air and water quality), welfare law (e.g., social security, social care, and education), consumer protection, and many more.6 This breadth is reflected in the “Health in All Policies” recommended by the US CDC or “mainstreaming” obligation requiring the EU to ensure a high level of public health protection in the development and implementation of all its policies.7 In the three chapters that follow, we embrace these broad and implicitly normative definitions. The discussions recognize that public health law, in contrast to healthcare law, seeks to protect health at a broad population, as opposed to an individual patient, level. They also acknowledge that the field of public health emphasizes prevention and health promotion, as opposed to the treatment of disease. However, as the definition by Goodman and colleagues implies with its reference to laws that have important consequences for health, the impact of public health laws is not always positive: they may fail to have the impact that lawmakers seek, or they may be misguided, even deleterious, to population health. In addition, as Gostin and Wiley suggest, we accept that the field of public health law, to which we hope these chapters are a contribution, is broader than a study of public health laws themselves. Instead, the field and the discussions that follow, seek to understand not only public health laws, but also the larger questions of how states can assure the conditions by which people can be healthy while respecting other critical human rights and values within their constitutional structures and adhering to the rule of law. As we shall see, both the United States and Europe grapple with these tensions, even as they often adopt different answers. Before turning to the overview of the three chapters that follow, it may be worth clarifying—particularly for readers who are not familiar with European governance— what we mean by “Europe.” We have, so far, used the term loosely to distinguish Europe from the United States, which together constitute the focus of this book. However, the term “Europe” is somewhat misleading in that, depending on how it is used, it can refer to different realities. For our purposes, three different organizations should be noted. First, the WHO regional office for Europe (WHO EURO) operates in 53 countries, covering a population of around 900 million people. As one of six WHO regional offices around the world, WHO EURO has published a series of regional action plans, technical guidance, and recommendations, and it supports the development, implementation, monitoring, and evaluation of effective public health policies, including public health 5  Lawrence O Gostin and Lindsay A Wiley, Public Health Law: Power, Duty, Restraint 4 (3d ed. 2016). 6  Hervey TK and McHale JV, European Union Health Law: Themes and Implications (Cambridge University Press 2015) 27. 7  Articles 9 and 168(1) of the Treaty on the Functioning of the EU and Article 35 of the Charter of Fundamental Rights of the EU ([2016] OJ C202); Office of the Associate Director for Policy and Strategy, Centers for Disease Control and Prevention, Health in All Policies, https://www.cdc.gov/ policy/hiap/index.html (9 June 2016).

72   Wendy E. Parmet et al. laws, in its 53 member countries (which include Russia and the other former members of the USSR as well as Israel and Turkey). Second, the Council of Europe (CoE) is the leading human rights organization on the continent. It sits in Strasbourg and comprises 47 Member States that have all signed up to the European Convention on Human Rights (ECHR)8 whose implementation in CoE Member States is overseen by the European Court of Human Rights (ECtHR). Several other CoE instruments are relevant to the development of public health laws and policies in the Member States, not least the European Social Charter.9 Third, the EU now comprises 27 Member States,10 covering a total population of around 450 million people. EU law, characterized by deeper economic and political integration, is interpreted by the Court of Justice of the EU (CJEU) that sits in Luxembourg. All EU Member States are CoE members; they are therefore bound by the ECHR as well as the Charter of Fundamental Rights of the EU (CFREU)11.

2  Communicable and Other Infectious Diseases A “communicable disease” is an infectious disease that “is transmitted from person to person by direct contact with an infected individual or by indirect means such as exposure to a vector, animal, fomite, product or environment, or exchange of fluid, which is contaminated with the contagious agent.”12 As the definition suggests, not all infectious diseases are communicable; for example, infection from toxins of the tetanus bacterium is not. Public health measures can be critical in protecting against both communicable and noncommunicable infectious diseases. Long before science understood the role that microorganisms play in causing illness, states enacted a variety of laws that were designed to protect a population from outbreaks of infectious disease. As Gostin, Burris, and Lazzarini note, contemporary infectious disease laws bear the influence of changing understandings of the etiology of infection as well as innovations in medicine and pharmacology.13 In both Europe and the United States, ancient laws relating to isolation and quarantine co-­exist with more modern laws that relate to surveillance or the regulation of new pharmaceutical innovations.

8  Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights, as amended) (signed 4 November 1950, entered into force 3 September 1953) CETS 5. 9  European Social Charter (Revised) (adopted 3 May 1996, entered into force 1 July 1999) CETS 163. 10  [2020] OJ L29 and C 34. 11  [2016] OJ C202/389. 12  European Parliament and Council Decision 1082/2013/EU of 22 October 2013 on Serious Cross-­Border Threats to Health and Repealing Decision 2119/98/EC [2013] OJ L293/1, as corrected by [2015] OJ L231/16, art 3 lit b. 13  Lawrence O Gostin, Scott Burris, and Zita Lazzarini, “The Law and the Public's Health: A Study of Infectious Disease Law in the United States” (1999) 99(1) Colum L Rev 59.

Introduction to Public Health Law   73 Both the United States and Europe confront some similar challenges. A key one is the distribution of authority (the term used in the United States) or competency (the term employed in the European Union) over infectious disease control. Should it be more centralized or remain rather decentralized? Another important question relates to how officials should respond to scientific uncertainty, which often exists at the outset of a new epidemic. A third, overarching question, is the relationship between individual autonomy and disease control. Historically, officials have often responded to epidemics by limiting the rights of individuals to travel or even to determine their own medical care. Reasonableness of public action, or the “principle of proportionality,” which has been shaped by the CJEU, is the key to keeping the right balance in this regard. Traditionally, human rights are classified as positive (right to benefits) or negative rights (a right to non-­interference).14 However, against the background of the broad notion of “public health,” limiting rights (in the sense of negative rights) might often not be enough, especially to prevent infectious diseases. Moreover, vulnerable populations have often borne the brunt of infectious disease control measures, such as quarantine. How to prevent these dangers while protecting public health in an increasingly integrated world that faces new risks of emerging infectious diseases remains a paramount question for both the United States and Europe. Gostin, Burris, and Lazzarini have described three different models of conceptualizing the role of the individual in the context of infectious diseases and, therefore, how government should shape its public health regulations.15 First, the “microbial model,” which sees disease as a “product of microbial infection,” where it is the task of “public health” “to identify the pathogen and to eliminate or contain it.”16 Second, the “behavioral model,” which focuses on human (risky) behavior, closely monitoring “the activities that give rise to morbidity and premature mortality,” as well as addressing the question of behavior that moves the germ “from person to person or that makes people susceptible to becoming ill when they encounter a pathogen.”17 Third, the “ecological model,” which emphasizes the role of socioeconomic factors in the spread of communicable disease.18 Taking the lens of these three models, Wendy E. Parmet analyzes the United States, and Markus Frischhut presents the EU situation and describes selectively to what extent these models can also be identified in the European Union.

3  Noncommunicable Diseases: Tobacco and Other Unhealthy Commodities The four main groups of NCDs—cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes—have become the most common cause of death and disability 14  Tom L Beauchamp and RR Faden, “The Right to Health and the Right to Health Care” (1979) 4(2) J Med Philosophy 118 120. 15  Gostin, Burris, and Lazzarini (n 13) 69–77. 16  Id., 70. 17  Id., 71–72. 18  Id., 74.

74   Wendy E. Parmet et al. worldwide, accounting for 71% of all deaths and more than three out of four years lived with a disability.19 The human and economic cost of NCDs is immense: they affect the health of individuals, significantly increase the cost to national health services, and entail broader societal costs such as lost productivity and absenteeism related to ill-­ health while increasing health inequities.20 However, many NCDs are eminently preventable by addressing proximal risk factors, including tobacco use and unhealthy diets. Determining whether, and if so, how to regulate the tobacco and other industries that manufacture, distribute, and promote harmful commodities requires a careful engagement with difficult, culturally sensitive questions which the paired chapters on NCDs attempt to address. A critical initial question is the extent to which individuals should take responsibility for their own health and what role, if any, governments should have in limiting the demand and supply for these commodities. As the paired chapters highlight, there is a marked difference between the approaches taken in the United States and in Europe on this question. US public authorities have been notoriously more reluctant to interfere with free market mechanisms than EU authorities, including with respect to tobacco control. This reluctance has been compounded by the US Supreme Court’s sweeping interpretation of the First Amendment to the US Constitution, which protects the right to free speech, including advertising and other forms of commercial expression. Parmet’s discussion of NCDs in the United States engages with the tensions between individual autonomy and public health this stance raises. By contrast, the European Union and its Member States have been, overall, less ideologically opposed to regulating the markets for tobacco products and other potentially harmful commodities—though one notes different stances from one Member State to another. Rather, the key questions that the European Union has faced relate to the nature of the intervention required and the sharing of competences between the Union and its Member States to adopt an effective regulatory framework to protect European citizens from NCDs associated with tobacco, alcohol, and unhealthy diet consumption. In her chapter, Amandine Garde reviews how the European Union and its Member States have shared powers to address the harm resulting from tobacco, alcohol, and unhealthy diets, looking both at the intensity and the nature of their intervention. The two sections of this paired chapter have a similar structure that is intended to facilitate comparisons between the US and EU approaches to the prevention of NCDs. Nevertheless, Parmet and Garde focus on the most topical issues in the United States and Europe, respectively. This explains why the US chapter focuses primarily on tobacco, unhealthy diets, firearms, and opioids, while the EU chapter focuses on 19  UN General Assembly Resolution 73/2 of 10 October 2018 (A/73/L.2) adopting the Political declaration of the third high-­level meeting of the General Assembly on the prevention and control of NCDs, following the high-­level meeting held on 27 September 2018 to undertake a comprehensive review of the prevention and control of NCDs. 20  In 2011, the Harvard School of Public Health and the World Economic Forum estimated that, on top of the social and psychological burdens of chronic disease, the cumulative loss to the global economy could reach $47 trillion by 2030 if things remained as they were: The Global Economic Burden of Non-­communicable Diseases (Geneva: World Economic Forum, September 2011).

Introduction to Public Health Law   75 tobacco, alcohol, and unhealthy diets without referring to opioids and firearms, which have not gained much prominence at the EU level.

4  Social Determinants of Health The WHO defines the social determinants of health as “the conditions in which people are born, grow, live, work and age.”21 These conditions are important contributors to health inequalities. “Socioeconomic health inequalities,” as opposed to inequalities due to biological variations or individual choice, are generally considered to be unnecessary and avoidable.22 However, whether public health law should or can alter these conditions and, if so, how remains highly contested, especially in the United States. In her chapter on the social determinants in the United States, Parmet explains the different pathways through which law can influence the social determinants of health and argues that law should itself be viewed as a critical social determinant. She then reviews some deep-­seated characteristics of US law, including the heavy influence of individualism and the prioritization of negative over positive rights,23 which may help to explain the underinvestment in the United States, as compared to Europe, in social welfare programs. The US section of the chapter concludes with a discussion of several initiatives and proposals in the United States to address social determinants, including “Health in All Policies,” which seek to promote cross-­sectoral collaboration, and recognition of the health impact of non-­health policies. However, such policies, as well as several others discussed in the section, do not engage with broader societal concerns like wealth and inequality. To address these issues, consideration must be given to other laws, including minimum wage laws, labor laws, and the federal tax code, that are not ordinarily thought of as “public health laws.” Whether such fundamental drivers of health should be understood as falling within the purview of public health law or, instead, should be addressed as questions of social justice remains deeply contentious. Brigit Toebes, in her chapter on the social determinants of health in Europe, observes that, at the EU level and within European states, dedicated policies and strategies have been adopted to reduce socioeconomic health inequalities. However, the efforts remain fragmented and there is a lack of evidence as to whether these measures have worked. She also observes that academic research on the role of law in relation to the social determinants of health in Europe is scant compared to what has been done in the United States. In Europe more than in the United States, human rights standards have been linked to the social determinants of health. Toebes argues that human rights law 21  World Health Organization, Social Determinants of Health, https://www.who.int/social_ determinants/sdh_definition/en. 22  WHO, Social Determinants of Health. Available at http://www.who.int/social_determinants/ sdh_definition/en/. Accessed January 2019. See also WHO, 10 Facts on Health Inequities and Their Causes, https://www.who.int/features/factfiles/health_inequities/en/. Accessed 17 January 2019. 23  See Beauchamp and Faden (n 14).

76   Wendy E. Parmet et al. ­ rovides a compelling moral and legal framework for assessing matters of social justice, p including socioeconomic health inequalities. Human rights law can serve as an overarching framework protecting the rights and interests of those affected by health inequalities and poor social conditions. Taking a human rights approach then also requires looking at the right to health in an interaction with other human rights, including rights to housing and social security, as well as labor rights. This chapter makes it clear that in Europe, and much more than in the United States, social rights are a driver for addressing socioeconomic health inequalities.

chapter 4

Com m u n ica bl e a n d Other I n fectious Diseases The EU Perspective Markus Frischhut

1  Introduction and Brief History The history of both EU1 health law in general, as well as more specific laws on ­communicable and other infectious disease, represents the characteristics of European integration at large. EU Member States have always been very reluctant to shift additional competences2 to the EU level. However, they did so when constrained by a common crisis that could not be tackled by a single country acting alone.3 For example, much EU law responds to the failure of the UK government to control the spread of bovine spongiform encephalitis (BSE; linked to new variant Creutzfeldt– Jakob disease) in the mid-­1980s. Another example is HIV/AIDS, due to different national approaches to tackle this issue, since the mid-­1980s. As a result, in 1992, the EU Member States were willing to transfer competences for public health to the European Union. These two examples not only highlight why Member States are willing to shift certain aspects of competences to the EU level: they also show the close link between the European Union’s economic ‘internal market’4 and other areas such as health protection. This is an example of EU integration in one field that breeds further integration in 1  For simplicity’s sake, in the following, reference will always be made to today’s terminology. 2  In US terminology: legal ‘authority’ or legal ‘power’. 3  This general reluctance applies even more in case of sensitive areas, such as public health. 4  An area without internal frontiers in which the free movement of goods, persons, services, and capital is ensured.

78   Markus Frischhut another field, the so-­called spill-­over effect.5 While previously the European Union could regulate to protect against HIV-­contaminated blood products only because of the contamination’s effects on the EU market (e.g., blood products), the European Union nowadays can set standards of quality and safety for blood, organs, and other substances of human origin because of their effects on public health. Communicable diseases not only cross borders, but they also often require measures that cross different areas of policy because of different vectors for disease transmission. The BSE crisis affected food safety, agriculture, and the environment. The relevant EU law cannot be attributed to one sectoral policy only, and thus various EU agencies participate in protecting public health. The key agency for this chapter is the European Centre for Disease Prevention and Control (ECDC). Other important agencies include the European Environment Agency; European Food Safety Authority; and the Consumers, Health, Agriculture and Food Executive Agency. The necessity for joint efforts in public health has led to an increasing body of EU law, not only in the field of communicable disease law as such, but also covering rules on food safety, animal health, and protection of the entire agri-­food chain. To be sure, EU integration has its limits. As in the United States, communicable disease laws are fragmented, with different agencies often acting independently rather than as one. While integration at the EU level has facilitated protection of the public’s health, it also has created potential conflicts among the different objectives of the European Union. The internal market promotes the free movement of products, but public health measures can require restrictions of trade, as with limits on the export of British meat in the wake of the BSE outbreak. Other conflicts can arise if protective public health measures (e.g., slaughtering of contaminated cows or forced blood testing in case of humans) conflict with individual human rights (e.g., right to property or right to bodily integrity). In addition, there are limits to the European Union’s ability to integrate public health law. Just as the United States divides authority for public health regulation between the national and state governments, so is authority for public health matters divided between the European Union and its Member States. Certain competences, including much of health law and the vast majority of criminal law, are still national competences, which can be challenging for the execution of the relevant rules. Conflicts between EU law and diverging national law are solved according to the principle of primacy (similar to the supremacy of national law over state law in the United States). Conflicting national law does not become invalid, merely inapplicable.6 Other similarities between EU and US infectious disease laws include that they very much rely on executive power and are usually reactive. 5  See Markus Frischhut and Scott L Greer, ‘EU Public Health Law and Policy: Communicable Diseases,’ in Tamara K Hervey, Calum Young, and Louise E Bishop (eds), Research Handbook on EU Health Law and Policy (Edward Elgar Publishing 2017) 320. 6  So relevant national provisions remain applicable in those situations which are not affected by EU law (e.g., with regard to third countries such as the United States or in purely internal situations which do not affect the internal market).

Communicable and Other Infectious Diseases   79 This chapter, written before COVID-19, explores the most important features of EU law on infectious diseases. Section 2 describes the allocation of competences between the European Union and its Member States, and Section 3 considers some of the potential conflicts of restrictive measures taken to control or even eradicate infectious diseases and the impact of this on individual’s rights. Against the background of the three elements (risk assessment, risk management, and risk communication) used to deal with the challenges that infectious diseases pose, Section 4 takes a closer look at risk assessment and Section 5 on the different ‘tools’ of risk management. Section 6, finally, turns to the external and ethical perspective and the role the European Union takes in global health.

2  Allocation of Authority in the European Union Although Member States have transferred certain ‘competences’ in ‘public health’ to the European Union, they are still competent for the definition of their health policy and for the organisation and delivery of health services and medical care. Therefore, measures such as isolation and quarantine, and other fields such as criminal law, still fall within national competence.7 Thus, for example, as the Court of Justice (CJEU8) has confirmed, Article 168 of the Treaty on the Functioning of the European Union (TFEU), which relates to the Union’s competence in the field of public health, does not impose any obligation on the Member States relating to the vaccination of minors.9 Vaccination of adults also is a matter for the Member States. Initially, the European Union passed health-­related legislation (e.g., a ban on advertising of tobacco products) based on the Union’s ‘economic’ competence for harmonisation of national laws in the EU single market, to some extent comparable to US ‘interstate commerce’. However, the CJEU has accepted harmonisation rules (e.g., on tobacco advertising) that aim at protecting public health. Indeed, the already mentioned spill-­over effect has led to more competences in the field of ‘public health’ being transferred to the European Union.10 That is why the 7  It is beyond the scope of this contribution to focus on the different approaches in the different Member States, also keeping in mind that federal countries can comprise various different approaches (one federal level and maybe many regions). 8  This abbreviation comprises not only the Court of Justice (CJ), but also the General Court (GC), including the former Court of First Instance. 9  Case C-­459/13 Široká ECLI:EU:C:2014:2120, para 19; see infra Section 5.4. 10  Typologically, public health protection can be divided into four fields of action: prevention, curation, rehabilitation, and terminal care; the EU is mainly competent for primary (first occurrence of diseases) and secondary prevention (individual protection against these diseases). However, due to difficulties in separating tertiary prevention (prevention of the worsening of diseases and disabilities and the prevention of consequential diseases) from curation, it is questionable whether EU competence also covers this tertiary form. Thorsten Kingreen, ‘Art 168 AEUV’ in Christian Calliess and Matthias Ruffert (eds), EUV/AEUV: Das Verfassungsrecht der Europäischen Union mit Europäischer Grundrechtecharta (5. Aufl. C.H. Beck 2016) 1809.

80   Markus Frischhut European Union now can pass legislation necessary ‘to combat the major cross-­border health scourges, measures concerning monitoring, early warning of and combating serious cross-­border threats to health’, namely the Decision on Cross-­border Threats.11 However, it is important to emphasize that this is only a ‘weak’12 competence to ‘support, coordinate or supplement’ the actions of the Member States, which excludes any EU harmonisation of national laws and regulations. The EU can also influence the health sector through financial support. To protect Union citizens from serious cross-­border health threats, EU funding (in total approximately €450 million) also includes identification and development of coherent approaches and their implementation for better preparedness and coordination in health emergencies (e.g., capacity building in Member States and cooperation with neighbouring countries). Under the European Union’s ‘Horizon 2020’ framework programme for research and innovation (2014–2020), the EU has provided €140 million on research on communicable diseases, such as Ebola. Issues about the allocation of authority also arise among different entities within the European Union. For example, the European Commission is the supranational and executive power within the Union and is structured in different Directorates General (DG). As the different directorates pursue their missions, their aims may come into conflict, as with the potential conflict mentioned earlier between a freely flowing internal market and a healthy public, thereby hindering the making of policy. Conflicts not only occur among the different government actors. There also can be conflicts between government action and individual rights. That is the topic of the next Section.

3  Individual Rights 3.1 Overview In EU law, individual rights are often seen as economic (internal market) rights or human rights. In the context of infectious diseases, internal market rights mainly concern the right to free movement of persons and products (e.g., export of meat). This includes the right not to be discriminated against based on citizenship, residence, or origin of the product (e.g., import ban on meat exported from the United Kingdom). 11  European Parliament and Council Decision 1082/2013/EU of 22 October 2013 on Serious Cross-­Border Threats to Health . . . [2013] OJ L293/1, as corrected by [2015] OJ L231/16 (Decision Cross-­border Threats). 12  One example of a ‘strong’ competence is TFEU, art 168 para 4. According to this provision, the EU has a ‘shared competence’ and can also legislate in the context of organs and substances of human origin, blood and blood derivatives (lit a), and measures in the veterinary and phytosanitary fields (lit b), as well as medicinal products and devices for medical use (lit c). The second category (lit b) is important for the context of BSE because the EU has a health-­related competence which is independent from the agricultural competence.

Communicable and Other Infectious Diseases   81 Human rights (e.g., right to liberty, respect for physical and mental integrity, data protection, nondiscrimination) are now mainly enshrined in the Charter of Fundamental Rights of the EU (CFREU). They can play an important role in the context of the ‘behavioural model’13, as when measures are taken that restrict individual liberty (e.g., controlling and limiting the movement or behaviour of infected people). Rights are subject to override for compelling reasons, and both internal market rights and human rights can be restricted by so-­called overriding reasons, such as public health (e.g., no free movement for BSE-­contaminated meat). However, those restrictive measures must be appropriate for ensuring the attainment of that objective and not go beyond what is necessary to attain the objective pursued (e.g., import ban for meat should apply only to BSE-­affected areas). Another typical example would be labelling requirements, which are clearly less restrictive compared to a total ban on the relevant product if the objective of public health protection can still be achieved. The wording of this principle of proportionality is reminiscent of the US case Jacobson v. Massachusetts, which disapproved government actions that ‘go so far beyond what was reasonably required for the safety of the public’.14 Several interventions are discussed later, especially in Section 5. For now, it is useful to consider some more general points. Due to the broad range of the ‘public health toolkit’, there are many potential infringements of human rights. Forced testing for a communicable disease can impede human dignity (CFREU Article 1, TEU Article 2) as well as the right to the integrity of the person (CFREU Article 3). Information forwarded in the event of a crisis can impede the protection of personal data (CFREU Article 8), and the slaughtering of BSE-­contaminated animals can interfere with the right to property (CFREU Article 17). A virus such as Zika and discussions about abortion can even trigger debates about the right to life (CFREU Article 2). EU law stipulates protection against unjustified discrimination on the basis of citizenship or residence. In addition, CFREU Article 21 para 1 protects against discrimination on the basis of race, sex, and other characteristics.

3.2  Right to Healthcare What seems like the most obvious human rights provision in the context of infectious diseases is CFREU Article 35 para 1 on ‘healthcare’, according to which everyone (thus not only an EU citizen) has the right of access to preventive healthcare and the right to benefit from medical treatment. However, there are two important limitations. First, this right can only be enjoyed ‘under the conditions established by national laws and practices’. Hence, it does not create a real ‘added value’. Second, beside this reference to national laws, CFREU Article 35 para 1 is merely seen as a negative right (to 13  See Parmet, Frischhut, Garde, and Toebes, ‘Introduction to Public Health Law,’ in this volume. 14  Jacobson v Massachusetts, 197 US 11 (1905), at 28. See chapter by Parmet, ‘Communicable Disease,’ in this volume.

82   Markus Frischhut noninterference15), where a positive right (to benefits) has only been accepted if the right to life (CFREU Article 1) or the right to integrity (CFREU Article 3) is at stake. Accordingly, in EU law, the right to healthcare can be invoked to oppose government interference with an individual’s efforts to obtain healthcare but not to require the government to provide the care. While the right to healthcare plays a limited role in public health law, other rights are significant, including individual rights to liberty, nondiscrimination, and property (including compensation), as well as a possible right not to be treated and procedural rights. These other rights will be discussed in following subsection.

3.3  Right to Liberty Under the European Convention on Human Rights (ECHR), which is also binding on EU Member States and their courts, individuals enjoy a right to liberty. However, individuals can be deprived of their liberty when they are a danger to ‘public safety’. Thus, the government can detain persons ‘for the prevention of the spreading of infectious diseases’.16 For example, the European Court of Human Rights (ECtHR) has balanced the interest of the individual in liberty with the public interest in safety to ensure that an infectious virus such as HIV does not spread. The key question is ‘whether detention of the person infected is the last resort in order to prevent the spreading of the disease, because less severe measures have been considered and found to be insufficient to safeguard the public interest’; in case these criteria are no longer fulfilled, ‘the basis for the deprivation of liberty ceases to exist’.17 There also is a right to liberty in the CFREU (Article 6), but so far it has led only to CJEU case law on procedural issues and not about the substance of this right because the European Union lacks public authority in restricting this liberty.

3.4 Nondiscrimination As a key principle, nondiscrimination prohibits the application of different rules to comparable situations or the application of the same rule to different situations. EU law not only prohibits discrimination based on citizenship, residence, or the origin of the product in the context of economic ‘internal market’ rights, but also based on other criteria (CFREU Article 21), such as ‘sex, race, colour, ethnic or social origin, genetic fea15  Tom L Beauchamp and RR Faden, ‘The Right to Health and the Right to Health Care’ (1979) J Med Philosophy 4(2), 118, 120. 16  ECHR, art 5 para 1 lit e (right to liberty and security). 17  Enhorn v Sweden App no 56529/00 (ECtHR, 25 January 2005), paras 43–44; in addition, such a detention has to be ‘lawful’ and free from arbitrariness (para 36).

Communicable and Other Infectious Diseases   83 tures, language, religion or belief, political or any other opinion, membership of a national minority, property, birth, disability, age or sexual orientation’. In the French case Léger concerning the exclusion from blood donation of men who have had sexual relations with other men, the CJEU had to deal with a case of alleged discrimination based on sexual orientation. A French court had referred the question to the CJEU. This case was, inter alia, about HIV. Asked whether this French policy constitutes discrimination based on sexual orientation, the CJEU left it to the national court to ascertain if there are effective techniques for detecting HIV in order to avoid transmission to recipients.18 However, the CJEU emphasised the objective of ensuring a high level of human health protection, which can justify a permanent deferral from blood donation. In reaching its decision, the CJEU took a close look at the specific epidemiological situation in France, took into account a report of the ECDC, and emphasized the need to consider ‘scientific and technical progress’. Hence, public health considerations can be important enough to justify the exclusion, if such a limitation is provided for by law, respects the essence of those rights and freedoms, and respects the principle of proportionality. In its follow-­up decision, the French court emphasized that such a general and permanent ban on blood donation is against the EU principle of proportionality19 and therefore unlawful.20 Discrimination is often linked to stigmatisation. In the context of the severe acute respiratory syndrome (SARS) outbreak in 2003, some people from Asia were, due to media reports, perceived as a threat purely because of their appearance. Such stigmatisation can also be counterproductive in that people actually affected may be discouraged from seeking timely medical assistance.

3.5  Right to Property and Compensation for Damages Another right often invoked in this context is the right to property (CFREU Article 17). According to this, ‘[n]o one may be deprived of his or her possessions, except in the public interest and in the cases and under the conditions provided for by law, subject to fair compensation being paid in good time for their loss’. As the CJEU has stated, ‘fundamental rights are not absolute rights but must be considered in relation to their social 18  Case C-­528/13 Léger ECLI:EU:C:2015:288, paras 44, 63. 19  This is reminiscent of the US ‘strict scrutiny’, where the state can override a fundamental right with a compelling public interest unless there is a ‘less restrictive alternative’. The author would like to thank the editors and Wendy E. Parmet for valuable feedback in this regard; on ‘strict scrutiny’; see Parmet, ‘Communicable Disease,’ in this volume. 20  Tribunal administratif de Strasbourg 8 March 2016 Case no 0903177 Follow-­up decision Léger. According to the Tribunal, there was no appeal against this decision (personal communication from 7 May 2019, on file with the author). This example of refusing the blood donation of Mr Léger on the ground that he had had sexual relations with another man can be attributed to the above-­mentioned ‘behavioural model’.

84   Markus Frischhut function’.21 Furthermore, ‘the objective of protection of health takes precedence over economic considerations . . . , the importance of that objective being such as to justify even substantial [!] negative economic consequences’.22 Infectious diseases can be transmitted not only by humans but also by animals. In a case of fish stocks infected by viral haemorrhagic septicaemia and infectious salmon anaemia, the CJEU had to decide whether the right to property should be interpreted as imposing on a Member State (the UK) the obligation to adopt measures providing for the payment of compensation for owners affected by those measures. According to the CJEU, EU law does not require compensation to be paid in all circumstances if this does not constitute disproportionate and intolerable interference impairing the very substance of the right to property.23 The measures taken in this case did not deprive farm owners of the use of their fish farms but enabled them to continue to carry on their activities as the immediate destruction and slaughter of all the fish enabled owners to restock the affected farms as soon as possible. Similar to the United States, in this context the CJEU grants the EU legislature a wide margin of discretion, here in the field of agricultural policy. In a Dutch case of foot-­and-­mouth disease, the CJEU has stressed several principles, including the danger of the disease to be controlled, the right to property, the principle of proportionality, and animal welfare.24 Based on the understanding that foot-­and-­ mouth disease constituted a ‘serious hazard’ for animals, Member States had the power to adopt control measures, including the power to order the slaughter of animals belonging to a holding adjacent to or within a specific radius of a holding containing infected animals.25 We now turn to some of the tools public authorities need in order to protect public health effectively. This includes the necessity to identify communicable disease threats quickly so that public authorities can take the steps necessary to prevent or contain outbreaks.

4  Risk Assessment Infectious diseases require public authorities to handle risk. The analysis of risk comprises risk assessment, risk management, and risk communication. The Ebola crisis was one prominent example where the necessity to strengthen risk assessment and risk 21  Joined Cases C-­20/00 and C-­64/00 Booker ECLI:EU:C:2003:397, para 68. 22  Case C-­151/17 Swedish Match ECLI:EU:C:2018:938, para 54. 23  Booker (n 21) paras 85–86. 24  Case C-­96/03 Tempelman and van Schaijk ECLI:EU:C:2005:145, paras 48–49. 25  So far, the discussion has centred on substantive rights. CFREU Article 47 (right to an effective remedy and to a fair trial) plays an increasingly important role in terms of procedural rights. This includes an effective remedy before a tribunal and a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law, as well as legal aid for those who lack sufficient resources.

Communicable and Other Infectious Diseases   85 management of serious cross-­border threats to health was one of the ‘lessons learned’ by the European Union. These lessons learned and identified by the EU health ministers comprise, amongst others, a better definition of EU medical evacuation capacities in preparation for possible future emergencies, as well as the necessity to strengthen preparedness research, especially with regard to diagnostic methods and vaccines and to improve coordination between the European and global research communities. Although the proper assessment of risk must be conducted by scientists, risk assessment involves, first of all, determining what level of risk is deemed unacceptable. This task lies with both EU institutions and Member States, which are legally required to achieve a high level of health protection.26 Second, risk assessment requires a scientifically based process, one that involves hazard identification, hazard characterisation, appraisal of exposure, and risk characterisation. Such scientific advice must be based on the principles of excellence, independence, and transparency. Once such advice is handed over from scientists to politicians, the competent public authority must be given sufficiently reliable and cogent information to allow it to understand the ramifications of the scientific issues raised and take policy decisions in full knowledge of the facts. This high-­quality scientific advice shall then enable the competent authority to choose the appropriate and necessary risk management measure from the broad range of the ‘public health toolkit’, which helps to prevent the risk from materialising. At the EU level, risk assessment is provided by EU agencies in accordance with their missions, respectively by the European Commission. The results are then communicated to the national competent authorities via the Early Warning and Response System (EWRS). The key agency for our topic is ECDC, and high-­quality advice is key, too. Thus, it is worth emphasising the requirements with regard to scientific excellence, where it is vital to ensure ECDC’s independence, high scientific quality, transparency, and efficiency. This is in line with the requirement of evidence-­based science and the avoidance of conflicts of interest. In case of scientific uncertainty about the existence or extent of risks to human health or the environment, the precautionary principle allows the European Union or the Member States to take protective measures without having to wait until the reality and seriousness of those risks become fully apparent or until the adverse health effects materialise.27 This principle is enshrined, although not defined, in EU primary law and has been qualified as a general principle of EU law28 that requires public authorities to take appropriate measures to prevent specific potential risks to public health, safety, and the environment by giving precedence to the requirements related to the protection of those interests over economic interests. This principle has to be seen against the background of a ‘high level’ of human health protection, which grants policymakers broad ­discretion. 26  Based on this ‘high level’ of health protection, for instance, the CJEU has accepted a possible exclusion from blood donation of men who have had sexual relations with other men. 27  Case T-­584/13 BASF Agro and Others v Commission ECLI:EU:T:2018:279, para 59. 28  For further details, see Frischhut and Greer (n 5) 331–333.

86   Markus Frischhut While it might be difficult to determine the exact threshold, there must be at least some discernible risks based on science.29 To sum up, the acceptable level of risk in a particular situation is determined by politicians. This is followed by independent scientific assessment to decide whether the health threat would exceed the legislatively determined level of acceptable risk. Finally, the task falls again to the politicians to decide on the corresponding risk management measures. While it is important that the steps of risk assessment and risk management are separate, there will often be an ‘ongoing exchange’ between them. The EU approach in this regard generally can be welcomed. However, the handling of the contested issue of glyphosate, a weed killer which the World Health Organisation (WHO) found as ‘probably’ carcinogenic, is one example where an application of the substance of the precautionary principle, as well as the requirement of striving for a ‘high level’ of human health protection, could and should have led to a more restrictive approach. In this context, the Commission itself has acknowledged the necessity of increasing transparency in case of scientific studies submitted by the industry.30

5  Risk Management Risk management comprises a broad range of tools (e.g., vaccination, isolation and quarantine, travel restrictions) which mainly fall within the competences of Member States. Since it is feared that Member States, in reality, want to take measures aiming at protecting their own economy from imported products, they must adhere to certain principles, such as proportionality, nondiscrimination, transparency, and consistency with similar measures already taken. However, de Ruijter has shown that even when a risk management measure concerns national competences such as vaccination strategies or communication with the public that don’t implicate trade, the European Union is ‘highly involved’.31

5.1  Emergency Preparedness As is often said, an ounce of prevention is worth a pound of cure. Preparedness and response planning is key for effective monitoring, early warning, and combating of serious cross-­border threats to health. This includes adequate preparedness of critical 29  The threshold for triggering the application of the precautionary principle is lower in case of provisional measures. 30  Commission, ‘Questions & Answers: Commission Replies to European Citizens' Initiative on Glyphosate and Announces More Transparency in Scientific Assessments’ (Fact sheet) MEMO/17/5192. 31  Anniek de Ruijter, ‘A Silent Revolution: The Expansion of EU Power in the Field of Human Health: A Rights-­Based Analysis of EU Health Law & Policy’ (PhD thesis, University of Amsterdam 2015) 205, 212.

Communicable and Other Infectious Diseases   87 sectors of society (energy, transport, communication or civil protection) which have to rely on well-­prepared public health systems, and vice versa. The cross-­cutting nature and the different types of infectious diseases (foodborne diseases, zoonotic diseases, waterborne diseases, and other) are well-­reflected in the information that Member States have to provide on preparedness and response planning. Based on the Member States’ competences32 in this field and their different resources and administrative and political structures, the way in which emergency preparedness measures can be designed in substantive terms can vary greatly. Although the European Union has no competence to take preparedness measures, it can at least coordinate the Member States’ efforts. Member States and the Commission consult each other within the Health Security Committee (HSC)33 to coordinate efforts in monitoring, early warning and assessment of, and response to serious cross-­border threats to health.34 This includes well-­known ‘soft tools’, such as the exchange of best practices and the promotion of interoperability. One example of coordinating Member States’ competences in this field is the development of training programmes for public health and preparedness professionals. The already-­mentioned cross-­cutting nature also can be seen in terms of ensuring interoperability between the health sector and other sectors, including the veterinary sector, that are identified as being critical in the case of an emergency. As one of the lessons learned from the BSE crisis, preparedness also involves the ability to know through all stages of production, processing, and distribution how and where problematic food, etc. has entered the system. The key word is ‘traceability’, which enables targeted and accurate withdrawals and the provision of information to the public. At the same time, beside the protection of public health, we can also clearly see the economic interest of avoiding unnecessary distortions of the single EU market. Another example to overcome the challenge of missing EU competences is joint procurement of medical countermeasures,35 such as flu vaccines, which is a voluntary possibility for Member States to achieve a better bargaining position and, consequently, a cheaper price.

5.2  Surveillance and Disease Reporting According to the WHO, public health surveillance ‘is the continuous, systematic collection, analysis and interpretation of health-­related data needed for the planning, implementation, and evaluation of public health practice’.36 32  While there are no explicit and specific preparedness principles, in this primarily intergovernmental field, Member States are bound to already covered general principles of EU law, such as proportionality, nondiscrimination, etc. 33  The HSC is an ad hoc committee composed of high-­level representatives of the Member States. 34  Decision Cross-­Border Threats, art 4. 35  Decision Cross-­Border Threats, art 5. 36  WHO, ‘Public Health Surveillance’. Available at http://www.who.int/topics/public_health_ surveillance/en. Accessed 29 August 2019.

88   Markus Frischhut As mentioned, the ECDC is the key EU agency for protecting against infectious diseases. The ECDC is in charge of operating a network for epidemiological surveillance of communicable diseases and related antimicrobial resistance issues. The ECDC has no regulatory powers. It collects, analyses, and disseminates surveillance data on infectious diseases and other cross-­border threats to health from the Member States. The ECDC has progressively taken over the epidemiological surveillance of infectious diseases and the operation of the EWRS, the European Union’s rapid alert system for serious cross-­ border threats to health. Alerts must be reported by the Member States or by the Commission if the emergence or development of a serious cross-­border threat to health fulfils the following three criteria: it is unusual or unexpected, it affects more than one Member State, and it requires a coordinated response at the EU level. A recent paper provides the Italian case study of extensively drug resistant tuberculosis (XDR-­TB) as an example fulfilling these prerequisites.37 In case of a report, the Commission also has to make promptly available to the national competent authorities and to the HSC, through the EWRS, a risk assessment of the potential severity of the threat to public health, including possible public health measures. Member States also must submit an alert via the EWRS if they notify the WHO of events that may constitute public health emergencies of international concern (PHEIC). Based on all this, the Member States then consult each other in the HSC and, in liaison with the European Commission, coordinate national risk management measures and risk and crisis communication (see Section 5.3). In cases of urgency, Member States can take measures before consultation and notify other States immediately upon adoption. In case a Member State cannot deal with a serious cross-­border threat to health, it can also request assistance from other Member States. As mentioned in Section 3, if Member States mandate that individuals be screened or tested for communicable diseases, they have to comply with the human rights of individuals, as enshrined in the CFREU. The US requirement that testing must be reasonable finds an equivalence in the EU principle of proportionality. The General Data Protection Regulation (GDPR) provides for special restrictions concerning ‘data concerning health’. While processing of such personal data is basically prohibited, the GDPR foresees, amongst others, an exception if the processing is necessary ‘for reasons of public interest in the area of public health, such as protecting against serious cross-­border threats to health’.38

37  Patrycja Dąbrowska-­Kłosińska, ‘Tracing Individuals Under the EU Regime on Serious, Cross-­ Border Health Threats: An Appraisal of the System of Personal Data Protection’ (2017) Eur J Risk Regulation 8(4), 700. 38  European Parliament and Council Regulation (EU) 2016/679 of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) [2016] OJ L119/1, as corrected by [2018] OJ L127/2 (GDPR), art 9.

Communicable and Other Infectious Diseases   89

5.3  Risk Communication When analysing literature on effective ‘risk communication’, an ECDC report revealed much room for improvement, where the example of the 2009 H1N1 situation draws a relatively negative picture on this tool which is key both for involving all stakeholders and for building trust. ‘Risk communication’39 can be seen as closely related to risk management, though it was addressed by the Commission as distinct from risk assessment and risk management. Nowadays, risk communication is key in the context of trust, can have an important impact on the effectiveness of the response from a public health perspective, and should be addressed (externally) to the public and key stakeholders, such as healthcare professionals. At the EU level, risk communication is a competence of both the HSC and the ECDC, which, in close collaboration with the Member States and the Commission, promote the necessary coherence in the risk communication process on health threats. In addition to its ‘external’ risk communication (i.e., to the public), ECDC is also responsible for ‘internal’ risk communication and especially coordination of the various mechanisms and structures due to the cross-­cutting dimension. As confirmed by a technical report of ECDC, to be successful, risk communication materials have to be easily comprehensible and comprise adequate visualisation.

5.4 Vaccination The ‘public health toolkit’ also includes vaccination of both animals40 and humans. Similar to the United States, many people in Europe are sceptical about vaccination. A recent Eurobarometer survey has revealed that while 80% of respondents state that vaccines are rigorously tested before being authorised for use, 55% state that vaccines overload and weaken the immune system, 49% think that vaccines can cause the disease against which they protect, and, finally, 41% of respondents answered that vaccines often produce serious side effects.41 Different countries have adopted various approaches to vaccination.42 For the usual childhood immunisations, such as polio, measles, mumps, pertussis, and tetanus, some 39  ‘Risk communication’ should be seen separately from ‘crisis communication’, where the first requires a permanent process, while the second is occasional. 40  On questions concerning the vaccination of animals against foot-­and-­mouth disease, see Case C-­189/01 Jippes and Others ECLI:EU:C:2001:420. 41  European Commission, ‘Europeans’ Attitudes Towards Vaccination: Special Eurobarometer 488’ (April 2019) 28. This report (p. 32) has also shown huge differences between EU countries; for example, on the statement of vaccines often producing serious side effects, ranging from 69% (Sweden) to 20% (Cyprus). 42  For the following overview and further details, see European Observatory on Health Systems and Policies, The Organization and Delivery of Vaccination Services in the European Union: Prepared for the European Commission (World Health Organisation 2018); and Elena Bozzola and others, ‘Mandatory

90   Markus Frischhut countries take a compulsory approach, while in other countries, the vaccinations are recommended but ultimately voluntary. Thus, for example, in the nine countries of Bulgaria, Croatia, Czech Republic, France, Hungary, Italy,43 Poland, Slovakia, and Slovenia, childhood immunisations generally are required, while countries such as Austria, Finland, the Netherlands, Norway, Spain, Sweden, and the United Kingdom take a voluntary approach. However, even among voluntary countries, vaccination may be mandatory in practice. In some countries, such as Greece, proof of immunisation is necessary when enrolling in kindergarten or primary school. Similarly, in Germany, vaccinations are voluntary, but children are only admitted to daycare facilities if they have received the standard vaccinations; in addition, in case of a measles or other outbreak, the daycare facility or school is allowed to exclude unvaccinated children from attendance. The recent measles outbreak has led to greater consideration of vaccination obligations. For adults, mandatory vaccination is uncommon. In the case of influenza, for example, vaccination for adults is voluntary in almost all EU Member States, with the exception of Slovakia, where vaccination against influenza is mandatory for persons living in social care facilities and for persons living or working in an area with an increased risk of infection. In a Slovakian case, the CJEU had to deal with a mother refusing to vaccinate her child, which under Slovakian law resulted in a fine of €100 (of a maximum possible amount of €331). This case would have entailed intriguing questions about whether there is a right not to be vaccinated or, in a broader sense, not to be treated. However, the Court refused to assert jurisdiction for this case, given the fact that TFEU Article 168 (the EU’s public health competence) does not impose any obligation on the Member States relating to the vaccination of minors.44 If the obligatory character were to change from a mere fine to an actually enforced vaccination, the right to physical integrity would also have to be discussed. However, as the European Union lacks competence in the field of vaccination of humans and EU rules on cross-­border healthcare exclude public vaccination programmes against infectious diseases, these questions have to be resolved in the various Member States. While the European Union cannot require vaccination, it can take steps to facilitate its availability, using its marketing authorisation powers for pharmaceuticals. Under certain circumstances defined in the Decision on Cross-­border Threats, the European Commission can declare a ‘public health emergency’ (PHE), for example, when the WHO has been informed and has not yet adopted a decision declaring a PHEIC.45 If Vaccinations in European Countries, Undocumented Information, False News and the Impact on Vaccination Uptake: The Position of the Italian Pediatric Society’ (2018) Ital J Pediatr 44(1), 67. 43  Walter Ricciardi, Stefania Boccia, and Roberta Siliquini, ‘Moving Towards Compulsory Vaccination: The Italian Experience’ (2018) Eur J Public Health 28(1), 2. 44  Široká (n 9) para 19; the CJEU also did not review this national law with regard to the CFR (paras 20–26). 45  World Health Organisation, International Health Regulations (2005) (2d ed, World Health Organisation 2008): art 1 defines a PHEIC, and art 12 provides for the procedure of determination of a PHEIC (see also Annex 2); see also art 57 para 3.

Communicable and Other Infectious Diseases   91 there is an ‘emergency situation’, a special procedure for authorisation of medicinal products comes into effect.46 De Ruijter has very well addressed potential problems of both determining priority groups for vaccination, as well as safety issues of vaccines authorised in an emergency procedure.47 In the case of normal and emergency procedures, safety issues of vaccination can lead to questions of liability, in this case not of the European Union or Member States (state liability), but of producers, etc. of vaccines according to the general EU product liability rules. Whereas in the United States vaccine manufacturers have received ‘significant liability protection’,48 the CJEU had to deal with such a case of potential product liability for an allegedly defective vaccination produced by Sanofi Pasteur. Mr W was vaccinated against hepatitis B through three injections, administered on 26 December 1998, 29 January 1999, and 8 July 1999 and then, from August 1999, began to present with various troubles, which led to a diagnosis of multiple sclerosis in November 2000. He finally died in October 2011. His family’s arguments were based on the short period between the vaccination and the appearance of the first symptoms of multiple sclerosis, in conjunction with the lack of any personal or family history of the disease. Thus, this raised the question of whether there was a causal link between the vaccine and his disease and eventual death precisely in a situation where ‘medical research neither establishes nor rules out the existence of a link between the administering of the vaccine and the occurrence of the victim’s disease’. The CJEU concluded that ‘serious, specific and consistent evidence’ allowed the national referring French court to base its decision on an assumption that ‘there is a defect in the vaccine and that there is a causal link between that defect and that disease’.49 This judgment addresses the safety issues described by de Ruijter in 2015. However, in the end, the applicants were denied compensation by the relevant French court.50 It is easy to complain about vaccination hesitancy while not guaranteeing transparency in authorisation procedures, addressing (hopefully past) practices of payments of the pharma industry to medical doctors, or sweeping possible side effects under the table. As the European Parliament has recently persuasively stated, ‘transparency in the process of evaluating vaccines and their adjuvants, and the funding of independent research programmes on their possible side-­effects, would contribute to restoring confidence in vaccination’.51 The EU Council of Ministers and the European Parliament recently addressed important issues such as researchers’ conflicts of interest and more 46  On medicines’ regulation, see chapters by Lietzan and Zettler and Mahalatchimy, in this volume. 47  Ruijter (n 31) 256–258. 48  Wendy E Parmet, ‘Pandemics, Populism and the Role of Law in the H1N1 Vaccine Campaign’ (2010) St Louis U J Health Law Pol 4, 113, 152–153. 49  Case C-­621/15 W and Others ECLI:EU:C:2017:484 para 43. Such an assumption (under French law) is possible as long as it does not apply automatically (i.e., if the causal link between defective vaccine and the damage ‘will always be considered’), para 55. 50  Estelle Brosset, ‘Distinguishing Between Law and Science in Terms of Causation and the Hepatitis B Vaccine: W v Sanofi Pasteur’ (2018) Common Market L R 55(6), 1899, 1914. 51  European Parliament resolution of 19 April 2018 on vaccine hesitancy and the drop in vaccination rates in Europe (2017/2951(RSP)), P8_TA(2018)0188 (EP resolution vaccine hesitancy), point 20.

92   Markus Frischhut transparency in EMA’s authorisation process of new vaccines, as well as, most importantly, ‘the importance of transparency in building and maintaining public trust in medicines’,52 also by ‘strengthen[ing] dialogue with citizens’.53 Beside transparency, also a ‘no-­fault compensation program for vaccine-­related injuries’54 could help to overcome vaccination hesitancy.

5.5  Isolation and Quarantine ‘Isolation’ and ‘quarantine’55 are amongst the oldest tools in the fight against communicable diseases. And they can be essential, especially with highly transmissible viruses such as Ebola. But their major interference with personal liberty can trigger serious human rights issues. Although to the author’s knowledge no cases were pending before the CJEU, measures in the field of isolation and quarantine can violate numerous rights. These include the right to physical and mental integrity (CFREU Article 3), the right to liberty (CFREU Article 6), and the respect for private and family life (CFREU Article 7), as well as procedural rights (CFREU Articles 47, etc.). The Ebola crisis stands out as a negative example where decisions to quarantine individuals returning from Africa ‘were based on fear rather than on evidence-­based risk assessment’.56 An important adjunct to isolation and quarantine is traceability, both of humans and animals (‘from farm to fork’). As in the case of vaccination, in the context of contact tracing it is up to the Member States to identify those persons already contaminated and those persons exposed to risk. Also similarly, contact tracing can trigger human rights concerns, particularly rights to privacy and data protection. De Ruijter mentions examples of tracing humans during the 2009 swine flu outbreak.57 There are other, less extreme steps than quarantine to interrupt the spread of communicable diseases. For example, public health authorities may want to close schools, suspend public transport, or ask people to stay at home. If such cross-­cutting action is needed, actions can be coordinated by either the HSC or another informal, intergovernmental, and executive body as the ‘friends of the presidency’ group. For the first time, school closures were unanimously adopted by the HSC and EWRS authorities with regard to pandemic influenza A (H1N1) in 2009. This coordinated EU 52  EP resolution vaccine hesitancy, points 5, 6, 14. 53  Council of the EU recommendation of 7 December 2018 on strengthened cooperation against vaccine-­preventable diseases [2018] OJ C466/1. 54  For the US example, see Parmet, ‘Communicable Disease,’ in this volume. 55  According to Zita Lazzarini, ‘Communicable Disease Law and Emerging Issues: Antibiotic Resistance’ in IG Cohen, Allison Hoffman, and William M Sage (eds), The Oxford Handbook of US Health Law (Oxford University Press 2017) 971, ‘isolation’ refers to confinement of a person known to have a communicable disease, while ‘quarantine’ refers to the confinement or separation of a person who has been exposed to a communicable disease in order to find out whether he or she is infected. 56  Aura Timen and others, ‘The Ebola Crisis: Perspectives from European Public Health’ (2015) Eur J Public Health 25(2), 187. 57  Ruijter (n 31) 210–211.

Communicable and Other Infectious Diseases   93 approach stated that there was no need ‘to enact pre-­emptive mass school closures’; however, it was recommended to ‘locally close schools reactively upon infection being found among students as beneficial to delay the transmission of the virus’.58 The document also addressed another issue: how to balance freedom of movement and public health protection for travellers with flu symptoms.

5.6  Travel Restrictions Freedom of movement is a key principle in EU law, one that can be restricted in exceptional circumstances (e.g., if necessary to safeguard ‘public health’). The challenge is thus to avoid arbitrary border measures while at the same time allowing Member States to carry out necessary checks on travellers. This notion of ‘public health’ is not within national discretion but is an autonomous notion of EU law, shaped by the CJEU and understood in a very narrow sense. In our context of communicable diseases, for this notion of ‘public health’, EU law sometimes even refers to the international level of the WHO. Such travel restrictions can occur both from an inbound, as well as from an outbound perspective.59 In case of EU citizens, EU law limits such inbound restrictions on the freedom of movement to, first, diseases with epidemic potential as defined by the relevant instruments of the WHO, such as Ebola or Lassa fever; here, the European Union delegates its task of defining those diseases to the WHO. Inbound restrictions are also permitted for other infectious diseases when a Member State has imposed analogous restrictions on its own citizens. In other words, EU law allows Member States to adopt restrictions on inbound travellers from other EU countries as long as they do so without discriminating against noncitizens. Within 3 months after the date of arrival, Member States can require persons to undergo medical examinations only in exceptional cases and free of charge. After those 3 months, expulsions cannot be based on diseases. Thus, reasons related to ‘public health’ can only restrict the entry and short-­term residence of EU citizens and family members. As can be clearly seen, this means that there is a very high threshold for restricting movement rights on the basis of health considerations. Similar rules of EU law, also referring to the WHO, apply to certain third-­country nationals (TCN). In case of an application for entry and residence of TCN family members in the context of family reunification, the relevant directive only refers to ‘public health’, without further reference to the WHO. Likewise, in case of a highly qualified TCN applying for an EU Blue Card, Member States can refuse admission or withdraw or decline to renew it if they pose a threat to ‘public health’. In today’s situation of migration, in the case of applicants for international protection (i.e., refugee status or subsidiary 58  Commission, ‘Influenza A (H1N1) 2009: EU Health Security Committee Agrees Statements on School Closures and Travel Advice’ IP/09/1234. 59  For further details concerning the next paragraphs, see Frischhut and Greer (n 5) 337–340.

94   Markus Frischhut protection), Member States can require medical screening of applicants on grounds of ‘public health’. Most Member States carry out such tests on a mandatory basis, where these tests have to be suitable, necessary, and proportionate, which, for instance, excludes mandatory HIV screening. In its 2001 proposal, the European Commission had mentioned tuberculosis testing to be carried out on a systematic basis. Recently, the European Union has adopted a European Travel Information and Authorisation System (ETIAS) for TCNs exempt from visa requirements when crossing EU external borders, enabling consideration of whether the presence of those TCNs on EU territory would pose ‘high epidemic risk’, where the latter refers to IHR, ECDC, or national definitions. Both the CJEU and the ECtHR have had to deal with foreigners from an outbound perspective; that is to say, expulsion and possible health-­related limitations. These cases centred on the question of whether a communicable disease such as AIDS could be a reason for wanting to remain in a country because it is the only place where appropriate medical treatment can be provided. Foreigners who are subject to expulsion in principle cannot claim any entitlement to remain in the territory in order to continue to benefit from medical, social, or other forms of assistance. However, under very exceptional circumstances, an expulsion depriving aliens of access to treatment can amount to ‘inhuman and degrading treatment’60 if a high61 threshold is reached. According to the ECtHR, this can cover the expulsion of an alien suffering from AIDS if his removal would expose him to ‘a real risk of dying under most distressing circumstances’.62 This case law has also been adopted by the CJEU.63

6  Global Health Law and Ethics Perspective Diseases cannot only cross borders between, but obviously also beyond EU Member States, thus the necessity to take into account the global perspective. The Ebola crisis, for instance, has triggered discussions about permanent European rapid response capability, comprising experts, laboratory support staff, epidemiologists, and logistics facilities, including mobile laboratories that can be swiftly deployed. Based on the various internal and other post-­Ebola reviews, some voices have recently called for an EU ‘global health strategy’, not only because pandemics are likely to arise from outside Europe. As often occurs in these discussions, the question is whether EU external health activities 60  ECHR, art 3. 61  N v UK App no 26565/05 (ECtHR, 27 May 2008), para 43. 62  Paposhvili v Belgium App No 41738/10 (ECtHR, 13 December 2016), para 177 (emphasis added). Even if a person’s life would not be at stake, this high threshold can be reached. This would be the case if a person subject to expulsion and thus resulting in a lack of appropriate treatment, ‘would face a real risk . . . of being exposed to a serious, rapid and irreversible decline in his or her state of health resulting in intense suffering or to a significant reduction in life expectancy’ (para 183, emphases added). 63  Case C-­353/16 MP ECLI:EU:C:2018:276, paras 40–41.

Communicable and Other Infectious Diseases   95 in the field of communicable diseases should focus on the EU’s security interests or on the priorities of the countries in need. There are various provisions tasking the European Union with external health-­related activities, such as cooperation between the European Union and its Member States on the one side, and third countries and competent international organisations, on the other. The European Union cooperates not only with WHO, but also with the World Bank, the Global Fund, and other bodies in the context of global health governance.64 In these forms of cooperation, the European Union must promote its values and interests, which not only comprise general values, but also the Union’s health values of universality, access to good-­quality care, equity, solidarity, and achieving a ‘high level’ of human health protection. In the European Union’s Blue Card regime for TCNs, we can find one example of an EU external paternalistic ethical approach. To guarantee ethical recruitment policies, Member States should refrain from pursuing active recruitment in developing countries in sensitive sectors suffering from a lack of personnel, such as the health sector, in order to avoid a brain-­drain in these countries.

7  Conclusion and Outlook Infectious diseases are a cross-­cutting issue as there are various ways they can be transmitted, including by animals and plants. From an external, inbound perspective, the European Union controls imports from third countries of plants, animals, and animal products. EU infectious disease law and policy offer a well-­known example of Member States’ reluctance to transfer additional competences to the EU level, although doing so when forced by public health crises. Such crises occur due to diseases crossing borders which initially have been opened for humans, animals, and plants based on the Union’s single-­market concept. EU infectious disease law often involves balancing divergent interests through the principle of proportionality, which can be found in Europe both at EU and at national levels (e.g., Léger case).65 Likewise, it applies both within the economic ‘internal market’ as well as within human rights. The CJEU finds a balanced solution in this respect. Considering economic ‘fundamental freedoms’, on the one hand, and other noneconomic interests such as health protection or human rights, on the other, case law has emphasised that the protection of human health ‘must take precedence over economic considerations’.66 From the different branches of power, the judiciary has very much shaped EU law on communicable diseases, as it has done in the United States. Important principles have 64  Frischhut and Greer (n 5) 338–339. 65  As a disclaimer, it should be mentioned that this paper did not analyse the case law of national courts in the different Member States. 66  Case C-­180/96 R United Kingdom v Commission ECLI:EU:C:1996:308, paras 91–92.

96   Markus Frischhut been developed by the CJEU, including in the context of communicable disease law, such as the (sometimes disputed) separation between risk assessment and risk management, individuals’ rights, the content of the precautionary principle, and the principle of proportionality. Likewise, at either side of the Atlantic, the executive branch plays a strong role, which we have seen throughout this chapter, for Europe, both at EU (European Commission, HSC, ECDC, etc.), as well as at national levels (health ministers). Beside a long-­standing scientific and technological cooperation between the European Union and the United States, both in the EU and within the US communicable disease control is a cross-­ cutting topic, thus very fragmented. In both jurisdictions, communicable disease law is often reactive to health crises. Consequently, in this context, we can identify more similarities between the European Union and the United States than one would initially expect. As Greer has put it, communicable disease control ‘oscillates between moments of extremely high salience and the near complete invisibility’.67 Still, authorities have to learn the lessons of past crises and act accordingly. Ebola, for instance, has proved that Member States have to improve the coordination of their national responses. Especially for both the executive (risk assessment vs. risk management) and the judiciary, it can be a challenge to import scientific information into the legal sphere. While in the past there has been a certain reluctance of courts to appoint experts,68 Alemanno and Garde have recently observed an emergence of an ‘evidence-­based judicial reflex’, according to which courts increasingly deal with knowledge stemming from other disciplines.69 The future of EU communicable disease law and policy can be expected still to be driven by the executive and the courts, with an increasing role of human rights due to the CFREU. Likewise, the European Union’s general or health values should play an increasing role.70 Following the afore-mentioned spill-over effect it remains to be seen, whether the EU will develop towards a European Health Union.

Acknowledgments The author would like to thank the editors for their valuable feedback, as well as the other authors of this section, especially Wendy E. Parmet, for excellent cooperation. This research (finalized in August 2019, before COVID-19) has been realized in the context of the Jean Monnet Chair on ‘EU Ethics & Values’, generously supported by the EU Erasmus+ programme.

67  Scott L Greer and Margitta Mätzke, ‘Bacteria Without Borders: Communicable Disease Politics in Europe’ (2012) J Health Politics Pol L 37(6), 887. 68  Eric Barbier De La Serre and Anne-­Lise Sibony, ‘Expert Evidence Before the EC Courts’ (2008) Common Market L Rev 45(4), 941, 961–962. 69  Alberto Alemanno and Amandine Garde, ‘The Emergence of EU Lifestyle Risk Regulation: New Trends in Evidence, Proportionality and Judicial Review’ in Hans-­W Micklitz and Takis Tridimas (eds), Risk and EU Law (Edward Elgar Publishing 2015) 166–167. 70  Markus Frischhut, The Ethical Spirit of EU Law (Springer International Publishing 2019) 34–36.

chapter 5

Com m u n ica bl e Dise ase L aw i n the U n ited States Wendy E. Parmet

1 Introduction The roots of contemporary health law in the United States lie with the colonial laws that responded to the horrific epidemics of smallpox, yellow fever, and other infectious diseases that regularly devastated the North American colonies. Long before modern medicine or any understanding of the etiology of infectious diseases, the colonists recognized that epidemics were public harms that demanded a legal response. Some of the laws they created called for prayer and fasting, but others would look more familiar to contemporary eyes familiar with disputes about vaccination, travel bans, and quarantine. For example, in 1622, the small settlement of East Hampton, Long Island, imposed what might have been the earliest recorded quarantine in Britain’s North American colonies.1 In the years that followed, colonial jurisdictions instituted numerous quarantines as well as sanitary laws to prevent the miasma that was thought to cause epidemics. After the Constitution was ratified in 1789, communicable disease control fell pri­ma­rily to the states. As the Supreme Court of the United States explained in 1824, “quarantine laws [and] health laws of every description” were among the legal powers that the Constitution left to the states.2 During the nineteenth century, states used these so-­ called police powers to implement a wide range of communicable disease laws. Many were challenged as harming out-­of-­state commercial interests. For the most part, the courts upheld the laws, underscoring the association between the states’ police power and communicable disease control,3 an association that continued after the Fourteenth 1 Wendy E. Parmet, Health Care and the Constitution, Public Health and the Role of the State in the Framing Era, 20 Hastings Cons’t L. Q. 267, 293 (1993). 2  Gibbons v. Ogden, 22 US (9 Wheat.) 1, 203 (1824). 3  Wendy E. Parmet, Populations, Public Health & the Law 82–83 (2009).

98   Wendy E. Parmet Amendment, with its promise of due process and equal protection of the law, was added to the Constitution.4 In the years that followed, courts continued to stress that states had broad powers to enact and enforce communicable disease laws, even when they adversely affected individuals or out-­of-­state commercial interests. Nevertheless, in the twentieth century, as the fear of infection declined, courts began to provide greater protections for individuals and vulnerable populations subjected to infectious disease laws. Moreover, the federal government began to play a more prominent and complex role in the control of infectious diseases. Still, many of the characteristics that marked communicable disease laws in the earlier period remain. First, as in the European Union, communicable disease laws are fragmented. The states play the primary role, and, despite efforts to promote uniformity between states through the promotion of “model” state laws, their laws vary widely. Moreover, states delegate significant authority to their local jurisdictions, whose laws also can differ dramatically. Second, infectious disease laws rely heavily on administrative bodies, generally state and local boards of health, but increasingly federal regulatory agencies. Indeed, to a large degree, communicable disease control relies on the work of local boards of health. Third, infectious disease laws are reactive; often they were enacted in response to specific epidemics.5 Thus the catastrophic cholera epidemics of the eighteenth century led to the establishment of standing boards of health and sanitary reforms.6 A hundred years later, the HIV epidemic helped create a greater appreciation of privacy, patient choice, and harm reduction. More recently, fears about bioterrorism, pandemic influenza, severe acute respiratory syndrome (SARS), and Ebola have left their imprint on American infectious disease law, as COVID-­19 will certainly do. Finally, US communicable disease laws reflect enduring tensions between protection of public health, scapegoating of vulnerable populations, and respect for individual rights. Indeed, perhaps to a greater extent than in the European Union, the story of American public health law is the story of the struggle to safeguard health while protecting individuals and vulnerable populations from those responses. This chapter, which was written before the COVID-­19 pandemic, explores these and other key features of American infectious disease law, many of which have endured ­during the COVID-­19 pandemic. The following section describes the allocation of authority between the states and federal government with respect to infectious disease control. The third section discusses the role that restraint on individual rights plays in infectious disease control and the limits that the Constitution and civil rights laws place on such restraints. The fourth section considers some of the specific tools that ­jurisdictions employ in response to infectious disease. The chapter concludes with a brief d ­ iscussion of the United States’ role in global public health. 4  Slaughter-­House Cases, 83 US 36, 62 (1873). 5  Lawrence O. Gostin, Scott Burris, & Zita Lazzarini, The Law and the Public’s Health: A Study of Infectious Disease Law in the United States, 99 Colum. L. Rev. 59, 101–118 (1999). 6  Charles E. Rosenberg, The Cholera Years: The United States in 1832, 1849, and 1866 (1987).

Communicable Disease Law in the United States   99

2  The Allocation of Authority: Public Health Federalism The traditional association between communicable disease control and the states’ police power continues to influence the allocation of legal authority with respect to infection control. Using their police powers, states remain the primary source of laws pertaining to disease reporting, isolation and quarantine, and vaccination, to name just a few areas. Moreover, states acting through statewide or local boards of health also regulate and inspect restaurants and food sellers, license healthcare providers, abate nuisances that breed disease-­spreading insects and rodents, and engage in a wide range of other legal interventions that seek to stem the spread of communicable diseases. In the nineteenth century, most constitutional challenges to state communicable disease laws were based on the so-­called dormant commerce clause, a court-­developed constitutional doctrine that limits states’ ability to regulate or interfere with interstate commerce, including the movement of goods across state lines. For example, in 1886, the Supreme Court reviewed a challenge to a Louisiana law that required vessels to pay a fee to state quarantine stations.7 The challengers claimed that the quarantine fee interfered with commerce and usurped federal authority. The Supreme Court disagreed, holding that, in the absence of a federal statute displacing their authority, the states had the power to impose quarantine laws even if they hampered interstate commerce.8 In the twentieth century, the Supreme Court’s approach to commerce clause challenges to state laws changed. Without reviewing the evolution of that doctrine, suffice it to say that today laws that discriminate against out-­of-­state commercial interests or try to impose state regulatory authority across states lines are generally unconstitutional.9 Thus in C. & A. Carbone, Inc. v. Town of Clarkstown, the Supreme Court found a municipal ordinance that required all solid waste within the town to be deposited at a local transfer station to unconstitutionally discriminate against out-­of-­state commerce.10 In contrast, health laws that neither discriminate against out-­of-­state interests nor have an extraterritorial reach are subject to the so-­called Pike balancing test, under which the Court asks whether the burden that the state law imposes on commerce is outweighed by its local benefits.11 In recent decades, the federal government has played an increasingly important and complex role in the assessment of, response to, and control of infectious disease. Under the Constitution, the national government can only enact laws that fall within one of its enumerated grants of authority. In most cases, these enumerated powers are broad enough to sustain federal communicable disease laws. Under Article I, for example, 7  Morgan’s Louisiana & T.R. & S.S. Co. v. Board of Health of State of Louisiana, 118 US 455 (1886). 8  Id. at 465–466. 9  Parmet, Populations, Public Health and the Law, at 82–90. 10  511 US 383, 390 (1994). 11  Pike v. Bruce Church, Inc. 397 US 137, 142 (1970).

100   Wendy E. Parmet Congress can regulate persons and goods in interstate or international commerce, as well as activities that substantially affect such commerce.12 Congress can also enact laws that are “necessary and proper for carrying into Execution” its other powers. As a result of these powers, Congress can establish the Food and Drug Administration (FDA) and require that pharmaceutical companies receive approval before marketing antibiotics, vaccines, and other pharmaceuticals and biologics that may be used to prevent or treat communicable and other diseases. Likewise, using its commerce power, Congress can establish international or interstate quarantines. Importantly, under the supremacy clause, federal laws that are enacted pursuant to the commerce power (as well as other sources of federal authority) supersede conflicting state laws. Moreover, Congress can use its Article I powers to block states from legislating in an area altogether, as it has done with vaccine safety.13 In addition to its regulatory authority under the commerce and necessary and proper clauses, Congress also has authority under Article I to tax and spend for the common defense and general welfare. This has proved to be an even more useful source of federal authority for infection control than the commerce power. For example, through the National Institutes of Health and the Centers for Disease Control and Prevention (CDC), Congress supports research into communicable disease prevention and treatment. Congress has also established the Strategic National Stockpile, which stores vaccines, pharmaceuticals and medical equipment that may be needed in the event of a pandemic or bioterrorist incident.14 Moreover, because the federal government has far greater financial resources than the states, federal agencies such as the CDC are able to provide expert guidance and support to the states. Despite the breadth of its Article I powers, there are implied structural limits to Congressional reach. Perhaps most importantly, Congress cannot use its power to regulate commerce to compel or commandeer state officials.15 Nor can Congress compel individuals to act, as opposed to refrain from acting.16 Hence, Congress probably could not use its commerce powers to require that individuals be vaccinated, nor could it compel states to mandate vaccination. Still, in many cases, Congress can achieve its policy goals by attaching conditions, as long as they are not unduly coercive, to the many federal grants, large and small, that support state health-­related activities. For example, in 1990, Congress required states that received certain federal HIV-­related grants to either criminalize the intentional transmission of HIV or demonstrate that their laws provide for prosecution in such cases.17 Likewise, although Congress almost certainly lacks the authority to establish nationwide vaccine mandates, it probably could condition the receipt of federal public health funds on states enacting such a mandate. To date, however, Congress has not chosen to do so. Rather, perhaps out of respect for tradition and 12  US v. Lopez, 514 US 549, 558–559 (1995). The case law focuses on interstate commerce, but the Constitution also provides Congress with authority to regulate international commerce. 13  Bruesewitz v. Wyeth, 562 US 223, 247 (2011). 14  42 USC. 247d-­6b. 15  US v. Printz, 521 US 898, 935 (1997). 16  Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 US 519, 557 (2012). 17  See Max D. Siegel, Lessons from a Plague, 4 William & Mary Pol’y Rev. 292, 302–303 (2013).

Communicable Disease Law in the United States   101 comity, Congress is generally content to leave states as the primary sources of regulatory interventions with respect to communicable diseases.

3  Communicable Disease Law and Individual Rights The central challenge of US communicable disease law is often viewed as the need to reconcile the rights of individuals against the public’s health. Without question, this challenge is critical. In the absence of vaccines or effective treatments, nonpharmaceutical interventions (NPIs), such as quarantines, travel bans, or stay-­at-­home orders which restrict limit individual liberty, may sometimes offer the best or only option. Nevertheless, it is wrong to assume that infection control only requires limitations on human rights. To the contrary, respecting human rights and providing supports to populations at risk may often help to stop the spread of an infectious disease.18 The extent to which infections are met with coercive or supportive laws depends not only on how a disease is transmitted and the availability of effective pharmaceutical interventions, but also on the social and political context through which epidemics are understood. The development of microbiology in the late nineteenth century led to what Professors Gostin, Burris, and Lazzarini have called the microbial model, in which disease was “seen as a product of microbial infection (or exposure to toxic substances of some other sort), and the job of public health is to identify the pathogen and to eliminate or contain it.”19 Under this model, individuals who are infected or are thought to be at risk of infection are viewed as vectors of disease. The job of infectious control law, therefore, becomes to reduce the risk posed by such individuals by controlling their movements or behavior. Hence liberty is seen as detrimental to public health, and coercive public health laws, including quarantine, travel bans, or mandatory vaccination, take center stage in the public health toolkit. Following the epidemiological transition in the twentieth century, noncommunicable diseases began to play a more significant role in determining population health. As they did so, public health began to focus more on the impact of individual behaviors, such as smoking or lack of exercise. With the advent of the AIDS epidemic in the 1980s, this behavioral model was transferred to infectious diseases, as public health experts warned against risky behaviors such as needle-­sharing or unprotected sex.20 This behavioral model supports divergent responses. Some are supportive; for example, laws may provide education or the means (such as condoms) for people to reduce the risk of contagion. Sometimes, however, the legal responses copy those suggested by 18  E.g., Jonathan Mann, Medicine and Public Health, Ethics and Human Rights, 27 The Hastings Center Rep. 6, 9–10 (2012). 19  Gostin, Burris, & Lazzarini, supra note 5, at 70. 20  Id. at 71–72.

102   Wendy E. Parmet the microbial model and seek to control individual behavior through isolation or quarantine orders or laws that criminalize behaviors that are deemed dangerous.21 A third model, which Gostin, Burris, and Lazzarini call the ecological model, responds to the recognition of the important role that social and environmental forces—the ­so-­called social determinants of health—play in establishing a community’s risks of infection. To some degree, this model echoes the pre-­microbial view that saw sanitation as the key to infection control. But the ecological model considers a far wider range of social factors. For example, during the AIDS epidemic, many public health experts came to appreciate that the infection was fueled in part by discrimination and stigmatization, and they accepted that the denial of human rights undermined efforts to stem the ­epidemic.22 This led to the view that civil rights laws and protections for individual ­liberty could abet, rather than impede, HIV prevention. It also led to the adoption of many policies that are more supportive than coercive, including increased access to care and services, educational campaigns, and harm mitigation strategies such as needle exchange programs.23 However, it is important to note that such approaches were implemented alongside more punitive policies, including HIV criminalization laws and bans on access to clean syringes.24 These divergent and complex understandings of the role of individual rights in infection control are reflected, but not to equal degrees, in court decisions regarding coercive communicable disease control laws. The seminal case is the Supreme Court’s 1905 decision in Jacobson v. Massachusetts,25 which concerned a Cambridge, Massachusetts, law requiring vaccination of all residents who had not been vaccinated since 1897. A Swedish immigrant, the Reverend Henning Jacobson, refused to be vaccinated and was convicted and ordered to pay a $5 fine. Jacobson appealed his conviction to the US Supreme Court, arguing that Cambridge had violated his rights under the Fourteenth Amendment. The Supreme Court disagreed. Writing for the Court, Justice John Marshall Harlan explained that the individual liberty secured by the Constitution is never absolute. Rather, “[t]here are manifold restraints to which every person is necessarily subject for the common good.”26 He added, “Upon the principle of self-­defense, of paramount necessity, a community has the right to protect itself as against an epidemic of disease which threatens the safety of its members.”27

21  Trevor Hoppe, Punishing Disease: HIV and the Criminalization of Sickness 101–132 (2018). 22 Mann, supra note 18, at 9–10. 23  See generally, Ronald Bayer, The Continuing Tensions Between Individual Rights and Public Health, 8 EMBO Reports 1099 (Dec. 12, 2007), http://embor.embopress.org/content/8/12/1099; see also, National Academies Press, HIV/AIDS Care in the Third Decade: Opportunities and Challenges in the Changing Epidemic, in Public Financing and Delivery of HIV/AIDS Care: Securing the Legacy of Ryan White (2005). 24 Hoppe, supra note 21, at 101–131. 25  See Jacobson v. Massachusetts, 197 US 11 (1905). 26  Id. at 26. 27  Id. at 27.

Communicable Disease Law in the United States   103 Although Jacobson affirmed Cambridge’s vaccine mandate and stressed the state’s right to limit liberty to control infectious diseases, it also marked an important turning point. For the first time, the high Court suggested that the Fourteenth Amendment provided some protections for individuals whose rights were limited by infectious disease laws. As the Court explained, “it might be that an acknowledged power of a local community to protect itself against an epidemic threatening the safety of all might be exercised in particular circumstances and in reference to particular persons in such an arbitrary, unreasonable manner, or might go so far beyond what was reasonably required for the safety of the public, as to authorize or compel the courts to interfere for the protection of such persons.”28 Under what circumstances that would occur, the Court did not say. In the years since, Jacobson has become something of a Rorschach test. Some see it as affirming that states have almost unlimited power to limit liberty in order to control an infectious disease. Others see it as establishing significant constitutional restraints on state power. For example, Professor Lawrence O. Gostin argues that Jacobson permits coercive public health powers “only if they are exercised in conformity with five standards, which I call public health necessity, reasonable means, proportionality, harm avoidance, and fairness.”29 Given the Court’s language, Gostin may read more specificity and firmer boundaries into the Court’s opinion than the Court intended. Nevertheless, Jacobson unquestionably established both the fact that states have broad power to restrain liberty to protect public health and that those powers are subject to some constitutional limits. Moreover, although Jacobson remains the starting point for discussions regarding individual liberty and coercive communicable disease laws, the courts’ approach to such questions has evolved in the subsequent century. Without recounting the full story of the evolution of constitutional rights since 1905, a few points are worth noting. First, courts today give far more consideration to the type and nature of the individual right affected by the state’s law. In Jacobson’s era, the Court had not yet concluded that the rights protected by the Bill of Rights, such as freedom of speech or religion, applied against the states. Today the rights guaranteed by the Bill of Rights, as well as certain recognized fundamental liberties protected by the Equal Protection and Due Process clauses of the Fifth and Fourteenth Amendments, trigger heightened judicial review. For many rights, the courts undertake what is known as “strict scrutiny,” in which they ask whether the restrictions are narrowly tailored to support a compelling state interest. As will be discussed further in the section on isolation and quarantine, this sometimes requires the state to show that the restriction on liberty is the least restrictive means for achieving its compelling interest. Although courts almost always accept without debate that public health is a compelling state interest, the hurdle of showing that a law is the least restrictive means of achieving it can be high. On the other hand, even with this rigorous test, courts do not require that states institute the type of supportive approaches 28  Id. at 28. 29  Lawrence O. Gostin, Public Health Law: Power, Duty, Restraint 126 (2d ed. 2008).

104   Wendy E. Parmet (such as the provision of healthcare, education, or suitable housing) that the ecological model might suggest. Even in its most stringent form, the least restrictive analysis requirement remains mired in the individualist paradigm of the microbial and behavioral models. The courts’ approach to laws that discriminate against vulnerable populations has also evolved substantially in the past hundred years. When epidemics strike, communities frequently blame and turn against minority and other vulnerable populations. The history of epidemics in the United States is rife with discrimination, stigmatization, and the disproportionate use of highly coercive public health laws, including quarantines and criminalization, against vulnerable groups. For example, a few years before Jacobson, San Francisco responded to a bubonic plague outbreak by requiring all people of Asian background to be inoculated with an experimental vaccine before leaving the city.30 After that law was declared unconstitutional, the city imposed a racially gerrymandered quarantine that applied to the homes of Chinese-­American residents but not to whites. That law was also ruled unconstitutional.31 In the years since, vulnerable populations have continued to face discrimination, stigma, and the disparate brunt of coercive public health laws.32 For example, studies suggest that racial minorities have been disproportionately prosecuted for HIV-­related offenses.33 Homeless people, immigrants, and people with mental illness have also been blamed for and detained to stop the spread of tuberculosis (TB).34 The ecological model suggests that such discriminatory measures are often counterproductive. Nevertheless, legal remedies remain limited. Under current doctrine, racial classifications are treated as suspect and are subject to strict scrutiny, under which the state must prove that that the classification is necessary to further a compelling state interest.35 Gender-­based classifications are also subject to heightened scrutiny, with the state having to show an “exceedingly persuasive” justification.36 The Supreme Court has not to date afforded classifications on the basis of sexual orientation similar treatment. However, it has struck down such classifications in cases in which it has found that the state law was motivated by anti-­gay animus.37 Despite these advances in the Court’s equal protection jurisprudence, constitutional challenges to discriminatory state communicable disease laws continue to face formidable 30  Wong Wai v. Williamson, 103 F. 1, 4 (C.C.N.D. Cal. 1900). 31  Jew Ho v. Williamson, 103 F. 10, 26 (C.C.N.D. Cal. 1900). 32 Hoppe, supra note 21, at 17–42. 33  Angela Perone, From Punitive to Proactive: An Alternative Approach for Responding to HIV Criminalization that Departs from Penalizing Marginalized Communities, 24 Hastings Women’s L.J. 363, 379–383 (2013). 34  Patricia Illingworth & Wendy E. Parmet, The Health of Newcomers: Immigration, Health Policy and the Case for Global Solidarity 61–72 (2017). 35  Adarand Constructors, Inc. v. Pena, 515 US 200, 227 (1995). 36  US v. Virginia, 518 US 515, 531 (1996). 37  See Obergefell v. Hodges, 576 U.S. 644 (2015)(striking down ban on same-­sex marriage): Roemer v. Evans, 517 US 620, 623 (1996)(striking down state law barring protections for gays and lesbians noting that it appeared to have been motivated by animus).

Communicable Disease Law in the United States   105 odds. Most importantly, in order to prevail, the plaintiff generally needs to show that the discrimination was either on the face of the statute or regulation or was otherwise intentional.38 This can be very hard to do as jurisdictions are seldom as blatant and open about their discrimination as was San Francisco in 1900. For example, even after the Supreme Court held that discrimination on the basis of gender triggered heightened judicial review, courts rejected Equal Protection challenges to communicable disease laws that targeted prostitutes on the theory that such laws do not necessarily discriminate on the basis of sex.39 Statutes barring discrimination on the basis of disability also play a significant role today in limiting the state’s power with respect to certain chronic, infectious diseases. In a case brought under the Rehabilitation Act of 1973, the Supreme Court ruled that a school board could not dismiss a teacher with TB unless it could demonstrate that she posed a significant risk to the health or safety of others.40 Borrowing a standard proffered by the American Medical Association, the Court explained that the determination of substantial risk should be based on “reasonable medical judgements given the state of medical knowledge, about a) the nature of the risk (how the disease is transmitted), b) the duration of the risk (how long is the carrier infectious), c) the severity of the risk (what is the potential harm to third parties), and d) the probabilities the disease will be transmitted and will cause varying degrees of harm.”41 This “direct threat” test has been applied in cases reviewing the segregation of HIV-­positive prisoners as well as the isolation of TB patients.42 Finally, in the years since Jacobson, the courts have more clearly distinguished and delineated the procedural protections required when states infringe upon individual rights. The notion that individuals have a right to a fair process, including notice, the right to a hearing, and the right to counsel, when their liberties are restrained has long been a central component of Anglo-­American law and is embedded within the Due Process clauses of the Fifth and Fourteenth Amendments to the Constitution. And, from the earliest period, courts have overridden coercive public health orders for procedural inadequacies. However, in 1905, the Supreme Court had not yet articulated a distinct doctrine of procedural due process. In the past century it has, holding that “what process is due” depends on the nature of the individual interest, the risk of an erroneous deprivation, and the probable value of additional safeguards.43 As discussed later, some courts have applied that test in cases of isolation and quarantine.44

38  Washington v. Davis, 426 US 229 (1976). 39  Reynolds v. McNichols, 488 F.2d at 1378, 1383 (10th Cir. 1973)(upholding law permitting detention and forced treatment of prostitutes); People v. Adams, 597 N.E. 2d 574, 585 (Ill. 1992)(upholding law requiring mandatory HIV testing of prostitutes). 40  School Bd. of Nassau Cty. v. Arline, 480 US 273, 287 (1987). 41 Id. 42  See, e.g., Henderson v. Thomas, 913 F. Supp. 2d 1267, 1289 (M.D. Ala. 2012); J.S. v. City of Newark, 279 N.J. Super. 178, 196–197 (1993). 43  Mathews v. Eldridge, 424 US 319, 333–347 (1976). 44  See infra text accompanying notes 73–79.

106   Wendy E. Parmet Taken together, these constitutional and statutory developments set important but limited restraints on coercive communicable disease laws. Nevertheless, as in Jacobson’s time, courts continue to grant public health authorities with broad deference when they restrain liberty in order to stop the spread of an epidemic. Perhaps more importantly, the Supreme Court continues to view the Constitution as a charter of negative liberties,45 one that neither compels that states protect public health nor requires them to act to create the conditions in which communities can be healthy. In contrast to the laws in the nations of the European Union, neither the Constitution nor laws of the United States guarantee that all Americans have access to healthcare when needed to diagnose, prevent or treat infections. Hence although the states and federal governments support insurance for many Americans and have implemented numerous initiatives from vaccine clinics to needle exchange programs designed to support communities and mitigate the risk of infection, US communicable disease remains mired in the challenge between government power and negative liberty. As a result, when the next significant infectious epidemic strikes, coercive measures—and the limited legal restraints placed on them—are apt to continue to dominate the debate.

4  Risk Management US jurisdictions utilize a wide range of legal tools—or powers—to detect and respond to infectious diseases. Space precludes a discussion of all of these tools, but several are worth discussing. Before doing so, however, it is important to note that many laws that are not ostensibly “about” infectious disease play an important role in preventing or responding to such diseases. Consider, the Affordable Care Act.46 That federal law is usually thought of as a health law rather than an infectious disease law. But by expanding access to healthcare, studies suggest that it has led to increased use of preventive services, including vaccinations.47 It also established the Public Health Prevention Fund, which provided federal funds to support local health agencies, including in their infectious disease control efforts.48 More broadly, the ecological model posits that many laws that affect the social determinants of health, positively or adversely, may impact infectious disease. In contrast, core infectious disease laws often reflect the microbial and behavioral models and tend to rely heavily on controlling individuals and populations, thus raising many of the human rights and constitutional law issues discussed earlier.

45  DeShaney v. Winnebago Ct’y Dep’t Soc, Serv, 489 US 189, 195 (1989). 46  Patient Protection and Affordable Care Act, Pub. L. No. 111–148, 124 Stat. 119 (2010). 47  Xuesong Han et al., Has Recommended Preventive Service Use Increased After Elimination of Cost-­Sharing as Part of the Affordable Care Act in the United States, 78 Preventive Med. 85, 85–91 (2015). 48  Patient Protection and Affordable Care Act § 4002.

Communicable Disease Law in the United States   107

4.1  Surveillance and Reporting The detection of infectious disease is key to its containment. In the United States, surveillance occurs primarily at the state and local levels. Although states increasingly rely on syndromic surveillance, which detects outbreaks using Big Data sources, all states have laws requiring healthcare providers and other specified parties (e.g., diagnostic laboratories or schools) to report cases of listed diseases to authorities. In deciding which diseases must be reported, states rely heavily on the guidance offered by the CDC and the Council for State and Territorial Epidemiologists.49 The CDC also requires state and local health departments to report certain diseases to it. The CDC also encourages state and local agencies to report other diseases voluntarily. For example, the CDC operates the National Outbreak Reporting System (NORS) which allows state and local health departments to report outbreaks of food- and waterborne diseases.50 States may also mandate that individuals be screened or tested for an infectious disease. For example, states have traditionally required that individuals be tested for syphilis prior to attaining a marriage license or TB prior to attending school or working in healthcare facilities. Under the Fourth Amendment, mandatory screening or testing constitutes a search and must be “reasonable.” However, in most infectious disease cases, in which the information obtained by the testing was not used for law enforcement purposes, the courts have upheld the mandates based on what is known as the “special needs” exception to the Fourth Amendment.51 Surveillance, screening, and reporting all raise issues of individual privacy. This became especially salient in the 1990s, when there were major debates as to whether states should require mandatory HIV screening of newborns or named reporting of HIV cases. Many felt that newborn screening was essential to assuring that infants receive prophylaxis before they become infected,52 as well as necessary to attain a complete and accurate picture of the scope of the epidemic. In turn, many AIDS advocates argued that mandatory screening violated a woman’s autonomy and that named reporting would deter individuals from being tested.53 These debates, which died down once HIV became a treatable condition, highlight the ways in which concerns about stigma and discrimination can conflict with health officials’ need to gather information. As in other areas of communicable disease law, no singular, national law governs all issues that relate to the confidentiality of patient information. In Whalen v. Roe, the Supreme Court rejected a due process challenge to a state law that required the mandated reporting of prescription drug use, at least where the state had made reasonable efforts 49  James G. Hodge, Public Health Law in a Nutshell 246 (3d ed. 2018). 50  Centers for Disease Control and Prevention, National Outbreak Reporting System (NORS), https://www.cdc.gov/nors/about.html (last visited May 2, 2018). 51  Lawrence O. Gostin & Lindsay F. Wiley, Public Health Law: Power, Duty Restraint 371 (3d ed. 2016). 52  Amitai Etzioni, HIV Testing of Infants: Privacy and Public Health, 17 Health Affairs 170 (1998). 53  Ronald Bayer & Amy Fairchild, The Limits of Privacy: Surveillance and the Control of Disease, 10 Health Care Analysis 19 (2002).

108   Wendy E. Parmet to protect confidentiality.54 In addition, the Health Insurance Portability and Accountability Act (HIPAA) privacy rules, which regulate the disclosure of personal health information by healthcare providers and insurers, specifically exempts state public health reporting laws.55 Nevertheless, most states have laws protecting the confidentiality of health information gathered by public health agencies, and most agencies take their obligation to respect patient confidentiality seriously. In the current era, flaws in computer security and the sharing of health information via social media probably pose a far greater risk to confidentiality than do state and local health departments.

4.2  Risk Communication Strictly speaking, risk communication raises issues more of public health practice than of law. Nevertheless, because risk communication plays an essential role in infectious disease control, it warrants brief consideration. Despite all of the legal tools public health officials have at their disposal, infection control depends to a large degree on the public’s faith in the information that it receives and its willingness to comply with the advice given. Officials may order home quarantines or mandate vaccinations, but in a large, complex society, force of law can never be sufficient. Many individuals will not follow the orders that are issued unless they trust the advice they receive and are confident that officials are neither overstating the dangers nor unfairly targeting them. In order to gain that trust and compliance without inciting panic or alarm, officials must communicate health-­related information in a clear, accessible, and culturally sensitive manner. They need to state what is known and what isn’t known, work with affected communities, and listen and respond to their concerns. Doing this can be challenging, even under the best of circumstances. With new epidemics, the information may be lacking, and the information available may change rapidly, which can create confusion or policies that cause more harm than good. For example, after several soldiers contracted a form of influenza in 1976 that appeared to resemble the strain responsible for the 1918 pandemic, health officials warned then-­ President Gerald Ford that there was a risk of a widespread pandemic. In response, he ordered an unprecedented campaign that urged all Americans to be vaccinated. More than 40 million were, yet the pandemic never materialized. Instead, news quickly spread of the potential association between the vaccine and cases of Guillain-­Barré syndrome.56 The fragmentation of governmental authority and the intervention of political leadership can exacerbate the challenge of responding with less than complete information. For example, during the 2014 Ebola outbreak, the Department of Defense undermined CDC’s efforts to reassure the public that widespread quarantines weren’t needed by 54  Whalen v. Roe, 429 US 589 (1977). 55  45 C.F.R. 164.512(b)(1)(i) (2018). 56  See Richard E. Neustadt & Harvey Fineberg, The Epidemic that Never Was: Policy-­ Making and the Swine Flu Affairs (1985).

Communicable Disease Law in the United States   109 ordering that all military personal leaving West Africa be quarantined even if they had no contact with Ebola patients.57 Likewise, the New York City Department of Health and Mental Hygiene undercut its own efforts to reassure people that they couldn’t contract Ebola by riding the subway when it decontaminated a bowling alley that had been used by a physician who had contracted Ebola.58 Such inconsistent messaging about how a disease can be transmitted can confuse the public and lead to panic and even demands for unnecessary and ineffective responses. Risk communication may be especially challenging in the current environment, which is characterized by high levels of political polarization and distrust of institutions, including the media. Reports such as those in 2017 that the White House told the CDC not to use seven words, including “science-­based” and “evidence-­based,” in their budget request may magnify the perception that public health has become politicized.59 This may prove problematic the next time public health officials have to communicate to the public about a dangerous new epidemic.

4.3 Vaccination Vaccines have long played a critical role in preventing infectious diseases. Thanks to vaccination, smallpox has been eradicated in nature, and many other communicable diseases, such as polio and diphtheria, are now exceptionally rare. Despite this record, many people follow Henning Jacobson’s footsteps in rejecting vaccination. Some do so simply because they no longer see the need to be vaccinated for diseases that are no ­longer common. Others cite religious objections or worries about vaccine safety. Since the 1990s, safety concerns have crystallized around the disproved theory that vaccines cause autism.60 Whatever the reason, when individuals reject vaccination for themselves or their children, they reduce a vaccine’s ability to achieve “herd immunity,” which occurs when a sufficient proportion of a population (which varies by pathogen) has become immune to a disease so that it can no longer be transmitted within the population. Herd immunity can be especially critical for individuals who cannot be vaccinated either because they are too young or have medical contraindications.

57  Robert Gatter, Ebola, Quarantine, and Flawed CDC Policy, 23 U. Miami Bus. L. Rev. 375, 397 (2015). 58  Marc Santora, Doctor in New York City Is Sick with Ebola, N.Y. Times, Oct. 23, 2014, https:// www.nytimes.com/2014/10/24/nyregion/craig-­spencer-­is-­tested-­for-­ebola-­virus-­at-­bellevue-­hospital-­ in-­new-­york-­city.html; Hunter Stuart, Brooklyn Bowling Alley “The Gutter” to Reopen After Ebola Cleanup, Huffington Post, Oct. 24, 2014, https://www.huffingtonpost.com/2014/10/24/the-­gutter-­ ebola_n_6041678.html. 59  Julia Belluz, CDC Banned from Using 7 Words, Including “Science-­based” In Budget Documents, Vox, Dec. 16, 2017, https://www.vox.com/2017/12/16/16784498/ cdc-­seven-­words-­science-­transgender-­fetus 60  Leila Barraza, Daniel G. Orenstein, & Doug Campos-­Outcalt, Denialism and Its Adverse Effect on Public Health, 53 Jurimetrics 307, 320–322 (2013).

110   Wendy E. Parmet The possibility of herd immunity underscores that vaccinations are a public good that benefit populations, not only the individuals who are vaccinated. It also offers another motivation for vaccine resistance: free-­riding. As long as herd immunity exists, individuals have little incentive to be vaccinated. They will be spared the infection simply because others are vaccinated. But if too many people elect that choice, herd immunity will dissipate, and the infection will return. Although vaccine rates remain relatively high in the United States in the general, many communities have elevated rates of re­sist­ ance and have experienced outbreaks in recent years of vaccine-­preventable infections such as measles and pertussis.61 Since the nineteenth century, states have responded to vaccine resistance and free-­ riding by requiring children to be vaccinated to attend school. Today all states have such laws, but they differ significantly as to what vaccines are covered, what actions trigger the mandate, and what exemptions exist. All states allow for medical exemptions, and 47 permit religious exemptions.62 Eighteen states permit parents to exempt their children from vaccination based solely on personal belief.63 In 2015, following a measles outbreak in Disneyworld, California became the third state to abolish all nonmedical exemptions. Other states in recent years have kept their exemptions but have tried to make them more difficult to attain, for example by requiring notarization.64 Although vaccine mandates remain very controversial, the courts have continued to affirm their constitutionality, relying on Jacobson and the 1922 Supreme Court case of Zucht v. King, which upholds a Texas vaccine mandate.65 The courts have also rejected claims that the lack of a religious exemption violates the First Amendment’s right to the free exercise of religion.66 Paradoxically, vaccine resisters have had more success challenging laws with religious exemptions either on the grounds that the exemptions discriminate against non-­believers or because the states’ application of the exemption either violates free exercise (by failing to recognize some religious claims) or impermissibly establishes religion (by recognizing other claims).67 While vaccine mandates remain the province of the states, the federal government regulates vaccine safety. Like other pharmaceuticals, vaccines are licensed by the FDA. A full discussion of FDA regulation is beyond the scope of this chapter. However, it is worth noting that, in 1986, in response to concerns about vaccine safety and shortages of 61  Varun K. Phadke et al., Association Between Vaccine Refusal and Vaccine-­Preventable Diseases in the United States, 315 JAMA 1149, 1149 (2016). 62  Michell M. Mello, David M. Studdert, & Wendy E. Parmet, Perspective- Shifting Vaccination Politics: The End of Personal-­Belief Exemptions in California, 373 N. Eng. J. Med. 785, 785 (2015). 63  National Conference of State Legislators, States with Religious and Philosophical Exemptions From School Immunization Requirements, http://www.ncsl.org/research/health/school-­immunization-­ exemption-­state-­laws.aspx. (last visited May 2, 2018). 64  The Network for Public Health Law, Compulsory Immunization Requirements 4–5, https://www. networkforphl.org/_asset/2b51ff/Elements-­of-­School-­Immunization-­Waiver.pdf. (last visited May 2, 2018). 65  Zucht v. King, 260 US 174 (1922). 66  See, e.g., Phillips. v. City of New York, 775 F.3d 538 (2d Cir. 2015). 67  See, e.g., Brown v. Stone, 378 So. 2d 218, 222–224 (Miss. 1979).

Communicable Disease Law in the United States   111 certain childhood vaccines, Congress passed the National Childhood Vaccine Injury Act (NCVIA).68 It established a passive surveillance system, the Vaccine Adverse Events Reporting System (VAERs) and a no-­fault compensation program for vaccine-­related injuries. Under the National Vaccine Injury Compensation program, claimants who have injuries that are well-­recognized and listed on a table as vaccine-­related can be compensated from a fund financed by an excise tax on vaccines without having to prove either fault or causation.69 Claimants with other injuries may seek relief before an administrative tribunal known as the “Vaccine Court.” Although that court’s decisions are reviewable by a federal court, claimants lose their right to bring product liability claims in state court.70 The combination of mandates on individuals and immunities for manufacturers may have helped to establish stability in the vaccine market and relatively high rates of vaccination. It has not, however, quieted resistance. Rather, the fact that the claims in the Vaccine Court are defended by the federal government, rather than the manufacturers, and that claimants are denied a right to a jury trial may have helped to fuel distrust and even conspiracy theories.

4.4  Isolation, Quarantine, and Travel Bans Jurisdictions employ many social distancing measures to keep people apart and reduce contagion. Among these measures are quarantine, isolation, and travel bans. Although the term “quarantine” has been used in American law to refer to all of these practices, public health practitioners today define “quarantine” as the separation of an individual or group of persons who are thought to have been exposed to a communicable disease.71 “Isolation,” in turn, refers to the separation of an individual or group who is thought to be infectious.72 Travel bans bar individuals or groups from entering or leaving a jurisdiction. Other social distancing measures include curfews or bans on public gatherings. Traditionally, isolation and quarantine have primarily been imposed by the states. Isolation is usually considered less subject to abuse than quarantine because the individual is ill and will generally want and benefit from isolation and treatment in a hospital. In the case of TB, however, isolation has been controversial because states frequently detain individuals even when they are not contagious if they fail to take their medications, which may cause them to develop and transmit a drug-­resistant form of the disease. During an outbreak of drug-­resistant TB in the 1990s, New York City and several other jurisdictions dramatically increased the use of isolation orders.73 Although some 68  42 USC. § 300aa-­1 et seq. (2018). 69  See 42 USC. § 300aa-­13(a)(1)(A). 70  Bruesewitz v. Wyeth, 562 US 223, 231–232, 243 (2011). 71  Centers for Disease Control and Prevention, Quarantine and Isolation, https://www.cdc.gov/ quarantine/quarantineisolation.html (last visited May 2, 2018). 72 Id. 73  Barron H Lerner, Catching Patients: Tuberculosis and Detention in the 1990s; 115 Chest 236, 236 (1999).

112   Wendy E. Parmet have credited the increased use of isolation with helping to stem the epidemic, critics worried that isolation orders would fall disproportionately on immigrants and other minorities.74 In the past few decades, several courts have examined the rights of individuals who are subjected to such orders. For example, in 1979, the West Virginia Supreme Court ruled that “involuntary commitment for having communicable TB impinges on the right to “ ‘liberty, full and complete liberty’ no less than involuntary commitment for being mentally ill,”75 and that the procedural due process rights applicable in cases of civil commitment, including notice, a hearing, a right to counsel, and the requirement that the state prove its case by clear and convincing evidence, must apply in cases of isolation for TB.76 A few years later, a New Jersey trial court held that detention for TB was justified only when it was the least restrictive alternative and that individuals who were detained must be provided with notice, a hearing, periodic review, and counsel.77 Directly observed therapy, in which a healthcare worker or someone else observes the TB patient taking his or her medications is a commonly used less-­restrictive alternative. Historically, many quarantines were instituted at ports, on navigable waters, or at state borders. These quarantines impeded commerce and met significant resistance from commercial interests, as was evident in the Morgan’s Louisiana & T.R. & S.S. Co. case discussed earlier. Quarantines of individuals have been relatively rare in recent decades, and the Supreme Court has not considered a quarantine case in more than a hundred years. In the absence of decided case law, many commentators had presumed that the standards articulated by the courts in the TB cases would apply to quarantine; however, cases outbreak in federal district courts in New Jersey and Connecticut arising from the 2014 Ebola cast some doubt on that conclusion. In both cases, the courts ruled that officials who had imposed quarantines on individuals who returned to the United States from West Africa were immune from civil judgment because the constitutional rights of individuals who were quarantined were not clearly established.78 The Connecticut case remains under appeal. The Ebola quarantines, as well as TB isolation orders, demonstrate the persistence of the microbial and behavioral models of communicable disease, which view individuals as the danger, and the problems created by the lack of contemporary case law. When frightening new outbreaks arise, health officials frequently face political pressure to 74  Illingworth & Parmet, supra 34, 58–60, 210. 75  Greene v. Edwards, 263 S.E.2d 661, 663 (W. Va. 980). Strictly speaking, Greene was not a constitutional case. The court interpreted the state’s TB control statute to require the same procedural protections that the court had previously demanded as a constitutional manner in the case of commitment for mental health. 76  Id. at 329 (quoting State ex rel Hawkes v. Lazaro, 202 S.E.2d 109, 112 (W. Va. 1974)). 77  J.S. v. City of Newark, 279 N.J. Super. 178 (1993). See also Best v. St. Vincents Hosp, 2003 WL 21518829 (S.D. N.Y. 2003), aff ’d in part, vacated in part as to one of two hospitals on evidentiary and procedural grounds; City of New York v. Antoinette R., 630 N.Y.S. 2d 1008 (N.Y. Sup Ct. 1995); City of New York v. Doe, 614 N.Y.S.2d 8 (N.Y. App. Div. 1994). 78  Hickox v. Christie, 205 F. Supp. 3d 579, 594 (D.N.J. 2016); Liberian Cmty. Ass’n v. Malloy, No. 3:16-­CV-­00201(AVC) 2017 WL 4897048 at *9 (D. Conn. Mar. 30, 2017).

Communicable Disease Law in the United States   113 institute “tough” individually-­focused actions, such as quarantine, even though evidence suggests that these measures are rarely effective in the contemporary context and pose significant burdens on human rights. The lack of clear constitutional guideposts means that there are few barriers preventing officials from succumbing to those pressures or ensuring that the rights of affected individuals are preserved. For example, US jurisdictions do not provide for compensation for individuals who are quarantined, despite the fact that other nations, such as Canada, came to realize the need for such policies during the 2003 SARS outbreak.79 Likewise, even when courts demand that that isolation or quarantine be the least restrictive alternative, they apply that phrase narrowly and focus only on alternatives that may reduce the risk posed by the individual who is detained, rather than the possibility of alternative, population-­wide policies that might be more effective in reducing the overall risk of infection. For example, no court has required a state seeking a quarantine order to demonstrate that it ensures sick pay, even though doing so might be more effective for preventing the spread of disease in the community as a whole.80 In fact, as of 2018, only ten states plus the District of Columbia have a law requiring that employers provide their workers with paid sick leave.81 The persistent allure of isolation and quarantine is also evident in the quarantine regulations that were promulgated by the federal government in 2017. The federal government has long had the power to bar individuals with a communicable disease from entering the United States82 and to quarantine individuals within the country to stop the spread a communicable disease across state lines.83 In recent decades, these powers have been seldom used. Although the CDC maintains quarantine stations at twenty ports of entry and border crossings, it has issued only one domestic quarantine order in the past sixty years.84 However, following the 2014 Ebola outbreak, the CDC published new quarantine regulations in 2017.85 These establish new requirements for air carriers to report passengers who may potentially have an infectious disease. The regulations also imagine a far more robust role for the federal government in enforcing domestic isolation and quarantine orders as well as interstate travel bans. They permit the CDC to authorize the apprehension, medical examination, quarantine, isolation, or conditional release of any individual for the purpose of preventing the introduction,

79  See Christine Coughlin, Public Health Policy: Revisiting the Need for a Compensation System for Quarantine, 7 Wake For. J. L. & Pol’y 415, 433 (2017). 80  Mark A. Rothstein & Meghan K. Talbott, Encouraging Compliance with Quarantine: A Proposal to Provide Job Security and Income Replacement, 97 Am. J. Pub. H. 549, 549 (2007). 81  National Conference of State Legislaturees, Paid Sick Leave (May 29, 2018), http://www.ncsl.org/ research/labor-­and-­employment/paid-­sick-­leave.aspx. 82  78 P.L. 410, 58 Stat. 682, 703 (1944). 83 Id. 84  Centers for Disease Control and Prevention, US Quarantine Stations, https://www.cdc.gov/ quarantine/quarantine-­stations-­us.html (last visited May 2, 2018); Mike Stobbe, US Quarantines Man with TB; Extraordinary Step Follows His Refusal to Avoid Flying, The Record, May 30, 2007, at A08. 85  82 Fed. Reg. 6890 (Jan. 19, 2017); 42 C.F.R. Parts 70 and 71 (2018).

114   Wendy E. Parmet transmission, and spread of quarantinable communicable diseases, as specified by Executive Order, based upon a finding that: (1) The individual is reasonably believed to be infected with a quarantinable communicable disease in a qualifying stage and is moving or about to move from a State into another State; or (2) The individual is reasonably believed to be infected with a quarantinable communicable disease in a qualifying stage and constitutes a probable source of infection to other individuals who may be moving from a State into another State.86

Although the regulations provide several rounds of internal administrative review and require the CDC to consider if the detention is the least restrictive alternative, they do not provide for an independent review, nor do they require the CDC to lift a quarantine or travel ban if it is not the least restrictive alternative. For the moment, the regulations only apply to the few diseases that the President has determined to be quarantinable: cholera, diphtheria, infectious TB, plague, smallpox, yellow fever, viral hemorrhagic fevers, and SARS.87 No law, however, prevents the President from adding to the list.

4.5  Emergency Preparedness Since the start of the AIDS epidemic, the United States has faced a plethora of emerging infectious diseases (EIDs), including multidrug-­resistant TB, Lyme disease, Zika, and Ebola, to name just a few.88 According to scientists, globalization, environmental degradation, modern agricultural practices, growing antimicrobial resistance (created by the overuse of antibiotics and antiviral medications), and the erosion of the public health infrastructure in many nations facilitate emergence and leave the world vulnerable to potentially catastrophic EIDs. Concern about this possibility has sparked the development of public health preparedness, which aims to improve the planning for, detection of, and response to public health emergencies. After 9/11 and the anthrax attacks on the US mail, the preparedness agenda focused to a large degree on terrorism (especially bioterrorism). The devastation of New Orleans following Hurricane Katrina led to further expansion of preparedness efforts and a more widespread embrace of what is known as the all-­hazards approach, which emphasizes preparedness for any type of disaster that might befall a community. Still, worries about EIDs, from Ebola to pandemic influenza, remain central to preparedness efforts.

86  42 C.F.R. § 70.6 (2018). 87  Centers for Disease Control and Prevention, Legal Authorities for Isolation and Quarantine, https://www.cdc.gov/quarantine/aboutlawsregulationsquarantineisolation.html (last visited May 2, 2018). 88  Centers for Disease Control and Prevention, Emerging Infectious Diseases, https://wwwnc.cdc. gov/eid/page/background-­goals (last visited May 2, 2018).

Communicable Disease Law in the United States   115 Emergency preparedness has influenced (and infiltrated) communicable disease law in many ways. First, it has increased the role of the federal government, as evidenced by the 2017 federal quarantine regulations as well as the plethora of federal statutes that have been enacted in recent years. These statutes have increased oversight of laboratories that worked with potentially dangerous pathogens,89 incentivized the development and use of vaccines and other biomedical countermeasures,90 and enhanced the federal government’s role in planning for and responding to an emergency.91 The focus on emergency preparedness has also reinforced the reliance on coercive public health powers. That was especially evident in the Model State Emergency Health Powers Act (MSEHPA) that was drafted by Professor Gostin at the bequest of the CDC shortly after 9/11. The MSEHPA, which purported to serve as a model for state emergency laws, sought to enhance the states’ power to respond to an emergency by seizing property, mandating medical examinations, and imposing isolation and quarantine.92 It was also evident when President Bush stated that the military might be needed during an influenza pandemic to enforce quarantines of American cities.93 Another notable feature of preparedness efforts has been the enhanced support for research and development of new biomedical responses, accompanied by the relaxation of regulations to provide significant liability protections for manufacturers and distributors of new vaccines and pharmaceuticals that may be used during an emergency. In 2004, Congress passed the Project BioShield Act, which permitted the Secretary of the Department of Health and Human Services to declare a public health emergency and authorize the emergency use of drugs, medical devices, and vaccines that have not yet been approved for use.94 The following year, Congress passed the Public Readiness and Emergency Preparedness Act, which provides extremely broad liability protections for manufacturers and distributors of countermeasures upon the Secretary’s declaration of a public health emergency.95 Public health preparedness efforts have also led to the reallocation of public health priorities, from common sources of morbidity and mortality (including common infections) toward emergency planning. In the past two decades, Congress has provided state

89  Antiterrorism and Effective Death Penalty Act of 1996; Pub. L. No. 104–132, 110 Stat. 1214, 1284–1285 (1996); Uniting and Strengthening America by Providing Appropriate Tools Required to Intercept and Obstruct Terrorism Act of 2001, Pub. L. No. 107–156, § 817, 115 Stat. 272, 385–386 (2001); Pub. L. No. 107–188, title II, 116 Stat. 594 (June 2002). 90  42 USC. § 262(a). 91  Benjamin Berkman et al. Assessing the Impact of Federal Law on Public Health Preparedness, 4 St. Louis J. Health L. & Pol’y 154, 159 (2010). 92 Lawrence O. Gostin, The Model State Emergency Health Powers Act 6 (2010), https://www. aapsonline.org/legis/msehpa2.pdf. 93  Associated Press, US May Use Military Quarantine to Contain Flu, NBC News (Oct. 7, 2005), http://www.nbcnews.com/id/9589897/ns/health-­infectious_diseases/t/us-­may-­use-­military-­quarantine-­ contain-­flu/#.Ws0W24jwZPY. 94  Pub. L. Nos. 108–276, 118 Stat. 835 (2004). 95  Pub. L. Nos. 109–417, 120 Stat. 2831 (2005).

116   Wendy E. Parmet and local governments with significant funding for preparedness.96 However, in recent years, federal support for public health has fallen, and local public health agencies have been forced to use general funds to respond to new infectious threats, such as Zika.97 Moreover, despite the broad agreement that preparedness efforts must be broad-­based, resources continue to be disease-­specific, and efforts to fight one infection compete with those to fight another. Thus CDC used money initially allocated to respond to Ebola to respond to Zika. In short, despite two decades of preparedness planning, infectious disease responses remain wedded to the older microbial and behavioral models. Far less attention has been paid to broader social determinants, including income inequality, poor education, and housing insecurity, that create vulnerabilities to communicable diseases, both emerging and endemic.98 Moreover, surge capacity remains limited, and the healthcare system remains fragmented, strained, and, in many ways, ill-­equipped to respond to a true emergency.

4.6  Global Health The US federal government is the world’s largest funder for global health, spending $10.4 billion in FYI 2017 on global health initiatives.99 Despite that relative generosity, the US approach to global health reflects many of the same tensions and weaknesses that are evident in domestic communicable disease policies. First, just as domestic communicable disease law has developed in reaction to different epidemics, US support for global health has resulted from concerns relating to specific diseases and threats. For example, almost half of US support for global health goes to support HIV, largely as a result of the US President’s Emergency Plan for AIDS Relief (PEPFAR).100 Likewise, in 2014 and 2015, the United States spent more than $2 billion fighting Ebola in West Africa.101 Once Ebola faded from the news, money allocated to it

96  Crystal R. Watson et al., Public Health Preparedness Funding: Key Programs and Trends from 2001 to 2017, AJPH Practice (Sept. 11, 2017), https://ajph.aphapublications.org/doi/full/10.2105/ AJPH.2017.303963. 97  A Funding Crisis for Public Health and Safety: State-­by-­State and Federal Public Health Funding Facts and Recommendations, Philanthropy News Digest (Mar. 18, 2018), http:// philanthropynewsdigest.org/connections/a-­funding-­crisis-­for-­public-­health-­and-­safety-­state-­by-­ state-­and-­federal-­public-­health-­funding-­facts-­and-­recommendations. 98  For a further discussion of the role of social determinants of health and policies that may address them, see Parmet,“Social Determinants of Health in the United States,” in this volume. 99  The US Government and Global Health, Kaiser Family Foundation (Aug. 8, 2017), https:// www.kff.org/global-­health-­policy/fact-­sheet/the-­u-­s-­government-­and-­global-­health/. 100  The US President’s Emergency Plan for AIDS Relief (PEPFAR), Kaiser Family Foundation (Dec. 19, 2017), https://www.kff.org/global-­health-­policy/fact-­sheet/the-­u-­s-­presidents-­emergency-­plan-­for/. 101  Jennifer Kates et al., The US Response to Ebola: Status of the FY 2005 Emergency Ebola Appropriation, Kaiser Family Foundation, Dec. 11, 2015), https://www.kff.org/global-­health-­policy/ issue-­brief/the-­u-­s-­response-­to-­ebola-­status-­of-­the-­fy2015-­emergency-­ebola-­appropriation/.

Communicable Disease Law in the United States   117 was reassigned to Zika.102 This disease-­specific approach impedes support for the development of a public health infrastructure that can respond to any number of diseases.103 Second, although many in the United States support funding for global public health because they believe that wealthy countries have an ethical obligation to protect the health of individuals in low-­income countries, the US involvement in global health has largely been spurred by more self-­interested concerns. Since the 1990s, infectious disease experts have repeatedly emphasized that germs know no borders and that epidemics left unchecked in one corner of the world can travel and wreak havoc in another. Thus support for global health has been presented as critical to safeguarding domestic health and even national security.104 Critics contend that this approach places disproportionate emphasis on communicable diseases that threaten developed countries to the neglect of more endemic diseases, including noncommunicable diseases, that extract a greater toll on local populations.105 The intermingling of support for global health with domestic security and preparedness also helps to explain a common feature of debates within the United States to overseas epidemics. While many argue that diseases such as Ebola or SARS should be “fought” overseas, others have called for travel bans and limits on migration, arguing that the dangerous diseases can and should be kept out of the United States.106 Thus while some see infectious diseases in low-­income countries as a reason for the United States to become more involved in and supportive of global health, others point to the same threats as a justification for limiting migration and increased isolationism. At the same time, many of the tensions between commercial interests and health that have been evident throughout the history of US communicable disease law play out today in the international arena. For example, while supporting global health initiatives, the United States has also taken strong positions regarding intellectual property rights, which critics contend has helped delay access to essential medications in low-­income countries.107 Despite self-­ interest, US support for global health always remains precarious. Although the 2018 Omnibus Budget Reconciliation Act appropriated 34% more funds for global health than the President requested, his proposal to slash funding evinces the uncertainty of future efforts in the absence of a prominent crisis.108 International aid in general remains a common target of the political right in the United States and is especially vulnerable now given the Trump administration’s “America First” stance. 102  Matiangai Sileaf, Ebola Does Not Fall from the Sky: Structural Violence and International Responsibility, 51 Vanderbilt J. Transnational L. 1, 67 (2018). 103  Id. at 75. 104  Id. at 76. 105 Id. 106  Illingworth & Parmet, supra 34, 41–44. 107  Intellectual Property and Access to Medicine, OXFAM (last visited May 4, 2018), https://policy-­ practice.oxfamamerica.org/work/trade/intellectual-­property-­and-­access-­to-­medicine/. 108  Kaiser Foundation Budget Summary Analyzes Global Health Aspects of FY 2017 Omnibus Released by Congress, March 2018, Kaiser Family Foundation (Mar. 22, 2018), https://www.kff.org/ news-­summary/kaiser-­family-­foundation-­budget-­summary-­analyzes-­global-­health-­aspects-­of-­fy-­ 2018-­omnibus-­released-­by-­congress/.

118   Wendy E. Parmet

5 Conclusion Our understanding of communicable disease has advanced dramatically since East Hampton, Long Island, quarantined ships in 1622. In the almost four hundred years since, scientists have learned far more about the causes of infection and have developed a pharmacopeia of vaccines and biomedical responses. Society has also become far more complex, as has public health law, rendering hazardous all short reviews such as this. Still, several enduring traits and challenges remain. Chief among them are the continued reactivity of communicable disease law to the latest epidemic, a centuries-­old pattern that has led not only to messy and inconsistent laws, but also to the neglect of broader, proactive strategies. Likewise, US law continues to display an excessive reliance on coercive laws that treat those who are infected as the vectors of disease to the detriment of interventions that aid those who are ill and enhance communities’ ability to resist infection. In this era of heightened nationalism and anti-­immigrant fervor, the temptation to respond to the next epidemic with shame, blame, and punishment may be especially strong. Whether US communicable disease law will be able to withstand that tide and protect both the public’s health and human rights remains to be seen.

chapter 6

US L aw R el ati ng to Noncom m u n ica ble Diseases Wendy E. Parmet

Noncommunicable diseases (NCDs) are the leading cause of death in the United States. According to the Centers for Disease Control and Prevention (CDC), heart disease is the number one cause of death in the United States, killing more than 630,000 Americans a year; cancer comes in second, killing another 595,000 Americans.1 More than 79,000 Americans die each year from diabetes, and more than 30 million Americans, or 9.4% of the population, live with that disease.2 More than 44,000 Americans die every year from suicide,3 almost half of which are by firearms.4 The incidence of suicide, drug overdoses, and obesity among adults is also rising.5 1  National Center for Health Statistics, Leading Causes of Death, Centers for Disease Control and prevention (Mar. 17, 2017), https://www.cdc.gov/nchs/fastats/leading-­causes-­of-­death.htm. Not all cases of cancer or heart disease are NCDs. Several infectious agents can increase an individual’s risk of cancer. National Cancer Institute, Infectious Agents (updated March 4, 2019), https://www.cancer.gov/ about-­cancer/causes-­prevention/risk/infectious-­agents. Infection may also play a role in heart disease. Daniel M. Musher, Micahel S. Abers, & Vincente F. Corrales-­Medina, Acute Infection and Myocardial Infarction, 380 N. Eng. J. Med. 171 (2019). For purposes of this discussion, however, cancer will be treated as noncommunicable because it is generally viewed as such, and most of the legal interventions that seek to prevent cancer generally focus on behavorial and environmental causes, rather than infection. 2  Centers for Disease Control and Prevention, National Diabetes Statistics Report (2017). 3  National Center for Health Statistics, Leading Causes of Death, Centers for Disease Control and Prevention (Mar. 17, 2017), https://www.cdc.gov/nchs/fastats/leading-­causes-­of-­death.htm. 4  National Center for Health Statistics, Suicide and Self-­Inflicted Injury, Centers for Disease Control and Prevention (May 30, 2016), https://www.cdc.gov/nchs/fastats/suicide.htm. 5  Suicide Rates Rising Across the U.S., Centers for Disease Control and Prevention (June 7, 2018), https://www.cdc.gov/media/releases/2018/p0329-­drug-­overdose-­deaths.html; Matt Richtel & Andrew Jacobs, American Adults Just Keep Getting Fatter, N.Y. Times (Mar. 23, 2018), https://www. nytimes.com/2018/03/23/health/obesity-­us-­adults.html; U.S. Drug Overdose Deaths Continue to Rise; Increase Fueled by Synthetic Opioids, Centers for Disease Control and Prevention (Mar. 29, 2018), https://www.cdc.gov/media/releases/2018/p0329-­drug-­overdose-­deaths.html.

120   Wendy E. Parmet Despite the heavy toll extracted by NCDs, the United States lacks a singular, binding strategy to address them as a group or even as separate diseases. Nor is there a clear consensus in the nation that law should be used to address those NCDS that are associated with consumer products, such as firearms and fast food. Despite widespread support for many public health laws relating to some NCDs, especially those that regulate the sale and use of cigarettes, many NCDs are widely viewed as resulting from individual choices. Hence laws that seek to protect Americans from such diseases are often thought to be paternalistic limitations on individual liberty. This individualistic framing significantly influences the legal tools that are employed. This chapter examines US laws that seek to reduce the prevalence of NCDs by targeting the sale, marketing, or use of the products with which they are associated. (For a discussion of the role of law relating to the social determinants of health, which also affect NCDs, see the Chapter on Social Determinants in the US). The chapter begins by reviewing some of the broad debates that have helped to shape the US response to NCDs; namely concerns about paternalism, individual responsibility, and the appropriate role for law regarding noncommunicable health threats. Next, the chapter examines the allocation of authority between US jurisdictions to address NCDs. The chapter then provides an overview of several key US laws and policies that relate to tobacco, unhealthy foods, firearms, and opioids. That survey is followed by a more in-­depth examination of the roles that the regulation of commercial speech and the First Amendment’s protection for speech play in efforts to reduce the threat of NCDs. The chapter concludes with a brief discussion of possible future developments.

1  Individualism, Paternalism, and the Role of Law The roots of US public health law lie in communicable disease control. Long before the discovery of the germ theory, states and localities turned to law to try to stem the tide of repeated epidemics of cholera, smallpox, yellow fever, and other infectious diseases.6 Although particular applications of the laws that were employed were sometimes controversial, few people questioned that government should try to protect the public from such diseases. By the middle of the twentieth century, the United States and other developed countries had experienced an epidemiological transition. Communicable diseases no longer posed the greatest threats to human health. Instead, heart disease, cancer, diabetes, and injuries were the primary killers. In response, public health officials and policy-­makers sought to employ a different set of legal tools to reduce morbidity and mortality. In the

6  See Chapter on Communicable Disease Law in the United States.

US Law Relating to Noncommunicable Diseases   121 US, these new tools proved to be far more controversial than were those that targeted infectious diseases.7 In part, the controversy stemmed from the fact that laws relating to NCDs are often adverse to the economic interests of industries that manufacture or sell products, such as cigarettes or guns, that are associated with NCDs. These industries tend to wield significant political influence.8 Some of these industries, such as the tobacco industry, have also poured substantial money into informational campaigns casting doubt on the risks associated with their products.9 Partially due to industries’ successful framing of public debates about their products, NCDs have frequently been viewed in the United States as the result of the type of private choices that a liberal society ought to respect. For many years, public health experts largely accepted this framing, treating NCDs as the result of risky individual behaviors. For example, in 1993, the Journal of the American Medical Association published an influential article by J. William McGinnis and William H. Foege entitled, “Actual Causes of Death in the United States.” In their paper, McGinnis and Foege stated that the most important contributors to death in the United States were tobacco, diet and activity patterns, alcohol, microbial agents, and illicit drug use.10 While acknowledging poverty’s role in creating these causes, McGinnis and Foege stressed “[t]here can be no illusions about the difficulty of the challenges in changing the impact these factors have on health status. Of those identified here, the three leading causes of death—tobacco, diet and activity patterns, and alcohol—are all rooted in behavioral choices.”11 This framing of the etiology of NCDs as the result of individual behavioral choices continues to be widely accepted in the United States and helps to explain why laws relating to NCDs are frequently viewed as inappropriately paternalistic.12 After all, if injuries and diseases result from individual choices, efforts to prevent them will often attempt in one way or another to restrict, for their own good, the choices available to individuals. Such paternalism is highly frowned upon in a society such as the United States that prioritizes individual liberty. To a significant degree, US laws that relate to NCDs have been shaped in response to the paternalism critique. For example, paternalism is generally considered to be 7  Several scholars have made the argument that laws relating to NCDs are inherently less appropriate than those relating to communicable diseases. See, e.g., Richard Epstein, Let the Shoemaker Stick to His Last: A Defense of the “Old” Public Health Law, 46 Perspectives Biol. Med. S199 (2003); Mark A. Hall, The Scope and Limits of Public Health Law, 46 Perspectives Biol. Med. S199 (2003). 8  For a discussion of the role that industry has played in causing NCDs and combatting public health interventions, see Nicholas Freudenberg, Lethal But Legal: Corporations, Consumption, and Protecting Public Health (2014). 9  See Erik M. Conway & Naomi Oreskes, Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warning 36–65, 136–168 (2010). 10  J. William McGinnis & William H. Foege, Actual Causes of Death in the United States, J. Am. Med. Assoc. 2207, 2007 (1993). 11  Id. at 2211. 12  David Adam Friedman, Public Health Regulation and the Limits of Paternalism, 46 Conn. L. Rev. 1687, 1691 (2014).

122   Wendy E. Parmet appropriate in the case of minors or others who are not mature or competent. Thus many US laws about NCDs seek to protect children or young adults (who are not considered fully mature). For example, US laws bar the sale of cigarette and e-­cigarettes to minors and the sale of alcohol to individuals under age 21. Likewise, efforts to ban billboard displays for products associated with NCDs, such as cigars, have been defended by public health advocates as necessary to protect minors. (However, it should be noted that the Supreme Court has rejected the idea that speech can be banned simply to protect minors.13) In addition, regulations tend to be stricter for products, including tobacco, alcohol, and opioids, that are addictive on the theory that once individuals are addicted they lose their ability to make free choices. As a result, laws regulating such products are less apt to be viewed as paternalistic limitations on individual liberty than are laws that relate to nonaddictive products. The desire to avoid the paternalism charge also helps to explain the popularity of laws that require warning labels or other disclosures. Although the efficacy of such measures depends significantly on the nature of the warning, product, and risk,14 with graphic warnings often more effective than warnings that simply employ plain text,15 disclosure laws are frequently defended as ways of providing adults with the information they can use to make their own choices. Thus calorie labels on menus, as is required by the Affordable Care Act in chain restaurants, can be viewed as more respectful of individual choice than laws that limit portion sizes or regulate the content of ingredients. Similarly, defenders of “sin taxes” and other measures that seek to influence but not dictate individual behaviors argue that such approaches merely “nudge” individuals to make the choices they would make if they were fully informed and fully rational.16 According to Thaler and Sunstein, such laws may be paternalistic, but they nevertheless adhere to libertarian values.17 The heavy role that individual responsibility plays in framing the etiology of NCDs also helps to explain why public health advocates often seek to portray the industries that manufacture and sell products that are related to NCDs as “bad actors” who are responsible for the health risks associated with their products. In a sense, this approach accepts the idea that disease is the result of individual responsibility but seeks to move 13  Brown v. Entertainment Merchants Ass’n, 564 U.S. 786, 796 (2011). 14  For a critique of the overuse of warning labels, see Lisa A. Robinson, W. Kip Viscusi, & Richard J. Zeckhauser, Efficient Warnings, Not “Wolf or Puppy” Warnings, Harvard Kennedy School of Gov’t, Faculty Research Working Series, Paper No. 16–033, 2016), https://papers.ssrn.com/sol3/papers. cfm?abstract_id=2839311. 15  Research on tobacco warnings suggests that graphic images are more effective in reducing consumption than are textual warnings. See, e.g., M. Bansal-­Travers, et al., The Impact of Cigarette Pack Design, Descriptors, and Warning Labels on Risk Perception in the U.S., 40 Am. J. Prev. Med. 674 (2011); D. Hammond et al., Text and Graphic Warnings on Cigarette Packages: Findings from the International Tobacco Control Four Country Study, 32 Am. J. Prev. Med. 202 (2007). 16 Wendy E. Parmet, Paternalism, Self-­Governance, and Public Health: The Case of and E-­Cigarettes, 70 U. Miami L. Rev. 879, 945–951 (2016). 17  Richard H. Thaler & Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness 4–6 (2008); Cass R. Sunstein, The Storrs Lectures: Behavioral Economics and Paternalism, 122 Yale L.J. 1826, 1865–1866 (2013).

US Law Relating to Noncommunicable Diseases   123 the responsibility from consumers to the commercial entities that profit from ­products that are associated with NCDs. For example, for many years, individuals who sued tobacco compqanies fared poorly due to an assumption of risk defense that emphasized the responsibility of the consumer. More recently, tobacco control advocates have largely succeeded in framing the problem as one of corporate, rather than individual, responsibility, opening the door not only to success in the courtroom, but also to a greater willingness on the public’s part to accept significant regulations. Many states and other jurisdictions are now trying to employ a similar strategy by suing pharmaceutical companies for their marketing of prescription opioids.18 Indeed, the prominent role that civil litigation plays in US NCD law19 can be understood as related to the fact that NCDs continue to be viewed primarily through an individualistic lens, which attributes responsibility to some actor, individual, or industry, as opposed to understanding the problem as one that derives from deeper social and environmental forces and warrants broader regulatory interventions.20

2  The Allocation of Authority Within the United States As in other areas of public health law, the legal authority in the United States to determine whether or how to regulate to prevent NCDs is divided between different jurisdictions and different branches of government. These different entities often adopt divergent and sometimes conflicting approaches, even with regard to the same disease. At the outset, it’s worth noting that international treaties and conventions only have the force of law in the United States if they have been ratified by the US Senate. Although the Senate has ratified the World Trade Agreement and several other key trade agreements that implicate products associated with NCDs, it has not ratified the Framework Convention on Tobacco Control (FCTC) or several other important international agreements that relate to health, such as the Convention on Social, Economic, and Cultural Rights.21 In general, international law plays a far lesser role guiding US domestic public health law and policy within the United States than it does in Europe. Within the United States, public health law is traditionally viewed as falling primarily within the province of the states.22 Unlike the federal government, states have general 18  See infra text accompanying notes 127–130. 19  See infra text accompanying notes 60–63, 82–84, 120–131. 20  See Chapter on Social Determinants. 21  WHO Framework Convention on Tobacco Control, United Nations Treaty Collection (May 21, 2003), https://treaties.un.org/pages/ViewDetails.aspx?src=TREATY&mtdsg_no=IX-­4& chapter=9&clang=_en 22  For a fuller discussion of the allocation of authority pertaining to public health, see Chapter on Communicable Disease Law in the U.S.

124   Wendy E. Parmet police powers that permit them to enact laws to safeguard the health and safety of their residents. Using these powers, states and their localities have utilized a wide variety of laws targeting NCDs, from excise taxes on tobacco and alcohol to mandated warning labels, such as a San Francisco law requiring health warnings on advertisements for sugary beverages.23 States also regulate the sale or use of products associated with NCDs, including firearms, tobacco, and motor vehicles, and they support a wide range of initiatives designed to inform people about and reduce the risks of NCDs. State power is especially broad with respect to alcohol due to the 21st Amendment to the Constitution, which ended Prohibition and grants states the power to prohibit the importation or use of alcohol. Despite the traditional primacy of the states over matters of public health, federal law plays a leading role with respect to regulating the manufacturing, marketing, and sale of products that are associated with NCDs, such as cigarettes or processed foods. There are two primary sources of constitutional authority for federal laws relating to NCDs. First, Congress’ constitutional power to tax and spend for the general welfare provides a powerful public health law tool that enables the federal government to leverage its considerable financial resources on behalf of public health. For example, using this power, Congress has established Medicare, Medicaid, and other federal healthcare programs that provide treatment and screening for NCDs, and it has financed biomedical research and public health grants that provide critical financial support for state, local, and private-­sector initiatives designed to address NCDs. In addition, as long as it is not unduly coercive,24 Congress can use its authority to tax and spend to require states to enact specific laws or policies as a condition of receiving federal funds. For example, in South Dakota v. Dole, the Supreme Court upheld a federal statute that required states to set a legal drinking age of 21 in order to receive federal highway funds.25 The federal government can also address NCDs by using to its so-­called commerce power, which grants Congress the authority to regulate international and interstate commerce. Although the Supreme Court in recent decades has placed some important limits on Congress’ commerce power, it remains very broad, permitting Congress to regulate all goods and services that travel across national or state lines as well as all activities that substantially affect interstate commerce.26 Using this power, Congress regulates tobacco products, alcohol, motor vehicles, and even the fats used in processed foods. The commerce power also sets some limits on the states’ police powers. This can happen in several ways. First, the Supreme Court has interpreted the commerce clause as creating an implied prohibition on state laws that disfavor out-­of-­state commerce. In the twentieth century, this so-­called negative or dormant commerce clause doctrine was 23  American Bev. Ass’n v. City and County of San Francisco, 916 F.3d 749 (9th Cir. 2019)(en banc). For a further discussion of the law pertaining to mandated warnings, see infra text accompanying notes 133–146. 24  National Federation of Independent Business v. Sebelius, 567 U.S. 519, 576–581 (2012). 25  South Dakota v. Dole, 483 U.S. 203, 211–212 (1987). 26  567 U.S. at 549.

US Law Relating to Noncommunicable Diseases   125 applied in numerous cases concerning state regulation of food products.27 Today, the Court permits states to enact health laws that implicate out-­of-­state commercial interests as long as they do not clearly favor in-­state economic interests.28 Still, state laws that discriminate against out-­of-­state commercial interests or seek to protect in-­state interests will usually be found in unconstitutional. For example, despite the existence of the 21st Amendment, which secures state power to regulate the manufacture, transportation, or sale of alcohol, the Supreme Court has struck down several state laws that prohibited out-­ of-­ state wineries, but not in-­ state wineries, from selling directly to consumers.29 Congress can also use its commerce power to explicitly “preempt” or block the states from regulating in areas that implicate interstate or international commerce. For example, the federal Cigarette Warning Act, which requires cigarette packages to include health warning labels, prevents states from requiring additional health warnings.30 Another federal statute, the Protection of Lawful Commerce in Arms Act, bars most state civil litigation against manufacturers or sellers of firearms.31 The Nutrition Labeling and Education Act of 1990 preempts state laws that impose labeling requirements on food products that are not identical to those established by federal law.32 Federal preemption of state law may also be implied. This can occur when courts determine that “it is impossible for a private party to comply with both state and federal requirements.”33 Relying on this reasoning, the Supreme Court has ruled that claims brought against generic drug companies for defective designs were subject to implied preemption because generic manufacturers could not comply with both federal regulations and the requirements of state tort law.34 Just as federal law may preempt state (and local) laws, state laws may override local public health ordinances and regulations. Traditionally, many states have given their cities and counties wide-­ranging authority over matters relating to public health,35 and some of the most innovative laws addressing NCDs, such as indoor smoking laws, were pioneered by local governments.36 Some of these innovations, such as bans on the sale of 27  E.g., Dean Milk Co. v. City of Madison, 340 U.S. 349, 351 (1951); H.P. Hood & Sons, Inc. v. Du Mond, 336 U.S. 525, 575 (1949). 28  City of Philadelphia v. New Jersey, 437 U.S. 617, 631–632 (1978). 29  Granholm v. Heald, 544 U.S. 460, 493 (2005). 30  Cippolone v. Liggett Group, Inc., 505 U.S. 504, 514 (1992) (construing 15 U.S.C. A. 1334(b)). For a further discussion of cigarette warning laws, see infra text accompanying notes 53–76. 31  Protection of Lawful Commerce in Arms Act of 2005, Pub. L. No. 109–192, 119 Stat. 2095 (2005). 32  Turek v. General Mills, 662 F.3d 423, 425–427 (2011) (discussing 21 U.S.C. Sec. 343-­1(a)(5)). 33  Mutual Pharmaceutical Co. v. Bartlett, 570 U.S. 472, 480 (2013) (citing English v. General Elec. Co., 496 U.S. 72, 79 (1990)). 34 Id. 35  Most states do not give localities so-­called home rule authority, meaning that local governments must rely on specific grants of power. Nevertheless, local boards of health have traditionally been granted broad powers. For a discussion of local authority regarding public health, see Lawrence O. Gostin & Lindsay A. Wiley, Public Health Law: Power, Duty, Restraint 179–184 (3d ed. 2016). 36 Paul A. Diller, Why Do Cities Innovate in Public Health? Implications of Scale and Structure, 91 Wash. U. L. Rev. 1219, 12224–12242 (2014).

126   Wendy E. Parmet tobacco products to youth or the banning of trans-­fats, have been copied by the federal government.37 However, many other local legal innovations have been met with strong political resistance, often spurred by lobbying from regulated industries. In response, some states have curbed the power of local governments to enact specific laws that relate to NCDs. Most states, for example, have laws limiting local governments’ authority to regulate firearms.38 Some states have laws limiting local governments’ ability to regulate food or beverage sellers.39 In one notable case, an Ohio appellate court struck down that state’s attempt to block the City of Cleveland from banning trans-­fats.40 The court found that the state legislature exceeded its authority to limit the city’s home rule power because the preemption provision was not part of a comprehensive statewide regulatory regime. Regardless of the jurisdiction, in an ideal world, the questions of whether and how to regulate the manufacture, marketing, or use of any product associated with an NCD would be based at least initially on the best evidence available regarding an intervention’s efficacy.41 However, in many cases, the efficacy of particular interventions, especially new and innovative ones, is lacking or incomplete. Moreover, the choice of which legal tools to use, or whether to use any at all, is usually entrusted in the first instance to political decision-­makers who are not generally bound by any particular standard of evidence and whose decisions are as likely to reflect ideology, the influence of powerful lobbies, and popular perceptions as to be based on a careful analysis of the scientific evidence. Although legislatures are not required to base their decisions on scientific evidence, when a statute is found to implicate a protected constitutional right, such as the right to free speech or privacy, the jurisdiction defending it will have the burden of presenting evidence demonstrating why the statute is needed to protect health. When some strongly protected rights, such as the right to free speech, are implicated, the evidentiary standard may be so high as to be virtually impossible for the jurisdiction to meet. For example, no amount of evidence regarding the dangers of smoking or the impact of misinformation on smoking rates would likely suffice to justify a law prohibiting newspapers from publishing articles claiming that tobacco was good for one’s health.

37  US Food and Drug Administration, Final Determination Regarding Partially Hydrogenated Oils (Removing Trans Fat), https://www.fda.gov/food/ingredientspackaginglabeling/ foodadditivesingredients/ucm449162.htm (federal regulations relating to trans-­fats). For a discussion of federal laws regulating the sale of tobacco products to youth, see infra text accompanying notes 56–64. 38  See Gostin & Wiley, supra note 35, at 181. 39 Id. See also Grace Thompson, Note—How Commonsense Consumption Acts Are Preventing “Big Food” Litigation, 41 Seattle Univ. L. Rev. 695, 696 (2018). 40  City of Cleveland v. State, 989 N.E.2d 1072, 1085 (Ohio Ct. App. 2013). 41  Nuffield Council on Bioethics, Public Health: Ethical Issues 31, http://nuffieldbioethics. org/report/public-­health-­2/policy-­process-­practice (2007), last visited June 24, 2019. Of course, other factors, including the laws’ cost and impact on protected rights and important values, should also be considered.

US Law Relating to Noncommunicable Diseases   127 In contrast to legislatures, administrative agencies, such as the Food and Drug Administration (FDA) or local boards of health, are expected to rely on scientific evidence in reaching their regulatory decisions. The precise standard of evidence to which they are held depends on the statute from which they receive their authority as well as on the administrative law in their jurisdiction. Under the federal Administrative Procedure Act, courts are expected to uphold agency regulations that are promulgated via notice and comment rule-­making, which is the main mechanism for establishing health regulations as long as the regulations are not “arbitrary and capricious, an abuse of discretion or otherwise not in accordance with law.”42 Although generally deferential, this stand­ ard requires agencies to show that they have considered and reviewed the relevant scientific evidence in promulgating their regulations. Other provisions of law require federal agencies in most instances to undertake a cost-­benefit analysis of proposed regulations.43 As with statutes, regulatory decisions are held to a higher standard when they infringe on protected constitutional rights.

3  Tobacco Control Tobacco control has rightly been called one of the top public health achievements in the United States.44 In 1955, more than 50% of American men older than 18 and almost 25% of American women older than age 18 smoked cigarettes.45 By 2016, only 17.5% of adult males and 13.5% of adult females in the United States smoked cigarettes.46 Youth smoking rates have also declined dramatically. Between 2011 and 2017, the rate of high school students who reported smoking within a 30-­day period fell from 15.8% to 7.6%.47 Among middle schoolers, the decline was similar, falling from 4.3% to 2.1%.48 Along with falling rates of smoking has come a significant drop in several smoking-­associated NCDs. Most notably, the death rate from lung cancer fell 23% from its peak in 1991 to 2012.49 These gains have been achieved through the application of a range of federal, state, and local supply- and demand-­side legal tools. A full discussion of these efforts is 42  5 U.S.C. § 706(2)(A)(2012). 43 Maeve P. Carey, Cost-­Benefit and Other Analysis Requirements in the Rulemaking Process, Congressional Research Service, Dec. 9, 2014, https://fas.org/sgp/crs/misc/R41974.pdf 44  Centers for Disease Control and Prevention, Ten Great Public Health Achievements—United States, 2001–2010, 60 Morb Mort Wkly Rpt 619, 619–621 (2011), https://www.cdc.gov/mmwr/preview/ mmwrhtml/mm6019a5.htm?s_cid=fb2423 45  Centers for Disease Control and Prevention, Achievements in Public Health, 1900–1999: Tobacco Use—United States, 1900–1999, 48 Morb Mort Weekly Rpt. 986 (1999), https://www.cdc.gov/mmwr/ preview/mmwrhtml/mm4843a2.htm 46  Centers for Disease Control and Prevention, Smoking & Tobacco Use: Fast Facts, https://www.cdc. gov/tobacco/data_statistics/fact_sheets/fast_facts/index.htm (last visited July 22, 2018). 47 Id. 48 Id. 49  Stacy Simon, Cancer Statistics Report: Death Rate Down 23% in 21 Years, American Cancer Society, Jan. 7, 2016, https://www.cancer.org/latest-­news/cancer-­statistics-­report-­death-­rate-­down-­23-­ percent-­in-­21-­years.html

128   Wendy E. Parmet beyond the scope of this chapter. A brief review, however, may help to illustrate key features of the US approach to NCDs because tobacco control remains, in many ways, the model for other public health efforts targeting NCDs. Importantly, the United States has not ratified the FCTC. Nevertheless, US laws regarding tobacco, taken as a whole, embody many of the Convention’s provisions, including those relating to reducing the demand for tobacco, protecting people from exposure to tobacco, and prohibiting misleading packaging and labeling. Other laws and initiatives provide support for tobacco cessation and regulate (though without fully banning) tobacco advertising. However, for constitutional reasons discussed further later, the United States does not require plain packaging or graphic warning labels. To reduce tobacco-­related diseases, US jurisdictions utilize almost all of the tools in the public health law toolkit, including private litigation, marketing restrictions, warning labels, and excise taxes, among others. US tobacco control laws also demonstrate the fragmentation of authority, the heavy influence of individualism and concomitant reluctance to impose command and control regulations, tensions between protecting public health and promoting commercial interests, and civil litigation’s prominent role in driving public health policy. Indeed, long before regulatory efforts commenced in earnest, hundreds of private civil lawsuits were filed in the 1950s and early 1960s claiming that cigarettes were responsible for the plaintiffs’ ill health.50 These early suits were uniformly unsuccessful due in large measure to the industry’s “no-­compromise strategy,” which rejected any settlements and relied on fighting every single claim as zealously as possible, in large measure to make it overwhelmingly expensive to sue tobacco companies. In terms of federal regulation, throughout much of US history, federal policy sought to support tobacco farming and the tobacco industry.51 For example, the Agricultural Adjustment Act of 1938 specifically stated that “[t]he marketing of tobacco constitutes one of the greatest basic industries of the United States.”52 That emphasis on protecting industry began to change, albeit slowly, following the release in 1964 of the first Surgeon General General’s report on the health effects of cigarette smoking.53 That report helped alter public perceptions about the health effects of smoking and led, in 1965, to the first federal act requiring warning labels on all cigarettes packages.54 In 1969, Congress amended that Act by strengthening the warning labels and barring television advertising for cigarettes.55 Tellingly, these acts not only sought to warn consumers, they also overrode stronger Federal Trade Commission (FTC) regulations that were set to require 50 Robert L. Rabin, A Sociolegal History of Tobacco Tort Litigation, 44 Stan. L. Rev. 853, 855 (1991). 51 See Food and Drug Admin. v. Brown & Williamson, 529 U.S. 120, 137 (2000). 52  Id., quoting then 7 U.S.C. Sec. 1311. 53  History of the Surgeon General’s Reports on Smoking and Health, Centers for Disease Control And Prevention (Jan. 11, 1964), https://www.cdc.gov/tobacco/data_statistics/sgr/history/index.htm 54  Federal Cigarette Labeling and Advertising Act of 1965, Pub. L. 89–92, 79 Stat. 282 (1965). 55  Public Health Cigarette Smoking Act of 1969, Pub. L. 91–222, 84 Stat. 87 (1969). If they were challenged, the advertising restrictions might not survive constitutional review under current doctrine. See infra text accompanying notes 136–147.

US Law Relating to Noncommunicable Diseases   129 warning labels on packages and advertisements and preempted stronger state laws.56 In effect, they served to protect the health of both Americans and the tobacco industry. Over the course of the next three decades, tobacco control slowly gained speed, especially as evidence accumulated regarding nicotine’s addictive qualities, the harmful effects of second-­hand smoke, and industry practices, especially the targeting of youth and the manipulation of nicotine levels to increase addiction. This evidence, which was skillfully deployed by public health and anti-­smoking advocates, helped to change public opinion about tobacco regulation, turning the problem from one of individual responsibility to one of corporate culpability.57 Interestingly, this framing was made possible in part due to information that came to light as a result of a new wave of tobacco litigation, which began in the 1990s. Many of these lawsuits focused on the addictiveness of smoking as a way of overcoming defenses based on the plaintiff ’s own choices or responsibility for smoking.58 Importantly, this litigation led to the discovery and publication of tobacco company documents that revealed the industry’s intentional manipulation of nicotine levels and marketing to minors.59 The publicity generated, and subsequent hearings before Congress, helped lay the popular foundation for further regulation, thereby demonstrating the complementary relationship between private litigation and public regulation that has often characterized US public health law. After the release of this damning evidence, the Federal Food and Drug Administration (FDA) in 1996 issued broad regulations regarding youth access, labeling, and cigarette marketing. These regulations were struck down by the US Supreme Court, which held that the FDA lacked statutory authority to regulate cigarettes.60 By that time, however, the tobacco companies had already settled civil litigation claims brought by state governments in a Master Settlement Agreement (MSA), under which the companies agreed to pay the states billions of dollars to cover some of the costs of tobacco-­related illnesses.61 Many tobacco control advocates initially hoped that the states would use the money they received from the MSA to promote tobacco control and related public health efforts. For the most part, however, that hasn’t happened as states have used the money for other purposes.62 Nevertheless, because of the fact that tobacco companies raised the price of cigarettes in order to pay the money owed under the settlement, the MSA may have played some role in reducing consumption.63 56  Cipollone v. Liggett Group, Inc., 505 U.S. 504, 513–515 (1992). 57  For a fuller discussion, see Parmet, supra note 16, at 934. 58 Wendy E. Parmet, Tobacco, HIV, and the Courtroom: The Role of Affirmative Litigation in the Formation of Public Health Policy, 36 Houston L. Rev. 1663, 1676–1678 (1999). 59  Id. at 1677. 60  Food and Drug Admin. v. Brown & Williamson Tobacco Corp, 529 U.S. 120, 161 (2000). 61  See Public Health Law Center at Mitchell Hamline School of Law, Master Settlement Agreement, http://www.publichealthlawcenter.org/topics/tobacco-­control/tobacco-­control-­litigation/master-­ settlement-­agreement (last visited July 22, 2018). 62 Id. 63  Derek Carr, Corey S. Davis, & Lainie Rutkow, Reducing Harm Through Litigation Against Opioid Manufacturers? Lessons From the Tobacco Wars, 133 Pub. Health Rpts. 207, 209 (2018).

130   Wendy E. Parmet Under the MSA, tobacco companies also agreed to significant restrictions on their advertising and marketing practices, especially as applied to youth.64 The voluntary nature of these restrictions has proved critical due to the increasing protection courts have given to commercial speech, as discussed further later. Also in the 1990s, in the absence of broad federal regulations, states and especially localities began to enact a wide array of laws regulating the sale and use of cigarettes and sometimes other tobacco products. In 1990, San Luis Obispo, California, became the first US jurisdiction to enact a comprehensive indoor smoking law.65 By 2017, 25 states and hundreds of cities and counties had laws banning smoking in workplaces, restaurants, and bars, and five additional states had laws barring smoking in just restaurants and bars.66 Although ostensibly designed to protect workers and patrons from second-­ hand smoke, these laws have also had the effect of denormalizing smoking, which may have reduced its allure.67 States and localities have also passed laws banning the sale of cigarettes to minors, as well as the addition of flavors thought to attract youth.68 Many states and localities have also significantly increased tobacco excise taxes to discourage consumption.69 In 2009, Congress finally enacted a broad tobacco control law. The Family Smoking Prevention and Tobacco Control Act (TCA) established a new center for tobacco control within the FDA and granted the agency broad authority over tobacco products.70 The Act also prohibited the sale of cigarettes and smokeless tobacco to minors, granted the FDA authority over tobacco marketing and labeling, and required FDA preapproval of new tobacco products, except if substantially equivalent products were on the market by 2007, to determine if they are detrimental to public health, mislabeled, or result from manufacturing processes that do not conform to good practices.71 In addition, the Act extended federal preemption of tobacco regulation by barring state laws relating to premarket review, misbranding, labeling, or the adulteration of tobacco products72 However, the TCA explicitly stated that it did not preempt state laws relating to taxation, sales, distribution, youth possession, use restrictions, or fire safety standards.73 Hence the TCA allowed states to continue to enforce indoor air laws as well as laws restricting the use by or sale of tobacco products to youth. 64  See id. 65 Diller, supra note 36, at 1229. 66  Campaign for Tobacco Free Kids, U.S. State and Local Issues: Smoke-­Free Laws, https://www. tobaccofreekids.org/assets/factsheets/0332.pdf, last visited Aug. 22, 2018. 67 Parmet, supra note 16, at 932. 68  Id. at 935. 69  Campaign for Tobacco Free Kids, State Cigarette Excise Tax Rates & Rankings, June 25, 2018, https://www.tobaccofreekids.org/assets/factsheets/0097.pdf 70  Family Smoking Prevention and Tobacco Control Act of 2009, Pub. L. No. 111–131, 123 Stat. 1776 (2009). For a critical review of tobacco regulation under the TCA, see Michah Berman, The Faltering Promise of FDA Tobacco Regulation, 12 St. Louis J. Health Policy 145 (2019). 71  21 USC § § 387j (c)(2)(2012). 72  Id. at § 387p(a)(2)(2012). Federal law previously prohibited states from regulating health-­related labeling. See Cippolone v. Liggett Group, Inc., 505 U.S. 504, 514 (1992) (construing 15 U.S.C. A. § 1334(b)). 73  21U.S.C. § 387p(a)(2)(2012).

US Law Relating to Noncommunicable Diseases   131 Since the TCA’s passage, two issues have become especially salient for tobacco control efforts. The first concerns the viability of advertising restrictions and mandatory warning labels in light of the evolving constitutional commercial speech doctrine.74 Laws regulating the advertising and packaging of tobacco products have been in effect since 1965 and are a well-­established component of tobacco control efforts; they are also called for by the TCA. However, as discussed further later, in recent years the federal courts have been giving greater protection to so-­called commercial speech, striking down on First Amendment grounds numerous laws that regulate advertising and marketing. Even before the TCA was enacted, these decisions had significantly hampered tobacco control efforts. For example, in 2001, the Supreme Court struck down a Massachusetts law banning outdoor advertising for smokeless tobacco products and point-­of-­sale displays within a 1,000 feet of a school as violating the commercial speech rights of tobacco companies and sellers.75 Then, in 2012, in R.J. Reynolds v. Food and Drug Administration, the US Court of Appeals for the District of Columbia struck down the FDA’s efforts pursuant to the TCA to require graphic warning labels on cigarette packages.76 Fearing that the decision would be upheld by the Supreme Court, the FDA did not seek review. However, in 2020, FDA finally issued new regulations, which industry promptly challenged in court. In May 2020, FDA and the plaintiffs agreed to postpone the regulations’ effective day to October 2021.77 The second issue relates to the regulatory response to lower risk tobacco products, such as e-­cigarettes and low nicotine cigarettes, which may be less harmful or addictive than traditional cigarettes. In recent years, as youth smoking rates have fallen, use of e-­cigarettes among youth has increased so that “e-­cigarettes are now the most commonly used tobacco product among youth, surpassing conventional cigarettes.”78 Although these products appear to be far less harmful than traditional cigarettes and they may help some smokers quit the habit, many public health experts worry that youth who use the such products, especially e-­cigarettes, may become addicted to nicotine and eventually begin to smoke traditional cigarettes.79 Tobacco control advocates also worry that the proliferation of “vaping” in public spaces may help to renormalize tobacco use, thus undermining decades of anti-­tobacco messaging. In contrast, some public health advocates believe that the proliferation of less dangerous tobacco products offers the possibility of a harm reduction strategy that relies on reducing the risk of tobacco use, as opposed to simply trying to limit exposure. 74  See infra text accompanying notes 136–147. 75  Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 554–571 (2001). 76  R. J. Reynolds Tobacco Co. v. FDA, 696 F.3d. 1205, 1221–1222 (D.C. Cir. 2012). 77  See U.S. Food & Drug Admin. Cigarette Labeling and Health Warning Requirements, https://www. fda.gov/tobacco-­products/labeling-­and-­warning-­statements-­tobacco-­products/cigarette-­labeling-­and-­ health-­warning-­requirements (2020). 78  U.S. Dept of Health and Human Serv., E-­Cigarette Use among Youth and Young Adults: A Report of the Surgeon General 5 (2015), https://e-­cigaretts.surgeongeneral.gov/documents/2016_SRG-­Full_ Report_non_508.pdf. 79  See also Parmet, supra note 16, at 925.

132   Wendy E. Parmet For several years after the adoption of the TCA, the FDA struggled to determine its approach to e-­cigarettes and other less harmful tobacco products. In 2016, the FDA finally issued broad regulations applicable to e-­cigarettes, cigars, pipe tobacco, and other tobacco products that had not previously been subject to its regulatory oversight.80 Although these regulations did not go as far as many anti-­tobacco advocates wanted, they did ban the sale of e-­cigarettes and other tobacco products to minors, require warning labels regarding the addictiveness of nicotine, ban free samples, and require pre-­ market review for certain new products.81 However, after the Trump Administration took office in 2017, under the leadership of then-­Commissioner Scott Gottlieb, the FDA announced a change of course to one promoting harm reduction and the marketing of less dangerous tobacco products. In furtherance of this approach, the FDA delayed many of the provisions of the 2016 regulations and announced that it would revisit many of the provisions in the interim.82 More recently, in response to a dramatic surge in e-­cigarette use among youth, the FDA altered its approach again, this time adopting a more aggressive stance by initiating enforcement actions against e-­cigarette manufacturers and issuing a guidance warning against the sale of certain flavored e-­cigarettes in locations open to minors.83 In addition, some states and private parties have filed lawsuits against Juul, the leading e-­cigarette manufacturer, claiming that it marketed its product to underage teens and deceived customers about the product’s safety.84 In the summer of 2019, San Francisco became the first large city to ban e-­cigarette sales.85 These developments suggest that attitudes toward e-­cigarettes are changing and support for them as tools for harm reduction is fading. Instead, e-­cigarette manufacturers, especially Juul (which is partially owned by tobacco giant Altria),86 are now being viewed as similar to tobacco companies in 80  Deeming Tobacco Products to Be Subject to the Federal Food, Drug, and Cosmetic Act, as Amended by the Family Smoking Prevention and Tobacco Control Act: Restrictions on the Sale and Distribution of Tobacco Products and Required Warning Statements for Tobacco Products, 81 Fed. Reg. 28974 (May 10, 2016). 81  US Food and Drug Admin., The Facts on the FDA’s New Tobacco Rule, June 16, 2016, https://www. fda.gov/ForConsumers/ConsumerUpdates/ucm506676.htm 82  US Food & Drug Admin., Extension of Certain Tobacco Product Compliance Deadlines Related to the Final Deeming Rule, Guidance for Industry (Revised), Nov. 2017, https://www.fda.gov/downloads/ TobaccoProducts/Labeling/RulesRegulationsGuidance/UCM557716.pdf/ 83  US Food and Drug Admin., Statement from FDA Commissioner Scott Gottlieb, M.D. on Advancing New Policies Aimed at Preventing Youth Access to, and Appeal of, Flavored Tobacco Products, Including E-­Cigarettes and Cigars (March 13, 2019), https://www.fda.gov/NewsEvents/Newsroom/Press Announcements/ucm633291.htm. For a discussion of the FDA’s approach to tobacco regulation under the Trump Administration, see Berman, supra note 70, at 158–162. 84  Deanna Paul, North Carolina Sues Juul, Setting Up a Fresh Legal Fight for the Embattled E-cigarette Company, Wash. Post, May 15, 2019, https://www.washingtonpost.com/health/2019/05/15/north-­ carolina-­sues-­juul-­setting-­up-­fresh-­legal-­fight-­embattled-­e-­cigarette-­company. Clyde Hughes, Lawsuit: Juul Violating Federal Law by Marketing to Teens, UP, April 15, 2019, https://www.upi.com/Health_ News/2019/04/15/Lawsuit-­Juul-­violating-­federal-­law-­by-­marketing-­to-­teens/4691555344870/ 85  Laura Klivans, San Francisco Bans Sales of E-­Cigarettes, NPR: Shots, June 15, 2019, https//www. npr.org/sections/health-­shots/2019/06/25/735714009/san-­francisco-­poised-­to-­ban-­sale-­of-­e-­cigarettes 86  Angelica LaVito, Tobacco Giant Altria Takes 35% Stake in Juul, Valuing E-­Cigarette Company at $38 Billion, CNBC, Dec. 20, 2018, https://www.cnbc.com/2018/12/20/altria-­takes-­stake-­in-­juul-­a-­pivotal-­ moment-­for-­the-­e-­cigarette-­maker.html

US Law Relating to Noncommunicable Diseases   133 terms of their responsibility for tobacco use, especially among minors. Moreover, the regulatory approach that developed in response to cigarettes, and which includes civil litigation and state and local use laws, as well as federal regulation coupled with partial preemption, now seems to be guiding the response to e-­cigarettes. Whether this model, which has undoubtedly helped to reduce the rate of cigarette smoking and NCDs in the United States over the past 60 years will offer a satisfactory response to the promise and challenge of e-­cigarettes and other reduced-­harm tobacco products remains to be seen.

4  Unhealthy Foods As rates of obesity and associated NCDs, including diabetes, have risen in recent decades, public health advocates have sought to apply the “tobacco model” to reduce the consumption of unhealthy foods, including sugary beverages and processed foods that are associated with these diseases.87 This has led to calls for a variety of supply- and demand-­side strategies to be implemented at the local, state, and federal levels, including taxing sugary beverages, mandating warning labels, and banning unhealthy ingredients. Private parties have also attempted to use the tobacco “playbook” by suing fast food companies.88 To date, however, these approaches have gained far less traction as applied to unhealthy foods than they have with tobacco. Currently, there is no comprehensive federal strategy to address the NCDs related to the consumption of unhealthy foods, nor is there even an agreed-­upon national commitment to address the problem.89 Moreover, civil litigation against fast food companies has for the most part been unsuccessful,90 and many local laws have been struck down by courts.91 Underlying this lack of success is the widespread perception that obesity and associated NCDs are matters of personal, rather than corporate, responsibility.92 In short, 87  E.g., Jonathan D. Klein & William Dietz, Childhood Obesity: The New Tobacco, 29 Health Aff. 388 (2010). 88  See Thompson, supra note 39, at 696. 89  It should be noted that the FDA has regulatory authority over most foods and beverages and is charged under the Food Safety and Modernization Act with requiring “comprehensive, science-­based preventive controls” to ensure the safety of the food supply. US Food & Drug Administration, Background on the FDA Food Safety Modernization Act (FSMA), https://www.fda.gov/newsevents/ publichealthfocus/ucm239907.htm. The FDA’s primary focus, however, is on preventing food-­borne illnesses and contamination and ensuring accurate labeling, not preventing obesity or reducing consumption of high-­calorie, low-­nutrition food. However, as discussed earlier, the FDA has recently moved to ban trans-­fats. The US Department of Agriculture has regulatory authority over the production and safety of most meat, poultry, and egg products. See Gretchen Goetz, Who Inspects What? A Food Safety Scramble, Food Safety News, Dec. 16, 2010, https://www.foodsafetynews. com/2010/12/who-­inspects-­what-­a-­food-­safety-­scramble/ 90  See Thompson, supra note 39, at 696. 91  Sarah A. Roache et al., Big Food and Soda versus Public Health: Industry Litigation Against Local Government Regulations Promoting Healthy Diets, 45 Fordham Urb. L. J. 1051, 1085 (2018). 92  Kelly D. Brownell et al., Personal Responsibility and Obesity: A Constructive Approach to a Controversial Issue, 29 Health Aff. (Mar. 2010), https://www.healthaffairs.org/doi/10.1377/ hlthaff.2009.0739

134   Wendy E. Parmet Americans, perhaps far more than many Europeans,93 tend to believe that weight and the NCDs associated with it are matters of personal choice and that laws that seek to control obesity are paternalistic interventions of the “nanny state.”94 Perhaps for this reason, the interventions that have been put in place have, for the most part, sought to offer individuals more healthy choices, for example, by facilitating farmers’ markets, setting up bike lanes, and providing individuals with information about healthy eating and exercising.95 As with tobacco, local governments have been at the forefront of efforts to reduce consumption of unhealthy foods and increase activity levels. Some of these efforts have been successful, but others have been either struck down in court or reversed following a political backlash, which has often been funded by food companies. On the success side, cities led the way in banning trans-­fats,96 a policy that was eventually extended to the nation by the FDA.97 Several cities, including New York and Philadelphia, have also enacted laws requiring restaurants to post calories, a policy that has also been extended, at least with respect to chain restaurants, nationwide.98 More recently, Berkeley, California, and Philadelphia, Pennsylvania, have taxed sugary beverages.99 However, Cook County, Illinois’ soda tax was repealed after surviving a court challenge.100 One notable failure was New York City’s 2012 regulation limiting portion sizes in sugary beverages sold in restaurants and movie theaters. The so-­called Portion Cap Rule ignited a strong public backlash with New York Mayor Michael Bloomberg widely derided as “Nanny Bloomberg.”101 The rule was eventually struck down by New York State’s highest court, which held that the city’s board of health exceeded the scope of its statutory authority. According to the court, “An agency that adopts a regulation, such as the Portion Cap Rule or an outright prohibition of sugary beverages, that interferes with common place daily activities preferred by large numbers of people must necessarily wrestle with complex value judgments concerning personal autonomy and economics. That is policymaking, not rulemaking.”102 In 2015, San Francisco enacted an ordinance requiring advertisements for sugary beverages to warn that such beverages contribute to “obesity, diabetes and tooth decay.” In 2019, this law was enjoined by the US Court of Appeals for the Ninth Circuit, sitting en 93  For a comparison of American and French attitudes and regulatory responses, see M. Neil Browne et al., Attacking Obesity: The Paternalistic Approach of France versus the Conservative Approach of United States, 39 Whittier L. Rev. 1 (2018). 94 Cindy D. Kam, The Polls-­Trends: Obesity, 81 Pub. Opinion Qrtly. 973, 985–988 (2017). 95  Id. at 288. 96 Friedman, supra note 12, at 1749. 97  US Food & Drug Administration, Final Determination Regarding Partially Hydrogenated Oils (Removing Trans Fat), https://www.fda.gov/food/ingredientspackaginglabeling/ foodadditivesingredients/ucm449162.htm (May 18, 2018). 98 Jason P. Block, The Calorie-­Labeling Saga—Federal Preemption and Delayed Implementation of Public Health Law, 379 New Eng. J. Med. 103 (2018). 99  Roache et al., supra note 90, at 1061, 1070–1078. 100  Id. at 1077–1078. 101 Friedman, supra note 12, at 1089. 102  New York Statewide Coalition of Hisp. Chambers of Commerce v. New York City Dept. of Health & Mental Hygiene, 16 N.E. 3d 538, 549 (N.Y. 2014).

US Law Relating to Noncommunicable Diseases   135 banc, for violating advertisers’ constitutional right to commercial speech.103 The court found that San Francisco could not prove that the requirement that the warnings occupy 20% of advertisements did not place an “undue burden” on the plaintiffs’ First Amendment rights. Although states have adopted some measures designed to reduce the incidence of NCDs associated with unhealthy eating, they have been far more supportive of the food industry than have many of their cities.104 For example, more than 20 states have enacted some version of the Commonsense Consumption Act, commonly known as the Cheeseburger Bill, which bans lawsuits holding fast food companies responsible for obesity and related illnesses.105 At the federal level, the FDA and the US Department of Agriculture are charged with ensuring the safety of the food supply. However, they tend to focus on preventing food-­ borne infections and have failed to take comprehensive measures to reduce the obesogenic nature of the American diet.106 To the contrary, some critics have argued that federal regulatory concerns have long prioritized promoting the food industry.107 For example, despite repeated calls for the FTC to regulate food marketing to children, the agency has failed to do so. Indeed, even the efforts of an Interagency Working Group on Food Marketed to Children that aimed to develop voluntary guidelines was effectively quashed by Congress in 2012.108 Despite the failure of that initiative, during the Obama Administration, First Lady Michelle Obama sought to shine a spotlight on the problem of childhood obesity with her Let’s Move initiative. Although that campaign emphasized individual, voluntary actions, such as exercising or growing vegetable gardens, it did draw attention to the unhealthy nature of the American diet and helped lead to the replacement of the federal government’s prior nutrition guidelines (known as the Food Pyramid) with a new symbol (MyPlate), which gives greater priority to fruits and vegetables over carbohydrates and dairy.109 The Let’s Move campaign also helped lead to the enactment in 2010 of the Healthy, Hunger-­Free Kids Act, which aimed to improve the healthfulness of school lunches.110 The Trump Administration, however, rolled back many of the targets established by that Act.111 103  Am. Bev. Ass’n. v. City and Cty. San Francisco, 916 F.3d 749 (9th Cir. 2019)(en banc). 104  For a discussion of why this may be the case, see Diller, supra note 36, at 1255–1269. 105  Christopher S. Carpenter & D. Sebastian Tello-­Trillo, Do Cheeseburger Bills Work? Effects of Tort Reform for Fast Food, 58 J. L. & Econ. 805 (2015) (finding that cheeseburger bills might make people be more careful about their won weight). 106  See supra note 89. 107  Marion Nestle, Food Politics: How the Food Industry Influences Nutrition and Health passim (2002). 108 William H. Dietz, New Strategies to Improve Food Marketing to Children, 32 Health Affairs 1652, 1654–1655 (2013), https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2012.1294 109  William Neuman, Nutrition Plate Unveiled, Replacing Food Pyramid, N.Y. Times (June 2, 2011), at https://www.nytimes.com/2011/06/03/business/03plate.html 110  Jenny Nelson, The Healthy, Hunger-­Free Kids Act: The Fight Must go On, 95 Denv. L. Rev. Online 42, 51 (2018). 111 Id.

136   Wendy E. Parmet Also during the Obama Administration, as part of the Affordable Care Act, Congress required food establishment chains to post calorie counts on menus.112 Due to stiff industry resistance and a protracted regulatory process, regulations implementing the law did not take effect until May 2018.113 In the meantime, the federal act required states and municipalities to halt calorie-­labeling laws that were not yet in effect, and some states even rescinded their laws in anticipation of the federal regulations.114 Given the current deregulatory agenda in Washington and the lack of broad political support for obesity control measures, new federal regulations are unlikely in the immediate future. Indeed, unless and until the problems of obesity, diabetes, hypertension, and related conditions come to be seen more widely as public health problems rather than matters of individual responsibility, legal efforts to reduce the incidence of such diseases are apt to remain limited and, for the most part, unsuccessful.

5  Looking to the Future: Opioids and Firearms Tobacco and unhealthy foods are not the only products associated with NCDs nor are they the only products that have been the subject of public health laws. In recent years, public health advocates and policy-­makers have turned their attention to several other products, including prescription opioids and firearms. In both cases, the “tobacco model” has proved highly influential. In particular, the legal response to both firearms and opioids demonstrate the influence of an individualistic framing; the diffusion of authority between local governments and states and the federal government, with innovation often coming at the local level; and a reliance by advocates on civil litigation. Space precludes a full discussion of either issue, but a brief review illustrates how the response to each compares to those employed against tobacco and unhealthy foods. With both opioids and guns, debates over individual responsibility have helped to frame the legal response. For example, the highly influential National Rifle Association (NRA) has been largely successful in framing gun violence as a problem caused by the acts of mentally ill or criminal individuals115 rather than lax gun laws or the manufacturing and marketing practices of gun-­makers. Rather than seek tighter restrictions on the marketing, manufacturing, or sale of guns, the NRA argues that access to firearms should be enhanced, so that people can protect themselves. This position received significant constitutional support from the Supreme Court’s decision in District of

112  Sec. 4205, Patient Protection and Affordable Care Act, Pub. L. 111–148, 124 Stat. 119 (2010). 113 Block, supra note 97, at 104. 114 Id. 115  James Downie, The NRA Is Winning the Spin Battle, Wash. Post. (Feb. 20, 2018), https://www. washingtonpost.com/blogs/post-­partisan/wp/2018/02/20/the-­nra-­is-­winning-­the-­spin-­battle/?utm_ term=.366921b2fa85

US Law Relating to Noncommunicable Diseases   137 Columbia v. Heller,116 which overturned decades of settled doctrine by holding that the Second Amendment protected an individual’s right to possess firearms for self-­ protection. Although the majority of gun control laws have survived constitutional challenge post-Heller, numerous cases have been brought, and many have been successful in the lower federal courts.117 Equally important, Heller has helped to reinforce the notion that individuals have a constitutional right to a gun in order to protect themselves. While no comparable constitutional right exists with respect to opioids, the opioid crisis has also been largely viewed largely through an individualist lens. But rather than protecting users’ rights, as with firearms, the legal response has been mostly punitive, with drug possession criminalized and the nation long engaged in the so-­called “war on drugs.”118 More recently, as with tobacco, the perception has shifted, and manufacturers and prescribing physicians have come to be seen as at least partially responsible for the problem.119 Legal responses to the opioid crisis and firearm violence also replicate the complex diffusion of legal authority that is evident with US responses to tobacco and unhealthy foods. In the case of firearm violence, Congress has failed in recent years to enact significant gun control measures despite polls showing that a majority of Americans support stricter gun laws.120 In the absence of Congressional action, most of the legal responses have come from states and localities. However, the highly polarized nature of the debate over gun violence has led to dramatically different responses across the country, with some jurisdictions moving to restrict gun sales and others moving to relax restrictions and enhance the rights of gun owners.121 As noted previously, preemption has also played a large role, with both federal law and many state laws, in overriding local gun control initiatives. A similar pattern exists with respect to opioids. Although federal law has long regulated the sale, possession, and prescription of opioids, and Congress enacted the SUPPORT for Patients and Communities Act of 2018, which provides increased

116  District of Columbia v. Heller, 554 U.S. 570 (2008). It should be noted, however, that the Court did not hold that all gun control laws were unconstitutional, and many have been found to be constitutional despite Heller. 117  Eric Ruben & Joseph Blocher, From Theory to Doctrine: An Empirical Analysis of the Right to Keep and Bear Arms after Heller, 67 Duke L.J. 1433, 1442 (2018). As this chapter is in press, the Supreme Court has agreed to review a case challenging a New York gun control law. See N.Y. State Rifle & Pistol Assn., Inc. v. Corlett, 818 Fed. Appx. 99 (2d Cir. 2020), cert. granted__ U.S. __, 2021 WL 1602643 (Apr. 26, 2021). 118  Leo Beletsky, America’s Favorite Antidote: Drug-­Induced Homicide in the Age of the Overdose Crisis (May 2018), https://ssrn.com/abstract=3185180 119  Nabarun Dapgupta et al., Opioid Crisis: No Easy Fix to its Social and Economic Determinants, 108 Am. J. Pub. Health 182, 183 (2018). 120  Steven Shepard, Gun Control Support Surges in Polls, Politico, Feb. 28, 2018, https://www. politico.com/story/2018/02/28/gun-­control-­polling-­parkland-­430099 121  Nathalie Baptiste, What You Need to Know About Red Flag Gun Laws, Mother Jones, Mar. 7, 2018, https://www.motherjones.com/politics/2018/03/what-­you-­need-­to-­know-­about-­red-­flag-­gun-­laws/

138   Wendy E. Parmet funding for prevention and treatment,122 there still is no comprehensive federal response to the rise in overdose-­related fatalities. Rather, as we have seen elsewhere, much of the innovation has come from states and municipalities. Many of these laws target physicians and prescribing practices123; others mirror communicable disease and mental health laws by allowing for the civil commitment and forced treatment of addicts.124 Inspired by the success of needle exchange programs to prevent HIV and other sexually transmitted infections, some local jurisdictions are contemplating harm reduction measures such as the establishment of safe injection centers. These centers, however, may run afoul of federal drug laws.125 In the absence of a comprehensive response to both the problems of firearm violence and opioids, advocates in state and local governments have also followed the tobacco model by commencing civil litigation. As Timothy Lytton explains, the litigation that was brought against the gun industry was “part of a larger trend to reframe violence as a public health problem”126 and a move to a response that focused more on regulating the industry than criminalizing individual behavior. However, despite some early victories, this litigation faced formidable doctrinal hurdles127 and was largely (though not completely) quelled by the 2005 federal Protection of Lawful Commerce in Arms Act.128 Cities and states, as well as private parties and Native American tribes, have also brought several hundred lawsuits against the manufacturers and distributors of opioids. As of the summer 2018, more than 700 claims were consolidated in a nationwide multidistrict litigation based in Ohio.129 Claims have also been brought in state courts around the country. In early 2019, the state of Oklahoma agreed to settle its claims against Purdue Pharmaceuticals for $270 million130 and Teva Pharmaceutical Industries for $85 million dollars.131 Both settlements included provisions to dedicate a significant share 122  Pub. L. 115–271, 132 Stat. 3894 (2018). 123  Leo Beletsky, Deploying Prescription Drug Monitoring to Address the Opioid Crisis: Ideology Meets Reality, 15 Ind. Health L. Rev. 49, 164 (2018). 124  Leo Beletsky, Elisabeth J. Ryan, & Wendy E. Parmet, Involuntary Treatment for Substance Use Disorder: A Misguided Response to the Opioid Crisis, Harvard Health Blog (Jan. 25, 2018), https:// www.health.harvard.edu/blog/involuntary-­treatment-­sud-­misguided-­response-­2018012413180 125  Heather Knight, San Francisco Would Defy Federal Law with Safe Injection Sites, San Fran. Chron., May 22, 2018, https://www.sfchronicle.com/news/article/San-­Francisco-­would-­defy-­federal-­ law-­with-­safe-­12932416.php 126 Timothy D. Lytton, Introduction: An Overview of Lawsuits against the Gun Industry, in Suing the Gun Industry: A Battle at the Crossroads of Gun Control & Mass Torts 4 (Timothy D. Lytton ed. 2005). 127 Stephen D. Sugarman, Torts and Guns, 10 J. Tort L. 3, 14–20 (2017). 128  See supra note 31. 129  Abbe R. Gluck, Ashley Hall, & Gregory Curfman, Civil Litigation and the Opioid Epidemic: The Role of Courts in a National Health Crisis, 46 J. Law Med. & Ethics 351, 361 (2018). 130  Jan Hoffman, Purdue Pharma and Sacklers Reach $270 Million Settlement in Opioid Lawsuit, N.Y. Times, March 26, 2019, https://www.nytimes.com/2019/03/26/health/opioids-­purdue-­pharma-­ oklahoma.html 131  Nate Raymond, Oklahoma Judge Approves Teva’s $85 Million Opioid Settlement, Reuters, June 24, 2019, https://www.reuters.com/article/uk-­usa-­opioids-­litigation/ oklahoma-­judge-­approves-­tevas-­85-­million-­opioid-­settlement-­idUSKCN1TP2M8

US Law Relating to Noncommunicable Diseases   139 of the funds to research and treatment. What will happen to the remaining claims, and whether future settlements can avoid the pitfalls faced by the MSA, remains to be seen.

6  Commercial Speech As noted earlier, the individualistic framing of NCDs in the United States has led policy-­ makers to focus much of their efforts on laws that seek to provide individuals with information about the health impact of their behavioral choices. Thus the first legal response to the Surgeon General’s report on smoking was the passage of the Cigarette Label Act, which required warning labels on cigarettes. In the decades since, there has been a proliferation of laws requiring warning labels or otherwise regulating the information that is given to consumers about the health impacts of products. Such laws are often defended as more respectful of individual liberty than those that prohibit the sale or use of dangerous products. They are also less burdensome on industry. Nevertheless, the strong pull of individualism that underlies US NCD laws and the paternalism critique have placed even these laws in constitutional jeopardy. For most of American history, the First Amendment’s prohibition on the regulation of speech did not apply to speech used to market or sell a service or product. That changed in 1975, when the Supreme Court in Bigelow v. Virginia overturned a publisher’s conviction for carrying advertisements for abortion.132 Emphasizing that commercial speech can serve the interests of consumers, the Court stated that “speech is not stripped of First Amendment protection merely because it appears in [the form of an advertisement].”133 In the years that followed, the Supreme Court struggled to develop a consistent approach for reviewing commercial speech laws. Without discussing all of the permutations, suffice it to say that, between 1975 and 2011, the Court most often subjected laws that banned commercial speech to what is known as “intermediate scrutiny” under the so-­called Central Hudson test.134 This allowed states to prohibit commercial speech that was misleading or related to unlawful activity and to limit truthful commercial speech when the state had a substantial interest (such as protecting health), and the regulation directly advanced that interest and was not more expansive or burdensome than necessary.135 Governments were given even wider latitude in compelling warning labels or disclosures. Noting the “material differences” between laws prohibiting and laws mandating speech, the Supreme Court in Zauderer v. Office of Disciplinary Counsel held that laws that compelled factual and uncontroversial commercial information were

132  421 U.S. 809 (1975). 133  Id. at 829 (1975). 134  See Central Hudson Gas & Elec. Corp. v. Public Serv. Comm., 447 U.S. 557,562–566 (1980). For a discussion of the evolution of the so-­called commercial speech doctrine, Wendy E. Parmet & Jason A. Smith, Free Speech and Public Health: Unraveling the Commercial-­Professional Speech Paradox, 78 Ohio S. L.J. 887, 889–892 (2017). 135  447 U.S. at 566.

140   Wendy E. Parmet constitutional.136 As a result, laws requiring warning labels on a wide range of products, such as cigarettes, were presumed constitutional. In recent years, the Supreme Court has changed course, substantially increasing the scrutiny given to regulations that implicate speech. For example, in 2010, in Citizens United v. FEC, the Court held that the First Amendment’s protections for speech extended to political spending by corporations.137 The following year in Sorrell v. IMS Health, Inc., the Court relied on the First Amendment to strike down a state law barring the sale of data pertaining to prescription practices.138 Without deciding if the speech was commercial, the Court stated that all laws that discriminate on the basis of content or speaker must be subject to the strictest form of judicial review, “strict scrutiny,” under which laws are almost always found unconstitutional. More recently, in National Institute of Family and Life Advocates v. Becerra (NIFLA), a 5–4 Court in an opinion by Justice Thomas reiterated its holding in Sorrell and ruled that Zauderer’s more lenient approach to mandatory disclosure laws was inapplicable to a California law that required licensed crisis pregnancy centers (CPCs) to disclose the factually accurate information that the state offered free or low-­cost reproductive health services, including abortion, because abortion itself is controversial.139 This reading of Zauderer’s requirement that the disclosure be noncontroversial suggests that even truthful, factual disclosures (such as laws listing the health dangers associated with a product) may be struck down if there is any controversy about the underlying issue.140 Moreover, the Court in NIFLA also appeared to enhance the scrutiny required even when Zauderer’s more lenient analysis applies, finding that a provision in the California law requiring unlicensed CPCs to disclose that they were not licensed was likely unconstitutional because the mandate was unduly burdensome.141 As a result, the assurance by the Court in NIFLA that it did not “question the legality of health and safety warnings long considered permissible, or purely factual and uncontroversial about commercial products,”142 seems unconvincing. As Justice Breyer noted in his dissent, the Court offered no test for distinguishing warning laws that would be upheld from those that would not be.143 Indeed, the Court’s admonition that the state could have used a public education campaign instead of a mandatory disclosure suggests that the Court may increasingly insist that the government use its own funds to educate the public about the health risks of dangerous products rather than mandate warnings on the product. Even before NIFLA, the Supreme Court’s increasingly strict approach to commercial speech regulations led commentators to argue that the First Amendment had become a “powerful deregulatory engine.”144 In recent years, lower courts have upheld First Amendment challenges to a wide range of laws relating to NCDs, such as FDA regulations 136  471 U.S. 626, 651 (1985). 137  558 U.S. 310, 473 (2010). 138  564 U.S. 552, 580 (2011). 139  138 S.Ct. 2361, 2372 (2018). 140 Sarah C. Haan, The Post-­Truth First Amendment, 94 Ind. L. J. 1351 (2018), at http://ssrn.com/ abstract=http://ssrn.com/abstract=3209366 141  138 S.Ct. at 2377. 142  Id. at 2376. 143  Id. at 2379, 2380–2381 (Breyer, J., dissenting). 144  Amanda Shanor, The New Lochner, 2016 Wisc. L. Rev. 133, 134 (2016).

US Law Relating to Noncommunicable Diseases   141 requiring graphic warning labels on cigarette packages and advertisements.145 And as noted earlier, following NIFLA, the full US Court of Appeals for the Ninth Circuit found a San Francisco ordinance requiring warning labels on sugary sodas likely to violated the First Amendment.146 Another lower court blocked implementation of FDA regulations requiring larger health warnings on packages and advertisements for cigars and pipe tobacco.147 Many more such decisions are likely in the wake of NIFLA, leaving open the question whether a once common form of regulation remains constitutionally viable.

7 Conclusion The increased difficulty that US jurisdictions face in regulating commercial speech raises the question of how they will address NCDs in the years to come. One possibility is that, deprived of the ability to regulate the information that consumers receive about products, lawmakers will be more willing to regulate the products themselves. For example, rather than require menu labeling, jurisdictions could, without raising constitutional alarms, limit the sale of unhealthy menu items to children. Such approaches would be far more draconian and would undoubtedly provoke strong resistance for being paternalistic. But if implemented, they might well be more effective in protecting Americans’ health than are warning labels. For the time being, however, the widescale adoption of product controls appears unlikely. Nor are we apt to see many significant new initiatives coming from the federal government, at least as long as Congressional gridlock remains a fact of life in Washington, DC. Nevertheless, the diversity of US jurisdictions and laws means that other paths to regulation and innovation remain open. For example, private parties and state attorneys general are apt to continue to file civil claims against industries whose products are associated with NCDs. Some of these lawsuits, no doubt, will unearth damning information about product safety, which will in turn increase political pressures for regulation. Moreover, despite preemption, many local governments are likely to explore and experiment with new legal interventions directed at NCDs. Community groups and nongovernmental organizations are also likely to continue to organize and demand new laws that protect the public from dangerous products. In addition, the cost pressures associated with treating NCDs may push health insurers, employers, and other business interests to support some regulations to prevent NCDS. Although these efforts may not, in the short term, result in significant new federal initiatives, they may well pave the way for the next generation of NCD laws in the United States. If history is any guide, some of these innovations will “catch on” and eventually be adopted at the national level. 145  R. J. Reynolds Tobacco Co. v. U.S., 810 F.3d 827 (D.C. Cir. 2016). 146  American Beverage Ass’n v. City and Ct’y San Francisco, 916 F.3d 749 (9th Cir. 2019) (en banc). 147  Cigar Assoc. of America v. FDA, 2018 WL 3304627 (D.D.C. July 5, 2018).

chapter 7

The L ack of Coher ence i n th e Eu ropea n U n ion ’s A pproach to Noncom m u n ica bl e Disease Pr ev en tion Amandine Garde

1 Introduction More than half a million people under the age of 65 die of heart disease, kidney failure, and other noncommunicable diseases (NCDs) in the European Union (EU) each year.1 Approximately one-­third of the EU population aged 15 and older, and nearly a quarter of the working age population lives with a chronic disease.2 While progress is made on reducing premature mortality from NCDs, longer lives have not necessarily translated into healthier lives. A significant gap exists between life expectancy and healthy life expectancy,3 which is arguably all the more problematic as the European population is aging.

1  OECD/EU (2016), Health at a Glance: Europe 2016—State of Health in the EU Cycle. 2  Id.; C. Scaratti et al., ‘Mapping European Welfare Models; State of the Art of Strategies for Professional Integration and Reintegration of Persons with Chronic Diseases’ (2018) 15(4) Int J Environ Res Public Health 781. 3 Eurostat, Healthy Life Years (2020): https://ec.europa.eu/eurostat/statistics-­explained/index.php/ Healthy_life_years_statistics#Healthy_life_years_at_birth

144   Amandine Garde Premature mortality from NCDs costs €115 billion per year to the EU economy, whilst €700 billion is spent in the EU each year on treating NCDs. Moreover, absenteeism, presenteeism, and sick leave resulting from NCDs negatively impact on productivity. Overall, 1.7% of gross domestic product (GDP) in the EU is spent on disability and paid sick leave each year, which is more than the expenditure on unemployment benefits.4 Nevertheless, a significant proportion of the direct and indirect costs of NCDs could be saved through the development of effective NCD prevention policies, particularly those intended to reduce the consumption of tobacco, alcohol, and unhealthy diets as three major NCD risk factors. The EU’s involvement in NCD prevention is still relatively recent. While the European Community adopted a few measures affecting the tobacco, alcohol, and food industries in its early days,5 these measures were isolated and not envisaged primarily as public health protection measures. However, the situation started to change in the mid-­1990s, following the introduction of a chapter on public health in the EU Treaties which led to the adoption of several EU public health programmes.6 These programmes recognize the need to tackle health determinants through the creation of supportive environments for healthy lifestyles and disease prevention.7 It is understood in Europe that good health is a prerequisite to making the EU a smart, sustainable, and inclusive economy promoting growth for all.8 However, despite the general recognition that prevention is better than cure, the involvement of the EU has differed in nature, scope, and intensity depending on the risk factor at stake. If its regulatory intervention has been far-­reaching in relation to tobacco control, it has been much more ‘light touch’ in relation to the promotion of healthy diets and even minimal in relation to alcohol control. To facilitate comparisons with the corresponding US chapter where possible, this chapter follows a similar structure. Before comparing the EU regulatory response to tobacco on the one hand (Part 3), and unhealthy diets and alcohol on the other (Part 4), the chapter discusses the powers that the EU derives from the EU Treaties to address the three main commercial determinants of health through the adoption of EU-­wide harmonizing rules (Part 2). It also reflects on the much less interventionist approach of the Court of Justice of the European Union (CJEU) in legislative choices than its American counterparts and what this means for the prevention of NCDs in Europe (Part 5). It concludes with a few remarks on what the future may hold in this policy area now that 4  Health at a Glance, supra note 1. 5  These measures included food labelling rules and excise duties on tobacco products and alcoholic beverages. 6  Decision 1786/2002, Decision 1350/2007, and Regulation (EU) 282/2014 of the European Parliament and of the Council of the European Union. 7  Article 2(2) and point 2.2 of the Annex of Decision 1350/2007. See also the White Paper Together for Health: A Strategic Approach for the EU 2008–2013, COM(2007) 630 final. The Lisbon Agenda on Growth and Competitiveness further strengthened the economic and social case for EU intervention by stressing that, in addition to good health being a valuable goal in itself, it also leads to better economic results, increased social cohesion, and, consequently, makes the European economy more competitive: European Council Conclusions, Lisbon 23–24 March 2000. 8 https://ec.europa.eu/health/funding/programme_en

The EU’s Approach to Noncommunicable Disease Prevention   145 the new EU Commission in post has announced its ‘Beating Cancer Plan’ and its ‘Farm to Fork Strategy’ (Part 6).9

2  The Powers of the EU to Prevent NCDs Under the principle of conferral, the EU can only intervene in a policy area insofar as it has been granted the competence to do so by the EU Treaties.10 Therefore, competences not conferred on the EU remain with Member States.11 As public health policy within the European Union began to develop, the Member States agreed that the EU should be given the legal competence to act in this field to reflect the public health activities that had been taking place at the European level for some time.12 This competence, introduced in the Maastricht Treaty (Treaty on European Union [TEU]), was subsequently strengthened when demands were made of the EU to step up its efforts to contain bovine spongiform encephalopathy (BSE).13 The provisions of the EU Treaties dealing specifically with public health grant the EU the powers to ‘carry out actions to support, coordinate or supplement the actions of the Member States’14 and adopt ‘incentive measures designed to protect and improve human health and in particular to combat the major cross-­border health scourges . . . and measures which have as their direct objective the protection of public health regarding tobacco and the abuse of alcohol’.15 Notwithstanding this express reference to tobacco and alcohol, they specifically exclude ‘any harmonization of the laws and regulations of the Member States’. Consequently, they do not provide a specific legal basis for the adoption of legally binding measures, thus preventing the EU from adopting a fully-fledged strategy addressing all aspects of NCD prevention comprehensively. Nevertheless, the EU is also under a duty to take, as a base, a high level of public health protection in the development and implementation of all its policies and activities.16 This ‘mainstreaming’ obligation acts as a ‘Health in All Policies’17 provision and acknowledges 9  This chapter does not discuss firearms and opioids that have not given rise to significant debates at the EU level—these matters being left primarily to Member States. 10  The EU Treaties comprise the Treaty on the European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU). 11  Article 4(1) TEU. 12  T Hervey, ‘Mapping the Contours of European Union Health Law and Policy’ (2002) 8(1) European Public L 69, 72. 13  S Greer, ‘Uninvited Europeanization: Neofunctionalism and the EU in Health Policy’ (2006) 13(1) J Eur Public Policy 134, 139. 14  Article 6(a) TFEU. 15  Article 168 TFEU. 16  Articles 9 and 168(1) TFEU. 17  Public health mainstreaming was first seriously addressed at EU level during the Finnish Council Presidency in 2006, with the introduction of Health in All Policies, a strategic initiative that was intended to galvanize policy-­makers to consider health determinants controlled in sectors other than health.

146   Amandine Garde that health interests should be considered and given significant value when pursuing potentially competing goals in other policy areas, particularly those where the EU has the power to adopt legally binding measures. The requirement to ensure that health concerns are accommodated to a high standard in all policy areas is further reflected in the wording of Article 3 of the TEU, which sets out the broad objectives of the EU.18 It is all the more relevant in relation to NCD prevention which requires a coordinated, multisectoral response from policy actors at national, regional, and global levels. To date, most EU legislation relevant to NCD prevention has been adopted on the basis of Article 114 of the TFEU, which empowers the EU to adopt common rules for the establishment and the functioning of the internal market.19 These harmonizing rules can explicitly pursue the dual objective of regulating trade between EU Member States and ensuring a high level of public health protection. Health can even be a decisive factor in policy choices (though it cannot be the exclusive one).20 In fact, Article 114 itself explicitly mandates the Commission, when proposing harmonizing rules concerning health, safety, environmental, and consumer protection, to take as a base a high level of protection,21 further highlighting the importance of ‘Health in All Policies’. The contours of EU internal market powers have given rise to an extensive body of case law from the CJEU. Key lessons stem from this case law which delineates the scope of EU powers under this provision. In particular, Article 114 TFEU can only be relied upon if three conditions are fulfilled: - there must exist an ‘internal market barrier’ resulting from the disparities in the legal systems of the Member States; - this market barrier must not consist of an ‘abstract risk of obstacles’, but should be ‘such as to obstruct the fundamental freedoms’ or create ‘distortions of competition’ within the internal market; and - the intended harmonization should ‘genuinely have as its object the improvement of the conditions for the establishment and functioning of the internal market’.22 The CJEU has interpreted these conditions generously. However, a mere finding of disparities between national rules is not sufficient to justify recourse to Article 114: the 18  Paragraph 1 requires that the Union should promote ‘the well-­being of its peoples’—of which good health is a precondition, whilst Paragraph 3 calls on the EU to establish an internal market that ‘shall work for the sustainable development of Europe’, and shall ‘promote protection of the rights of the child’—which are negatively affected by tobacco use, alcohol consumption, and unhealthy diets. 19  Beyond the internal market powers the EU derives from Article 114 TFEU, it also has powers in the field of taxation, environmental and consumer protection, health and safety at work, and agricultural policy These could all be used more effectively to prevent NCDs and promote better health for all. 20  Opinion of AG Fennelly, in Tobacco advertising I, at paragraph 149. On this balancing, see Case C-­376/98 Germany v Council and EP (Tobacco Advertising I) [2000] ECR 1–8419 [88]; Case C-­380/03Germany v Council and EP (Tobacco Advertising II) [2006] ECR I-­11573, at paragraph 39; Joined Cases C-­154 and 155/04 Alliance for Natural Health [2005] ECR I-­6451, at paragraph 30; Case C-­547/14 Philip Morris, at paragraph 60. 21  At paragraph 3. 22  Case C-­58/08 Vodafone Ltd. [2010] ECR I-­4999, at paragraph 32.

The EU’s Approach to Noncommunicable Disease Prevention   147 differences between the laws, regulations, or administrative provisions of the Member States must be such as to obstruct the fundamental freedoms and thus have a direct effect on the functioning of the internal market.23 Consequently, and as discussed later, the EU can regulate product packaging or cross-­border marketing, such as broadcast or digital advertising, but it cannot regulate more static forms of advertising, such as advertising in cinemas or on billboards; nor can it impose EU-­wide smoking bans in public places or limit the presence of fast food outlets in the vicinity of schools. Overall, therefore, the EU Treaties call on the EU to ensure a high level of public health protection in all its policies, particularly its internal market policy, whilst they also circumscribe the powers of the EU to adopt harmonizing public health measures. As a result, the EU has had no option but to resort extensively to Article 114 TFEU to pursue public health objectives and prevent NCDs more specifically, thus revealing the ambiguity of the EU as a federal project and the lack of clarity of the division of competences between the EU and its Member States.24

3  Towards a Comprehensive EU Tobacco Control Policy? Despite considerable progress made in recent years, the number of smokers in the EU is still high: 26% of the overall population and 29% of young Europeans aged 15–24 smoke. Tobacco use remains the leading preventable cause of cancer mortality in the EU and the most significant cause of premature death in the EU. Around 50% of smokers die prematurely (on average 14 years earlier),25 and 27% of all cancer deaths are linked to smoking, not least lung cancer—90% of which are directly related to tobacco smoking— as the most common cause of cancer death among men and the second most common among women.26 In 2017, the economic costs of smoking were estimated at 4.6% of EU GDP.27 Very aware of the harm smoking causes and determined to address it, the EU has adopted various harmonizing measures intended to limit the uptake of smoking throughout its territory (Part 4.1), although the limited public health powers it derives from the EU Treaties together with diverging views from Member States lead to an unavoidable degree of diversity of tobacco control laws from one country to another (Part 4.2).

23  Case C-­547/14 Philip Morris Brands and Others (Philip Morris) [2016] ECLI:EU:C:2016:325, at paragraph 58. 24  V Delhomme, ‘Between Market Integration and Public Health: The Paradoxical EU Competence to Regulate Tobacco Consumption’, Research Paper in Law 01/2018 (College of Europe 2018). 25 https://ec.europa.eu/health/tobacco/overview_en 26  Lung cancer in the EU accounted for some 273,000 deaths out of 1,320,000 deaths from cancer in total in 2015. OECD/EU, Health at a Glance: Europe 2018. 27  Statement by Commissioner Kyriakides ahead of World No Tobacco Day, 29 May 2020: https:// ec.europa.eu/cyprus/news/20200529_3_en.

148   Amandine Garde

3.1  The EU’s Strong Regulatory Intervention to Nudge Consumers Away from Smoking Over the years, the EU has adopted various tobacco control measures, incrementally tightening its regional grip on the tobacco industry. In 2005, the EU also ratified the World Health Organization (WHO) Framework Convention on Tobacco Control (FCTC), which calls for tobacco tax increases, protections on exposure to second-­hand smoke, and other measures, thus confirming the EU’s status as a major actor on the public health scene at the global level. The FCTC’s full implementation is integral to the commitments states have made to achieving a reduction in premature deaths from NCDs, including a 30% relative reduction in the prevalence of tobacco use in persons aged 15 years and older by 2025.28 The two main building blocks of the EU’s regulatory tobacco control arsenal are the Tobacco Advertising Directive and the Tobacco Products Directive. They are based on Article 114 of the TFEU, and both have a long and controversial history.29 The Tobacco Advertising Directive imposes an EU-­wide ban on cross-­border tobacco advertising and sponsorship in all media.30 The Tobacco Products Directive lays down diverse rules governing the manufacture, presentation, and sale of tobacco and related products.31 In particular, it • bans characterizing flavours in cigarettes and roll-­your-­own tobacco products; • mandates combined health warnings covering 65% of the front and back of cigarette and roll-­your-­own tobacco packages; • bans promotional and misleading claims on tobacco products, e-­cigarettes, and herbal products for smoking (e.g., ‘light’, ‘mild’ . . .); • introduces EU-­wide tracking and tracing to combat the illicit trade of tobacco products; • sets out safety, quality, and notification requirements for electronic cigarettes; and • introduces extensive notification and reporting requirements on the tobacco industry.

28  Resolution WHA 66.10 adopting the WHO Global Action Plan for the Prevention and Control of NCDs 2013–2020, at para 36. 29  On the various methods that the tobacco industry has used against its regulation, particularly in the EU, see the excellent Tobacco Tactics database run by the University of Bath: http://www. tobaccotactics.org. 30  Directive 2003/33/EC. Only publications intended for professionals in the tobacco trade and publications from non-­EU countries which are not principally intended for the EU market are exempt. Also note that the Tobacco Advertising Directive does not apply to advertising and sponsorship on television and other audiovisual media services which are covered by the Audio-­visual Media Services Directive: Directive 2010/13/EU, as recently amended by Directive 2018/1808/EU. 31  Directive 2014/40/EU. Tobacco products include cigarettes, roll-­your-­own tobacco, pipe tobacco, cigars, cigarillos, smokeless tobacco, electronic cigarettes, and herbal products for smoking.

The EU’s Approach to Noncommunicable Disease Prevention   149 The EU’s rationale for the extension of the scope of the Tobacco Products Directive was to address internal market fragmentation in light of new scientific evidence, in line with the EU’s international obligations under the FCTC and its mandate under Article 114(3) of the TFEU to take as a base a high level of health protection, considering in particular any new development based on scientific facts.32 The EU has also adopted rules establishing minimum excise duties on tobacco products.33 Overall, the paradigm underpinning EU tobacco control policy has increasingly consisted in ‘nudging’ people, particularly young people, away from temptation. The policy of informing consumers of the risks associated with smoking through mandatory textual and pictorial health warnings is complemented by measures limiting the use of certain ingredients in tobacco products, thus limiting the placement of some products on the EU market. In particular, the latest EU-­wide tobacco control measure to take effect bans the use of cigarettes and roll-­your-­own tobacco with a characterizing flavour other than one of tobacco (e.g., fruit, spice, herbs, alcohol, candy, menthol, or vanilla) on the basis that such products can facilitate the initiation of tobacco consumption or affect consumption patterns.34 The Tobacco Products Directive even prohibits manufacturers from information about the nicotine, tar, or carbon monoxide content of the tobacco product’,35 the idea being to nudge smokers towards believing that tobacco products are all equally dangerous.

3.2  The Relationship Between EU and National Tobacco Control Laws Even if the scope of EU powers has been interpreted extensively—and the EU has been able to implement several FCTC provisions at regional level—the fact remains that the EU cannot regulate tobacco products and commercial practices comprehensively alone: it is only if Member States also regulate tobacco products at the national level that the FCTC can be fully implemented in the EU. For example, the EU-­wide ban on all forms of cross-­ border advertising and sponsorship has been complemented by various 32  When the Tobacco Products Directive was challenged, the CJEU attached great probative weight to the FCTC and to the WHO Guidelines which were intended to assist the contracting parties (which include the EU and all its Member States) in implementing the binding provisions of the Convention. Even though the Guidelines are nonbinding, they were drawn up on the basis of the very best scientific evidence, reflected an international consensus, and are therefore capable of exerting ‘decisive influence’ and must be recognized as being of ‘particularly high evidential value’: see Case C-­547/14 Philip Morris, at paragraphs 111–113 and 175. 33  Article 113 TFEU allows the EU to adopt common rules harmonizing the laws of the Member States on direct taxation. On the harmonization of excise duties of tobacco products, see Council Directive 2011/64/EU. 34  Article 7(1) and Recital 16 of the Tobacco Products Directive. On the interpretation of Article 7, see Case C-­220/17 Planta Tabak-­Manufaktur Dr Manfred Obermann Gmbh & Co. KG v Land Berlin, ECLI:EU:C:2019:76. 35  Article 13.

150   Amandine Garde national restrictions on forms of advertising and sponsorship arrangements that the EU cannot regulate itself due to its limited public health powers. The CJEU has ruled that the EU cannot regulate static advertising (e.g., advertisements in hotels, on billboards, umbrellas, ashtrays, and similar items), advertisements screened in cinemas, and the sponsorship of events that do not have any cross-­border appeal on the basis of Article 114 of the TFEU.36 These issues are therefore left to each Member State. Some of them have been very proactive in restricting tobacco marketing. In particular, Slovenia has banned all forms of direct and indirect advertising, whilst several others are very close to having complete tobacco advertising bans in place, including Croatia, Cyprus, Finland, France, the Netherlands, and the UK.37 However, other EU Member States have not led by example and failed to regulate tobacco advertising bans beyond what EU rules imposed. Most notably, Germany, which notoriously opposed the adoption of common EU rules on tobacco advertising and sponsorship,38 still allows tobacco companies to advertise their products in cinemas and outdoors.39 This is particularly problematic bearing in mind that, when bans are not comprehensive, tobacco companies exploit legal loopholes or simply shift their investments to forms of promotion that are not banned 164, 175, 176. The EU has also adopted ‘soft law’ provisions in policy areas where it does not have the conferred powers to adopt EU-­wide harmonizing legislation. Therefore, the Tobacco Products and the Tobacco Advertising Directives have been complemented by recommendations to Member States, not least the 2009 Council Recommendation on ­smoke-­free environments,40 as well as EU-­wide anti-­smoking campaigns. Nevertheless, these Recommendations have been implemented differently from one Member State to another.41 Moreover, political will has not always been sufficient to ensure the adoption of all the tobacco control measures that the EU would have had the necessary powers to adopt under Article 114 to reduce smoking rates across the EU. Some Member States have therefore used the discretion they derive from EU law to fill in some of these outstanding gaps. For example, the Tobacco Products Directive grants some freedom to Member States to regulate the presentation and labelling of tobacco products at the national ­level.42 The debates surrounding the implementation by some EU Member States of 36  Case C-­376/98 Germany v Council and the European Parliament [2000] ECR I-­8419, in which the CJEU annulled the First Tobacco Advertising Directive: Directive 98/43. The EU subsequently adopted Directive 2003/33, the Tobacco Advertising Directive discussed earlier, whose validity was upheld by the Court: Case C-­380/03 Germany v Council and the European Parliament [2006] ECR I-­11573. 37  WHO Report on the Global Tobacco Epidemic (WHO 2019), 104. 38 See supra, note 36. 39  The German government recently announced it would change the law. https://tobaccoreporter. com/2020/07/06/germany-­to-­curb-­tobacco-­advertising/. 40  OJ 2009 C 296/4. See also Council Recommendation of 2 December 2002 on the prevention of smoking and on initiatives to improve tobacco control [2003] OJ L 22/31. 41  In 2013, the European Commission reported that 17 EU countries had comprehensive smoke-­free legislation in place: SWD(2013) 56 final/2. 42  Article 24.

The EU’s Approach to Noncommunicable Disease Prevention   151 tobacco plain packaging schemes has put in sharp focus the question of the scope of this freedom. In their challenge to the validity of both the UK Tobacco Plain Packaging Regulations and the Tobacco Products Directive, Philip Morris and other tobacco manufacturers called on the CJEU to ‘clarify’ the regulatory discretion Member States retained to adopt national laws exceeding the requirements of the Directive. The Court upheld the validity of the Directive. In particular, it ruled that the Directive did not harmonize all aspects of tobacco products packaging (it is a measure of partial as opposed to full harmonization) and that Member States could therefore adopt national measures imposing the plain packaging of tobacco products under Article 24(2) of the Directive. Even though the wording of Article 24(2) does not contain a specific statement to that effect—the Court noted—‘such an interpretation is consonant with the objective and general scheme of the directive’.43 Ultimately, the EU could have gone further but preferred not to. In the absence of an EU-­wide tobacco plain packaging scheme, it has allowed Member States to adopt such schemes at national level, which the UK, France, Ireland, Slovenia, Hungary, Belgium, and the Netherlands have done.44 Finally, the reliance on the technique of minimum harmonization has also contributed to the differences among national tobacco control policies. This approach aims to set the minimum standard of protection with which all Member States must comply. It can therefore be a useful regulatory mechanism to strike a compromise when Member States hold different views. On the one hand, the minimum requirements bind all parties and establish a regional level playing field within which all business actors must operate. This, in turn, limits unfair competition by making it clear that some commercial ­practices are not acceptable anywhere in the EU. On the other hand, a minimum harmonization approach does not stifle national initiatives in countries where there is enough political will to adopt more ambitious tobacco control strategies. Nevertheless, such an approach only makes sense if the level of minimum protection it achieves is sufficient to protect public health effectively. The health mainstreaming obligations resting on the EU mandates that minimum harmonization shall not be envisaged as minimal harmonization. The issue is strikingly illustrated by the EU’s imposition of minimum excise duties for tobacco products.45 The prices of tobacco products vary significantly from one Member State to another (the average price of a 43  Case C-­547/14 Philip Morris ECLI:EU:C:2016:325, at paragraphs 38, 73. This means that if a national scheme was challenged under Articles 34–36 TFEU, the CJEU would be likely to rule that plain packaging is a Measure of Equivalent Effect to a Quantitative Restriction (MEEQR) that can be justified on public health grounds under Article 36 TFEU. In this sense, the CJEU has preempted the outcome of challenges based on the free movement provisions regarding the compatibility of national tobacco control measures relating to aspects of the packaging which are not regulated by the Second Tobacco Products Directive. In so doing, it has maintained the rather broad margin of discretion Member States have under Article 36 TFEU. 44  Similarly, several aspects of the regulation of e-­cigarettes have been left to Member States. This is not discussed in this contribution due to space constraints, but see L Gruszczynski ed., The Regulation of E-­Cigarettes: International, European and National Challenges (Edward Elgar 2019), particularly ch 7: A Pudlo, L Gruszczynski, ‘Regulating E-­Cigarettes at the EU Level’. 45  Directive 2011/64.

152   Amandine Garde pack of cigarettes ranging from €2.57 to €11.37) and represent a sufficient economic incentive for unintended high levels of cross-­border shopping. In most cases, higher taxes on tobacco products lead to higher prices which, in turn, lead to lower consumption and prevalence and result in a reduction of mortality and morbidity and thus in the improved health of the population. The inverse relationship between price and tobacco use has been demonstrated by numerous studies and is not contested.46 Therefore, as an FCTC Party, the EU should ‘recognize that price and tax measures are an effective and important means of reducing tobacco consumption by various segments of the population, in particular young persons’47 and amend its regulatory framework accordingly. There are signs that a reform of the EU regulatory framework on tobacco taxation may be forthcoming. In February 2020, the European Commission published an evaluation of its functioning where it concluded that it was not effective at deterring consumption.48 Moreover, the launch by the new Commission of its Europe’s Beating Cancer Plan highlights the pivotal role of taxation in reducing tobacco consumption and deterring young people from smoking.49 These differences between national tobacco control laws and the many reasons explaining them should not, however, detract from the conclusion that progress has been significant in this policy area over the past 20 years. As a result of the EU’s strong regulatory involvement in this area, tobacco control has been at the forefront of a ‘federal’ experimentation helping to delineate the scope of EU competences as well as the relevance of the principles of subsidiarity and proportionality for EU law and ­ policy-­making.50

4  The EU’s Stubborn Reluctance to Regulate the Alcohol and Food Industries In contrast with its approach to tobacco control, the EU has been extremely reluctant, overall, to regulate the alcohol and food industries. As a result of the paucity of EU rules 46  Guidelines to Article 6 FCTC, at paragraph 2. 47  Article 6(1) FCTC. See also Guidelines to Article 6, at paragraph 2.1 (referring to estimates ‘that young people are two to three times more responsive to tax and price changes than older people’. 48  SWD(2020) 32 final (also highlighting ‘the lack of any explicit quantifiable public health objective in the Directive,’ page 64). 49  https://ec.europa.eu/info/law/better-­regulation/have-­your-­say/ initiatives/12154-­Europe-­s-­Beating-­Cancer-­Plan 50  A Alemanno, A Garde, Regulating Lifestyles in Europe How to Prevent and Control Non-­ Communicable Diseases Associated with Tobacco, Alcohol and Unhealthy Diets? (Swedish Institute of European Policy Studies 2013) 19.

The EU’s Approach to Noncommunicable Disease Prevention   153 intended to improve food environments, regulatory diversity across Member State is even sharper in this policy area than for tobacco control. This is not to say, however, that food and alcohol have not featured prominently in the EU’s public health discourse. Alcohol-­related harm is a major public health concern in the EU, accountable for more than 7% of all ill health and early deaths,51 whilst unhealthy diets have led to rising rates of obesity. More than half of the adult population are now overweight, contributing to a high prevalence of diet-­related NCDs and related healthcare costs. It is estimated that, in the EU in 2017, more than 16 million lost healthy life years were attributable to unhealthy diets, mainly through cardiovascular diseases and cancers.52 Following several Council Conclusions,53 the EU recognized that alcohol consumption and unhealthy diets were harmful to EU citizens and the EU economy and needed to be addressed. The EU adopted the Alcohol Strategy 2006–2012 in 2006, shortly before publishing the Obesity Prevention White Paper in 2007, followed seven years later by the EU Action Plan on Childhood Obesity 2014–2020.54 Even though these strategies are not legally binding, they have set the tone and indicated what the EU is willing to prioritize and the means to achieve its objectives. As it has done for tobacco control, the EU has recognized the importance of implementing coordinated, multisectoral strategies to address the harm stemming from unhealthy diets and alcohol consumption. Nevertheless, this is where the analogy stops between the EU’s approach to tobacco control and its approach to unhealthy diets and alcohol consumption. The few legislative measures that the EU has adopted in relation to food and alcohol have consisted nearly exclusively in information measures intended to ‘empower’ consumers to make healthier choices. Hardly any legislation has been adopted to more radically change the environment in which choices are made to promote the availability, accessibility, and affordability of healthier food whilst reducing that of unhealthy food and alcoholic beverages (Part 3.1). Moreover, the EU has engaged extensively with the food and alcohol industries, preferring to rely on the adoption of ‘pledges’ and other voluntary commitments by industry actors, alongside the exchange of best practice between Member States, rather than the adoption of a comprehensive, legally binding NCD prevention strategy at EU level (Part 3.2). This, in turn, has increased the burden on Member States wishing to regulate more effectively the food and alcohol industries to defend legal changes based on EU internal market law (Part 3.3).

51  https://ec.europa.eu/youth/policy/youth-­strategy/health-­wellbeing_en 52  COM(2020) 381 final, at pages 3, 13. 53  Some of these calls have focused specifically on one risk factor, whilst others have tended to be more horizontal in nature. Examples of the latter type include the Council Conclusions of December 2003 on Healthy Lifestyles; the Council Conclusions of June 2004 on Promoting Heart Health; the Council Conclusions of June 2006 on the Promotion of Healthy Lifestyles and the Prevention of Type II Diabetes. 54  COM(2006) 625 final; COM(2007) 279 final; https://ec.europa.eu/health/sites/health/files/ nutrition_physical_activity/docs/childhoodobesity_actionplan_2014_2020_en.pdf

154   Amandine Garde

4.1  The Food Information Paradigm Requiring that consumers be provided with specific information about a product or serv­ice is a regulatory technique that has proven popular in the development of EU measures affecting the protection of consumers’ interests and their health, as it has in the United States. It places the onus on consumers to decide what is good for them and their families, expecting them to take their personal circumstances into account. The approach of improving transparency by providing enough relevant information has the advantage of minimizing interference with individual choices. The provision of information is therefore seen as a compromise: protection is provided as a result of the introduction of duties on businesses to inform consumers of the qualities of their goods and services while avoiding intrusive controls, such as bans on particular types of contract, which may unduly diminish consumer choice.55 As regards the consumption of food and alcohol more specifically, the assumption is that if consumers are aware of what food (defined in the EU to include alcoholic beverages56) contains, they should be trusted to apply their knowledge and make appropriate choices. Rather than require the reformulation of a given food or ban its marketing due to its high sugar, fat, or salt content, or increase its price through taxation, consumers are informed about this content so that they can decide whether to buy it and integrate it into their diet. This reflects the view that in a society where individuals are autonomous and dislike limitations, they must take primary responsibility for the consumption choices they make on their own and their children’s behalf. It also explains the focus which the EU has placed on the importance of providing information to consumers.57 The regulation of how the information is provided and what it should contain allows public authorities to support individual purchasing decisions without interfering too much with consumers’ personal freedom to buy what they want and with business operators’ freedom to respond to, or even shape, consumer demand through advertising.58 However, the information paradigm promoted at the EU level requires that the information made available to consumers be of sufficient quality to guide their choices and effectively allow them to ‘protect’ themselves. The responsibility for healthy choices is therefore shared primarily between consumers on the one hand, who are expected to process the information made available to them when purchasing goods and services, and regulatory authorities on the other, which must ensure that enough information is made available to consumers and that such information is neither false nor misleading.59 Thus, the Food Information Regulation and the Food Claims Regulation are intended to ensure that the information on food provided to consumers 55  S Weatherill, EU Consumer Law and Policy (Edward Elgar 2005) at 84. 56  Article 2 of Regulation (EC) 178/2002. 57  This is clearly noted in the Commission’s White Paper laying down a strategy for Europe on nutrition, overweight, and obesity-­related health issues, COM(2007) 279 final. 58  Id. at 5. 59  The food industry also has a role to play by complying with disclosure requirements laid down by law and ensuring that the information it provides to consumers is not misleading.

The EU’s Approach to Noncommunicable Disease Prevention   155 is clear, sufficient, and trustworthy and, as such, these constitute major components of the EU’s nutrition strategy.60 The Food Information Regulation requires the disclosure of information intended to help consumers make ‘informed’ food choices, referring specifically to the list of ingredients and the nutrition declaration. It also imposes legibility and other presentation requirements.61 As such, it is very much in line with the ‘information paradigm’ at the heart of EU health and consumer protection strategies. However, while the Food Information Regulation does require, as its predecessor did, that alcoholic beverages containing more than 1.2% by volume of alcohol should indicate their actual alcoholic strength by volume, it exempts them from the disclosure obligation that consumers should be informed of the list of ingredients and the nutrition declaration on the basis of their ‘specific nature’.62 The Commission has instead called on industry operators to provide voluntary nutrition information.63 This exemption has been vividly criticized as it fails to provide the level playing field required both to promote the proper functioning of the internal market and ensure a high level of public health protection. The debate is ongoing.64 The Food Claims Regulation regulates the information provided voluntarily by food operators on their products, specifically nutrition claims and health claims. It defines a claim as ‘any message or representation, which is not mandatory under [Union] or national legislation, including pictorial, graphic or symbolic representation, in any form, which states, suggests or implies that a food has particular characteristics’.65 As claims are voluntarily provided by food operators, they must be distinguished from the information disclosed as part of a legal requirement, such as the nutritional declaration provided for in the Food Information Regulation. Food claims are often used by industry operators as a means to promote the characteristics of the foods they have placed on the market, thus constituting a potentially powerful marketing tool to distinguish their goods from competing goods and influence consumer behaviour. However, claims may ‘encourage consumers to make choices which directly influence their total intake of individual nutrients or other substances in a way which would run counter to scientific advice’.66 Consequently, the Food Claims Regulation lays down the overriding principle that nutrition and health claims may only be used if they are not misleading and if they are scientifically substantiated. Nutrition claims are only permitted if they are listed in the Annex and if they respect precise and quantifiable values. For example, a claim that a food is ‘low in sugars’ may only be made where the product contains no more than 5 g of sugars per 100 g for solids or 2.5 g of sugars per 100 mL for 60  Commission’s White Paper laying down a strategy for Europe on nutrition, overweight, and obesity-­related health issues, COM(2007) 279 final, 5. 61  Regulation (EU) 1169/2011, Articles 9(1), 13, and 29 to 35. 62  Id. at Articles 9(1)(k) and 16(4) and Recital 40. 63  COM(2017) 58 final. 64 https://ec.europa.eu/food/safety/labelling_nutrition/labelling_legislation/alcohol_en 65  Regulation (EC) No 1924/2006, Article 2(2). The definitions of ‘nutrition’ and ‘health’ claims are provided in Articles 2(4) and 2(5) respectively. 66  Regulation (EC) No 2004/2006, Recital 10 of the Food Claims Regulation

156   Amandine Garde liquids.67 Health claims are only allowed if the producer can verify the link between the claim and a given product (e.g., ‘calcium is required for the maintenance of normal bones’). The Commission has established an EU register of nutrition and health claims made on food which is regularly updated to include the list of permitted nutrition and health claims together with the conditions applying to them, as well as a list of rejected health claims and the reasons for their rejection.68 As far as alcohol more specifically is concerned, the EU has banned the use of nutrition and health claims on alcoholic beverages of more than 1.2% by volume of alcohol.69 As discussed later, this prohibition was challenged both during the legislative process that led to the adoption of the Food Claims Regulation and after its adoption before the CJEU. Importantly, the Food Claims Regulation provides the only example where the EU has effectively recognized that ‘alcoholic beverages represent a special category of foods that is subject to particularly strict regulation’.70 Even though the regulation of food claims is welcome to ensure that the information provided to consumers on food, including alcohol, is not misleading, the EU’s regulatory framework remains largely insufficient to address NCDs stemming from unhealthy diets and alcohol consumption. The EU could (and should) do more to protect EU citizens from the harmful use of alcohol and unhealthy diets considering the significant cross-­border trade in food and beverages. This is particularly so as the information measures that the EU has adopted are neither comprehensive nor able, on their own, to promote healthier food environments. Too much information may result in information overload, which may in turn lead to failures on the part of consumers (1) to pay attention to the most relevant information, (2) to process this information, and (3) to act on it. Furthermore, the notion of ‘consumer empowerment’ may not be helpful to certain groups of particularly vulnerable consumers who, due to their age, disability, lack of education, or high credulity, cannot process the information made available to them through disclosure requirements. This is particularly true when addictive substances are involved.71 More fundamentally, behavioural research, by highlighting the complex cognitive framework in which people make decisions, increasingly questions the ability of back-­of-­pack nutrition labelling schemes to effectively inform decisions. After documenting the systematic errors people make in thinking (cognitive biases) and tracing those errors to the ‘design of the machin-

67  Regulation (EC) No 1924/2006, Recital 9, Articles 3–6, 8(1), and Annex. Amendments to the Annex may be made subject, where appropriate, to the consultation of EFSA. To date, six nutrition claims have been added to the original list of 24. 68 http://ec.europa.eu/nuhclaims/ 69  Article 4(3). This provision exempts nutrition claims which refer to a reduction in the alcohol or energy content from the scope of the prohibition. 70  Case C-­544/10 Deutsches Weintor [2012] ECLI:EU:C:2012:526, at paragraph 48 (emphasis added). 71  For example, the WHO Global Strategy to Reduce the Harmful Use of Alcohol clearly states: ‘Some vulnerable or at-­risk groups and individuals have increased susceptibility to the toxic, psychoactive and dependence-­producing properties of ethanol’ (at paragraph 5).

The EU’s Approach to Noncommunicable Disease Prevention   157 ery of cognition’,72 behavioural research shows that individual choices are conditioned by, inter alia, context (e.g., the shelf where the product is placed), social influence (e.g., the eating and drinking habits of the people you socialize with), and framing (e.g., how the nutritional information is conveyed). In particular, evidence suggests that salient, vivid, and colourful indications are more effective than statistical and abstract information sets and that front-­of-­pack nutrition labelling with interpretational aides (e.g., words, colours, or symbols) is more likely to be used and understood by consumers.73 Consequently, the WHO European Food and Nutrition Action Plan 2015–2020 identified the introduction of interpretative, consumer-­friendly front-­of-­pack nutrition labelling as a priority policy issue for European states. Many attempts have recently been made at the national level to turn nutritional information into simplified front-­of-­pack nutrition labelling schemes.74 However, few countries have interpretive systems in place that provide judgments about the relative healthfulness of food. Moreover, despite the growing evidence that front-­of-­pack nutrition labelling is far more likely to influence consumer choices and therefore contribute to public health objectives and the ­cross-­border dimension of food packaging regulation, the EU has not adopted an ­EU-­wide front-­of-­pack nutritional labelling scheme to date—though the Commission has recently acknowledged that it would be appropriate if it did.75 This is all the more problematic as the EU Food Information Regulation and the EU Food Claims Regulation restrict the freedom of Member States to adopt mandatory front-­of-­pack nutrition labelling schemes at the national level: they can only endorse voluntary schemes provided they comply with various conditions.76 Even if information may work, there are external impediments—such as lack of availability or price—which could prevent consumers from acting on such information. Nevertheless, beyond the adoption of (incomplete) information rules, the EU has not responded, overall, to the repeated calls of civil society for evidence-­based regulatory interventions: rather than harmonize national laws, the EU has focused nearly ­exclusively on promoting self-­regulation and the exchange of best practice between Member States.77 72  See D Kahneman, A Twersky, ‘Prospect Theory: An Analysis of the Decision Under Risk’ (1979) 47 Econometrica 263. 73  B Kelly, J Jewell, What Is the Evidence on the Policy Specifications, Development Processes and Effectiveness of Existing Front-­of-­Pack Food Labelling Policies in the WHO European Region? (WHO 2018). 74  For the different front-­of-­pack nutrition labelling schemes in use in the EU Member States, see the Report from the Commission regarding the use of additional forms of expression and presentation of the nutrition declaration, COM(2020) 207 final. 75  Supra, note 74. 76  See, in particular, Article 35 of the Food Information Regulation. 77  The only significant exception is the recent adoption of an EU-­wide legal limit on the content of industrially produced trans-­fat (fats not naturally occurring in fat of animal origin) in food of 2 grams per 100 grams of fat to take effect by April 2021. This limit was adopted on the ground that trans-­fat increases the risk of heart disease more than any other nutrient: Commission Regulation (EU) 2019/649.

158   Amandine Garde

4.2  How Genuine Are the EU’s Efforts to Prevent NCDs at Regional Level? As Roger Magnusson has put it, ‘with large profits at stake, there is a struggle for regulatory control’,78 and history has shown that the tobacco, alcohol, and food industries deploy a broad range of corporate tactics, including portraying themselves as key ‘partners’ in the fight against NCD prevention. The European food and drink industry has used this tactic extensively. It is arguably not surprising bearing in mind its economic clout: not only is it the EU’s biggest manufacturing sector in terms of jobs and value added, but the EU also boasts an important surplus in trade in food with third countries.79 The active involvement of the food and drink industry in the development and implementation of EU food policy has been facilitated by the absence of any equivalent to Article 5(3) of the FCTC which mandates that ‘in setting and implementing their public health policies with respect to tobacco control, Parties shall act to protect these policies from commercial and other vested interests of the tobacco industry in accordance with national law’.80 The failure of the WHO to draw similar red lines regarding the role of the food and alcohol industries in promoting health81 has encouraged these industries to portray themselves as key players in the prevention of NCDs at national, regional, and global levels. Even though they are known to be a major contributor to the problem of growing rates of NCDs, they have made a range of voluntary pledges and thus purported themselves to be reactive and act faster than states, and even act in lieu of public ­regulatory intervention. This dynamic is particularly prevalent at the EU level. For more than a decade, the EU relied extensively on the EU Platform on Diet, Physical Activity, and Health, which was set up in 2005 and gathered a broad range of members from industry operators to consumer and public health organizations. To participate in the Platform’s activities, every one of its members had to commit to the adoption of at least one specific, concrete commitment to help address unhealthy diets. These commitments are published on a dedicated database.82 Similarly, the EU set up the EU 78  R Magnusson, ‘What’s Law Got to Do with It? Part 1: A Framework for Obesity Prevention’ (2008) 5 Australia N Zealand Health Policy 10. 79  In the past 10 years, EU food and drink exports have doubled, reaching more than €90 billion and contributing to a positive balance of almost €30 billion: https://ec.europa.eu/growth/sectors/food_en 80  As the British High Court noted, this provision takes as its starting point ‘what may fairly be described as an expression of profound distrust about the motives of the tobacco industry’ and ‘assumes a history of deliberate subversion by the industry of governmental health policies’: BAT and Others v. Secretary of State for Health [2016] EWHC 1169 (Admin), at paragraph 170. 81  Even the most recent UN Political Declaration on the Prevention and Control of NCDs calls on states and the international community to ‘[e]ngage with the private sector, taking into account national health priorities and objectives for its meaningful and effective contribution to the implementation of national responses to non-­communicable diseases in order to reach Sustainable Development Goal target 3.4 on non-­communicable diseases, while giving due regard to managing conflicts of interest’: Resolution 73/2 of 10 October 2018 (A/73/L.2), at paragraph 43. 82 https://ec.europa.eu/health/nutrition_physical_activity/platform_en

The EU’s Approach to Noncommunicable Disease Prevention   159 Alcohol and Health Forum in 2007 as the ‘cornerstone’ of the EU Alcohol Strategy. It was conceived as a ‘Forum for action’ and gathered a broad range of stakeholders ranging from industry and hospitality operators to consumer and public health organizations. The founding Charter of the Forum required its members to ‘devote an increasing level of effort’ to the commitments made within the Forum to reducing alcohol-­related harm, and to demonstrate how their commitments were contributing to reducing alcohol-­related harm in a ‘transparent, participatory and accountable way’.83 However, neither the EU Platform nor the EU Forum lived up to the expectations they had set. Supposedly comprised of ‘experts’, their membership actually had a disproportionately large number of industry representatives who sought to use their position to shift policy-­making towards weak, smokescreen interventions. This was all the more so as many commitments, particularly in relation to alcohol, related to education and information campaigns rather than to more meaningful changes to the environment. Ultimately, in 2015, all the public health nongovernmental organizations (NGOs) resigned from the Forum on the basis that it had failed to deliver the meaningful and durable contribution to addressing alcohol-­related harm that had been promised.84 The lack of any ambition in the EU Alcohol Strategy 2006–2012 to implement a Health in All Policies approach to alcohol control and the Commission’s subsequent failure to develop a new Strategy are astonishing. Unfortunately, this failure has not been compensated by the establishment of other fora coordinated by the European Commission to facilitate the exchange of best practice between the EU and Member States, such as the High Level Group on Nutrition85 and the Committee on National Alcohol Policy and Action.86 Public–private partnerships with the food and alcohol industries have inherent limits. Clear rules of engagement are needed at regional and global levels to ensure that real, potential, and perceived conflicts of interest are acknowledged and carefully managed.87 This is all the more fundamental as independent research over the years has established that voluntary pledges adopted by food and alcohol business actors to address the harm resulting from the consumption of alcohol and unhealthy food have proven ineffective. The EU should adopt laws and regulations establishing the level playing that the European food and drink industry needs to operate fairly in a globalized world whilst ensuring a high level of public health protection. The myriad of Council Conclusions recently adopted on the harm resulting from alcohol and unhealthy diets highlight how much these issues are identified as major

83 https://ec.europa.eu/health/alcohol/events_en#anchor1 84  https://www.euractiv.com/section/health-­consumers/news/health-­ngos-­walk-­out-­of-­eu­alcohol-­forum/ 85  http://ec.europa.eu/health/nutrition_physical_activity/high_level_group/index_en.htm. The High Level group and the Platform used to meet regularly to improve the coordination of their activities. 86 https://ec.europa.eu/health/alcohol/events_en#anchor0 87  Some WHO instruments use the language of conflicts of interest avoidance rather than conflicts of interest management.

160   Amandine Garde concerns across all EU Member States.88 It is regrettable that these concerns have still not yet led to the adoption of effective EU-­wide legislation. Member States are left with no option but to address unilaterally issues that could be addressed more effectively at the regional level as a result of their inherently cross-­border nature.

4.3  Leaving Member States to Defend Their Measures from Trade-­Related Challenges As a result of the paucity of EU rules intended to limit unhealthy diet and alcohol consumption, several Member States have had no option but to address related NCDs at a national level via the adoption of national laws. Where the EU does not exercise its internal market powers, Member States can introduce further NCD prevention measures subject to their obligation to comply with the general free movement provisions of the EU Treaties, and Articles 34–36 of the TFEU on the free movement of goods more specifically.89 As a result, several EU Member States have adopted national measures intended to limit the availability, affordability, and acceptability of alcoholic beverages and unhealthy food. For example, France, Finland, and Lithuania have adopted extensive restrictions on the marketing of alcoholic beverages; the UK was the first to adopt restrictions on the marketing of unhealthy food on television and is set to extend these restrictions to the online environment; and many Member States have also adopted measures intended to increase the price of unhealthy commodities, not least alcoholic and sugar-­sweetened beverages. However, the more robust and effective are states regulatory measures, the more Member States will have to prepare to meet the vigorous opposition of business actors that manufacture, distribute, and promote alcoholic beverages and unhealthy food. Although Member States are granted a broad margin of discretion under EU internal market law to adopt measures that may be trade-­restrictive in order to protect public health, the extent of this discretion is subject to the proportionality principle on which many free movement cases involving alcoholic beverages have been decided following legal challenges by alcohol manufacturers or distributors to national public health measures restricting the availability or marketing of alcoholic beverages.90 Member States bear the burden of demonstrating that they have paid due regard to free movement and 88  See, in particular, Council Conclusions to contribute towards halting the rise in Childhood Overweight and Obesity of 15–16 June 2017 (Maltese Presidency), Council Conclusions on cross-­border aspects in alcohol policy of 8 December 2017 (Estonian Presidency), and Council Conclusions on healthy child nutrition of 22 June 2018 (Bulgarian Presidency). 89  The CJEU established this principle in its seminal judgment in Case 120/78 Rewe-­Zentrale AG v Bundesmonopolverwaltung für Brantwein [1979] ECR 649. 90  Well-­known legal challenges to national rules concerning the sale and distribution of alcoholic beverages include Case C-­189/95 Franzen ECLI:EU:C:1997:504, [1997] ECR I–5909; Case C-­405/98 Gourmet [2001] ECLI:EU:C:2001:135; Case C-­170/04 Rosengren ECLI:EU:C:2007:313; and Case C-­333/14 Scotch Whisky ECLI:EU:C:2015:845.

The EU’s Approach to Noncommunicable Disease Prevention   161 that the restrictions they have imposed on public health grounds are proportionate. This may be onerous, and one of the most important difficulties facing EU Member States remains the cost and uncertainty involved in defending the adoption of strong NCD prevention measures intended to curb the consumption of alcohol and unhealthy diets, which in turn may lead to significant delay in the implementation of the national measures in question.91 Overall, the EU has not responded effectively and coherently to the challenge of growing rates of NCDs because the cross-­border impact of multinational food and drink companies remains insufficiently addressed at the regional level. This approach, which is very much at odds with the approach that the EU has adopted for tobacco products, gives rise to a tension between what is expected from Member States (who are primarily responsible for the health of their citizens) and the support that the EU has provided them in trying to meet the commitments they have made to reduce alcohol-­ related harm and end childhood obesity (by failing to regulate harmful commercial practices with cross-­border impact). However, the constitutional principles at the heart of EU Treaties do not account for this failure as they would allow for a much more robust EU regulatory response to NCDs. Not only does the EU have extensive powers to adopt harmonizing internal market rules regulating intra-­EU trade in alcohol and food, but it has a broad margin of discretion (and even a duty) to use them to promote a high level of public health protection in the development and implementation of all its policies.

5  The CJEU’s Consistent Rejection of the Challenges to EU NCD Prevention Measures as Infringing the Fundamental Rights of ‘Vector of Disease’ Industries Industry actors have systematically mounted judicial review challenges against ­EU-­wide measures likely to reduce their profit margins. Tobacco manufacturers were the first to argue, when challenging tobacco control legislation, that EU harmonizing rules regulating the content, presentation, or promotion of their products infringed the fundamental rights they derived from EU law, not least their freedom of (commercial) expression and information, their freedom to conduct a business, and their right to (intellectual) property. Since then, the alcohol industry has joined the ‘vectors of disease’ chorus and mounted similar challenges, alleging infringements of fundamental rights. These claims have never succeeded before the CJEU: the margin of discretion that the Court is 91  It is only six years after Scotland adopted legislation imposing a minimum unit price on alcoholic beverages that this legislation entered into force.

162   Amandine Garde prepared to grant the EU legislature is significantly more extensive than what the US Supreme Court has done. The rights of the tobacco and alcohol industries to manufacture, sell, and promote their goods, services, and brands are not absolute and can be limited on grounds of public health protection. This is all the more so as, in such judicial review cases, the starting point of the CJEU does not have to be whether a given measure disproportionately impedes the principle of free movement, as in the cases discussed above involving the compatibility of national NCD prevention measures with the general free movement provisions of the EU Treaties.

5.1  Freedom of Commercial Expression The tobacco industry has repeatedly claimed before the CJEU that the imposition of tobacco marketing restrictions infringes the right to free expression. In particular, British American Tobacco challenged the first Tobacco Products Directive which imposed an EU-­wide ban on the use of texts, names, trademarks, and figurative or other signs on tobacco products which suggest that a particular tobacco product is less harmful than others (e.g., ‘light’, ‘mild’ . . .).92 The Court dismissed the challenge and noted that these descriptors were liable to mislead consumers, not least because ‘the use of descriptions which suggest that consumption of a certain tobacco product is beneficial to health, compared with other tobacco products, is liable to encourage smoking’, and individuals need to be given objective information concerning the toxicity of tobacco products.93 Similarly, when challenged by Germany, the CJEU dismissed the argument that the Tobacco Advertising Directive constituted an unlawful interference with freedom of expression. After recalling its settled case law that the EU legislature should be granted a broad margin of discretion in areas entailing political, economic, and social choices on its part, and in which it was called upon to undertake complex assessments, the Court concluded that the measures under review banning cross-­border tobacco advertising and sponsorship were not disproportionate. In its judgment, the Court relied explicitly on the case law of the European Court on Human Rights (ECtHR) on Article 10 of the European Convention on Human Rights (ECHR).94 After upholding the principle of freedom of expression as a general principle of EU law, it stated that the freedom of individuals to promote commercial activities derived not only from their right to engage in economic activities and the general commitment, in the EU context, to a market economy based on free competition, but also from their inherent entitlement as human beings freely to express and receive views on any topic, including the merits of the goods 92  Directive 2001/37. 93  Case C-­491/01 British American Tobacco and Imperial Tobacco [2002] ECR I-­11453. 94  Case C-­380/03 Germany v Council and the European Parliament [2006] ECR I-­11573 (citing Markt Intern v Germany Series A no 165 (1990) 12 EHRR 161; Groppera v Switzerland Series A no 173 (1990) 12 EHRR 321; and Casado Coca v Spain Series A no 285 (1994) 18 EHRR 1.

The EU’s Approach to Noncommunicable Disease Prevention   163 or services which they market or purchase.95 Nevertheless, the Court also explicitly stated that commercial expression was a lesser form of expression than political or artistic expression, that it could therefore be restricted on public health grounds, and that the EU legislature should retain a broad margin of discretion in determining what was legitimate and necessary to protect public health. Even though the outcome of these cases promotes public health, it is nonetheless regrettable that the Court did not use the opportunity to both challenge the information paradigm and explain why such restrictions were indeed proportionate.96

5.2  The Right to Property and the Freedom to Conduct a business The right to property and the freedom to conduct a business have often been invoked in tandem. The CJEU has highlighted that neither of those rights constitutes an unfettered prerogative but should be viewed in light of their social function and could be restricted, provided that the restrictions imposed correspond to objectives of general interest pursued by the EU and do not constitute a disproportionate and intolerable interference with the very substance of the rights thus guaranteed.97 In its British American Tobacco judgment, the CJEU dismissed the argument that the EU had unlawfully interfered with the right to property of tobacco manufacturers and their freedom to pursue a trade or profession by prohibiting the use of trademarks incorporating descriptors such as ‘light’ or ‘mild’. The Court noted that tobacco producers could continue to use other distinctive signs on the packs to distinguish its product by using other distinctive signs.98 EU institutions enjoy a margin of discretion in the choice of the means required to achieve their policies; and no business should ­legitimately expect that patterns of trade will remain unchanged.

5.3  Article 35 EU Charter and the Mainstreaming of Public Health Following the entry into force of the Lisbon Treaty in December 2009, the CJEU has continued to emphasize the limits to tobacco manufacturers’ commercial rights, 95  Opinion of Advocate General Fennelly in Case C-­380/03 Germany v Parliament and Council [2006] ECR I-­11573, at paragraph 154. 96  It hardly engaged in the balancing exercise: A Garde, ‘Freedom of Commercial Expression and the Protection of Public Health in Europe’ (2010) Cambridge Yearbook Eur Legal Studies 225. 97  See, in particular, Case 44/79 Liselotte Hauer [1979] ECR 3727; Case 52/81 Werner Faust [1982] ECR 3745; Case 265/87 Hermann Schräder [1989] ECR 2237; Case 5/88 Wachauf [1989] ECR 2609; Case C-­280/93 Germany v Council [1994] ECR I-­4973; Case C-­293/97 Standley and Others [1999] ECR I-­2603. 98  Case C-­491/01 British American Tobacco [2002] ECR I-­11453, at paragraphs 149, 150, and 152.

164   Amandine Garde although it now relies explicitly on the EU Charter and, in particular, Article 11 (freedom of expression), Article 16 (freedom to conduct a business), and Article 17 (right to property), rather than on the general principles of EU law. In particular, in its Sky Österreich decision, the Grand Chamber ruled that the EU legislature was entitled to give priority, in the necessary balancing of the rights and interests at issue, to overriding requirements of public interests over the private economic interests such as those protected by Articles 16 and 17 of the EU Charter, provided that a fair balance had been struck between several rights and fundamental freedoms protected by the EU legal order with a view to reconciling them.99 In relation to health more specifically, the CJEU was asked in Deutsches Weintor whether, by prohibiting the description of a wine as ‘easily digestible’, the Food Claims Regulation violated the freedoms of a German winegrowers’ cooperative to choose an occupation and conduct a business under Articles 15 and 16 of the EU Charter. In its judgment, the Court relied on Article 35 of the EU Charter, which requires that ‘a high level of human health protection be ensured in the definition and implementation of all the European Union’s policies and activities’, to dismiss the claim. After referring to the EU’s mainstreaming obligation laid down in Article 9 TFEU, the Court pointed out that ‘in view of the risks of addiction and abuse as well as the complex harmful effects known to be linked to the consumption of alcohol, in particular the development of serious diseases, alcoholic beverages represent a special category of foods that is subject to particularly strict regulation’. Thus, even if the claim is ‘substantively inherently correct in that it indicates reduced acidity levels’, it nonetheless remains ‘incomplete’ in that it is ‘silent as to the fact that, regardless of a sound digestion, the dangers inherent in the consumption of alcoholic beverages are not in any way removed, or even limited’. Consequently, the EU legislature was ‘fully entitled’ to take the view that such claims on alcoholic beverages are misleading and that ‘the prohibition of such claims is warranted in the light of the requirement to ensure a high level of health protection for consumers’. No mention was made in this case of a possible violation of the rights of winegrowers to free commercial expression; however, it is most unlikely that the outcome of the case would have changed if such mention had been made.100 Similarly, in its Philip Morris decision, the Court rejected the claimants’ argument that the revised Tobacco Products Directive infringed their right to freedom of expression under Article 11 of the EU Charter on the ground that ‘human health protection . . . ­outweighs the interests put forward by the claimants in the main proceedings’. Indeed, ‘as is apparent from the second sentence of Article 35 of the Charter and Articles 9, 114(3) and 168(1) of the TFEU, a high level of human health protection must be ensured in the definition and implementation of all the European Union’s policies and activities’.101

99  Case C-­283/11 Sky Österreich [2013] ECR I-­28, at paragraph 60. 100  Case C-­544/10 Deutsches Weintor, at paragraphs 48, 49, 51, and 52. See also Case C-­157/14 Neptune Distribution [2015] EU:C:2015:823. 101  Case C-­547/14 Philip Morris ECLI:EU:C:2016:325, at paragraphs 156, 157.

The EU’s Approach to Noncommunicable Disease Prevention   165 It is notable that the CJEU has relied extensively in this suite of cases on the duty of the EU to ensure a high level of public health protection in the development and implementation of all its policies. Arguably, in balancing the economic interests of tobacco or alcohol business actors against the protection of public health,102 the EU has indirectly recognized the need to protect the right to the enjoyment of the highest attainable stand­ ard of health. Even if the threshold of what would constitute ‘a high level of public health protection’ remains undefined, these provisions nonetheless require that the EU should place health concerns at the centre of the policy process and give them significant consideration when balancing them against other interests, not least the economic interests of the tobacco, alcohol, and food industries.103 In light of its ‘considerably greater importance in the value system under EU law’, human health protection can indeed justify even substantial restrictions to these interests.104 The EU is committed to the protection of human rights.105 Nevertheless, its reference to its public health mainstreaming obligation can only constitute, at best, an implicit recognition that human rights should guide EU NCD prevention law and policies. To date, there has been little reflection at the EU level on the added value of an explicit reliance on human rights to regulate the tobacco, alcohol, and food industries more effectively and therefore promote better health. If it is true that the wording of Article 35 of the EU Charter suggests that it may have a more limited scope than the provisions on the right to health in several international human rights instruments,106 the EU could nonetheless rely more explicitly on human rights as a justification for the implementation of NCD prevention regulatory measures. The very fact that the EU Charter ­contains a provision dedicated to health recognizes that health is indeed an important EU value. The EU should therefore recognize the onus that this provision, as complemented by the other health mainstreaming Treaty obligations and other relevant provisions of the EU Charter,107 places on its institutions to ensure that all EU policies do indeed protect the  right to health and negatively affected by tobacco, alcohol, and unhealthy food consumption.

102  See, for example, Recital 3 of Directive 2003/33 on tobacco advertising, and Recital 8 of Directive 2014/40 on tobacco products: ‘Tobacco products are not ordinary commodities and in view of the particularly harmful effects of tobacco on human health, health protection should be given high importance, in particular, to reduce smoking prevalence among young people.’ 103  A Garde, B Toebes, ‘Is There a European Human Rights Approach to Tobacco Control?’ in ME Gipsen, B Toebes eds., Human Rights and Tobacco Control (Edward Elgar 2020) ch 6. 104  Case C-­547/14 Philip Morris ECLI:EU:C:2016:325, at paragraph 179 of Advocate General Kokott’s Opinion. 105  Article 6 TEU. 106  Article 35 provides: ‘Everyone has the right of access to preventive health care and the right to benefit from medical treatment under the conditions established by national laws and practices. A high level of human health protection shall be ensured in the definition and implementation of all Union policies and activities’. This wording differs from the language used in several international human rights instruments, not least Article 12 ICESCR and Article 24 CRC, which refer to the ‘highest attainable standard of health’. 107  For example, Article 2: ‘Everyone has the right to life’.

166   Amandine Garde Mainstreaming is particularly important if the issue at hand is as complex as NCD prevention and requires a multisectoral response to the problems tobacco, alcohol, and unhealthy food consumption raises. Such consumption should be treated consistently across multiple policy fields when input from multiple policy fields—and therefore Directorates-­General of the European Commission—is required. In particular, a more explicit and systematic emphasis on the EU’s mandate to protect human rights (within the scope of its attributed powers) should lead the European Commission to reframe the discussions on the taxation of harmful commodities and propose to use EU legislative powers to increase the level of health protection across the EU. Similarly, the references to the rights of the child have multiplied in recent Council Conclusions calling on the EU and its Member States to do more to prevent child obesity, improve school food environments, and, in particular, protect children from exposure to alcohol and unhealthy food marketing.108 Nevertheless, the impacts these calls have had on the regulation of harmful commercial practices remains very limited indeed, and the gap is growing between the EU’s rhetoric and its actual implementation of rights-­based approaches.

6 Conclusion It is welcome that the Commission intends to design its Europe’s Beating Cancer Plan to cover the entire cycle of the disease starting from prevention and early diagnosis to treatment and quality of life of patients and survivors. Even though up to 40% of cancers may be preventable, only 3% of health budgets on average are spent on prevention across the EU. With the current impact of COVID-­19 on health in general, and on cancer prevention and care specifically, the Commission has recognized the urgent need to do more and has proposed €9.4 billion in EU funding for its EU4Health programme, hoping to significantly increase its support for cancer prevention and care. Over the years, the EU has adopted far-­reaching tobacco control legislation, and, in sharp contrast to the US Supreme Court, the CJEU has systematically rejected fundamental rights claims from the tobacco industry, explicitly giving preference to human health protection. Unfortunately, however, the EU has failed to use the powers it derives from the EU Treaties to effectively regulate the commercial practices of the food and alcohol industries and thereby ensure a high level of public health protection at the regional level, despite the fact that alcohol and food are traded extensively within the EU. Instead, it has preferred to rely on ineffective, voluntary commitments of industry actors and the exchange of best practice between Member States to address the disease burden stemming from alcohol and unhealthy food consumption.

108  See, in particular, Council Conclusions to contribute towards halting the rise in Childhood Overweight and Obesity of 15–16 June 2017 (Maltese Presidency) and Council Conclusions on healthy child nutrition of 22 June 2018 (Bulgarian Presidency).

The EU’s Approach to Noncommunicable Disease Prevention   167 On 20 May 2020, the Commission published its EU Farm to Fork Strategy,109 indicating a willingness to develop policies promoting sustainable food and healthier nutrition, where necessary through the adoption of EU-­wide legislation. Of interest is the Commission’s intention to improve the availability and price of sustainable food and to promote healthy and sustainable diets in institutional catering through the establishment of minimum mandatory criteria for sustainable food procurement. However, this exception aside, the Strategy only refers to the imposition of a front-­of-­pack nutrition labelling scheme, and still makes no mention at all of alcohol. The focus remains very much on ‘empowering consumers’ with more nutrition information.110 It is therefore unlikely that we will see in the near future the adoption of evidence-­based EU-­wide rules that would replace the limited, largely ineffective ‘pledges’ that the food and alcohol industries have made. The Commission’s promised new EU Children’s Rights Strategy provides at least another opportunity to remind European institutions (yet again) that they have an obligation to mainstream public health and children’s rights concerns in the development and implementation of all EU policies, including those that should involve the regulation of harmful commercial practices.111 The gap between what public health civil society organizations are calling for112 and what the EU seems to be prepared to do seems as wide as ever.

Bibliography A Alemanno, ‘EU Public Health Law and Policy—Tobacco’ in T Hervey et al. eds., Research Handbook on EU Health Law and Policy (Edward Elgar 2017), ch 14. A Alemanno, ‘Informing the Non-Communicable Diseases Agenda with Behavioural Insights’ in A Alemanno, A Garde eds., Regulating Lifestyles—Europe, Alcohol, Tobacco And Unhealthy Diets (Cambridge University Press 2015). A Alemanno, ‘Out of Sight Out of Mind: Towards a New European Tobacco Products Directive’ (2012) 18 Columbia J Eur L 197. A Alemanno, E Bonadio eds., The New Intellectual Property of Health: Beyond Plain Packaging (Elgar 2016). A Alemanno, A Garde, Legal Opinion on Standardised Tobacco Packaging (ASH, 2015): http:// ash.org.uk/information-and-resources/packaging-labelling-information-and-resources/ standardised-plain-packaging/legal-opinion-on-standardised-tobacco-packaging/ A Alemanno, A Garde, Regulating Lifestyles in Europe How to Prevent and Control NonCommunicable Diseases Associated with Tobacco, Alcohol and Unhealthy Diets? (Swedish Institute of European Policy Studies 2013). 109  COM(2020) 381 final. 110  Id. at page 13. On front-­of-­pack nutrition labelling, see COM(2020) 207 final. 111  https://ec.europa.eu/info/law/better-­regulation/have-­your-­say/initiatives/12454-­Delivering­for-­children-­an-­EU-­strategy-­on-­the-­rights-­of-­the-­child 112  See, in particular, the paper published jointly by the European Chronic Diseases Alliance, the European Public Health Alliance, and the NCD Alliance, Towards an EU Strategic Framework for the Prevention of NCDs (Brussels, May 2019): https://alliancechronicdiseases.org/wp-­content/uploads/ Final-­NCD-­Paper-­full-­version.pdf

168   Amandine Garde A Alemanno, A Garde eds., The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets (Cambridge University Press 2015). A Alemanno, A Garde, ‘The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets’ (2013) 50(6) Common Market L Rev 1745. L Azoulai, ‘The Question of Competence’, in L. Azoulai ed., The Question of Competence in the European Union (Oxford University Press 2014) 11. O Bartlett, A Garde, ‘On the Rocks: A Few Sobering Thoughts on the Growing EU Alcohol Problem’ in T. Hervey et al. eds., Research Handbook on EU Health Law and Policy (Edward Elgar 2017), 283. O Bartlett, A Garde, ‘The EU Platform and the EU Forum: New Modes of Governance or a Smokescreen for the Promotion of Conflicts of Interest?’ in A Alemanno, A Garde eds., The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets (Cambridge University Press, January 2015), 283. O Bartlett, A Garde, ‘Time to Seize the (Red) Bull by the Horns: The EU’s Failure to Protect Children from Alcohol and Unhealthy Food Marketing’ (2013) 38(4) Eur L R 498. JM Bauer, LA Reisch, ‘Behavioural Insights and (Un)healthy Dietary Choices: A Review of Current Evidence’ (2019) 42 J Consumer Pol 3. R Calo, ‘Code, Nudge or Notice?’, University of Washington Research Paper No. 2013–04. V Delhomme, ‘Between Market Integration and Public Health: The Paradoxical EU Competence to Regulate Tobacco Consumption’, Research Paper in Law 01/2018 (College of Europe 2018). B De Witte, ‘Non-Market Values in Internal Market Legislation’ in N.  Nic Shuibhne ed., Regulating the Internal Market (Edward Elgar 2006) 61. B De Witte, ‘A Competence to Protect: The Pursuit of Non-Market Aims Through Internal Market Legislation’ in P. Syrpis ed., The Judiciary, The Legislature and the EU Internal Market (Cambridge University Press 2012) 25. EU Network of Independent Experts on Fundamental Rights, Commentary on the Charter of Fundamental Rights of the European Union, Article 35—Protection de la Santé (June 2006), at page 304. M Friant-Perrot, A Garde, ‘From BSE to Obesity—EFSA’s Growing Role in the EU’s Nutrition Policy’ in A Alemanno, S Gabbi eds., Foundations of EU Food Law and Policy: Ten Years of the European Food Safety (Routledge 2014), ch 8. S Galbraith-Emami, T Lobstein, ‘The Impact of Initiatives to Limit the Advertising of Food and Beverage Products to Children: A Systematic Review’ (2013) Obesity Reviews. A Garde, ‘Advertising Regulation and the Protection of Children-Consumers in the European Union: In the Best Interest of . . . Commercial Operators?’ (2011) 19 Int’l J Children’s Rights 149. A Garde, B Toebes, ‘Is There a European Human Rights Approach to Tobacco Control?’ in ME Gipsen, B Toebes eds., Human Rights and Tobacco Control (Edward Elgar 2020) ch 6. A Garde, EU Law and Obesity Prevention (Kluwer Law International 2010). A Garde et al., ‘Implementing the WHO Recommendations Whilst Avoiding Real, Perceived or Potential Conflicts of Interest’ (2017) 8(2) Eur J Risk Regulation 237. A Gilmore, ‘Public Health, Corporations and the New Responsibility Deal: Promoting Partnerships with Vectors of Disease?’ (2011) 33(1) J Public Health 2. I Goldner Lang, ‘Public Health in European Union Food Law’ in T. Hervey et al. eds., Research Handbook on EU Health Law and Policy (Edward Elgar 2017) ch 16. N Gokani, ‘Regulation for Health Inequalities and Non-Communicable Diseases: In Want of (Effective) Behavioural Insights’ (2018) 24 Eur L J 490.

The EU’s Approach to Noncommunicable Disease Prevention   169 L Gruszczynski, M Melillo, ‘The FCTC and Its Role in WTO Law: Some Remarks on the WTO Plain Packaging Report’ (2018) 9 Eur J Risk Regulation 564. T Hervey, J McHale, European Union Health Law: Themes and Implications (Cambridge University Press 2015), 403. G Howells, The Tobacco Challenge (Ashgate 2011). G Howells, ‘The Potential and Limits of Consumer Empowerment by Information’ (2005) 32(3) J L Society 349, 356–358. J Jensen, K Ronit, ‘The EU Pledge for Responsible Marketing of Food and Beverages to Children: Implementation in Food Companies’ (2015) 69 Eur J Clin Nutrition 896. M Koivulsalo, ‘The State of Health in All Policies in the European Union: Potential and Pitfalls’ (2010) 64 J Epidemiol Comm Health 500. D Ludwig, M Nestle, ‘Can the Food Industry Play a Constructive Role in the Obesity Epidemic?’ (2008) 15 JAMA 300. C MacMaoláin, Food Law: National, European and International Perspectives (Hart 2015). M Melillo, ‘Evidentiary Issues in Philip Morris v Uruguay: The Role of the Framework Convention for Tobacco Control, and Lessons for NCD Prevention’ (2020) 21 J World Investment and Trade 724. R Moodie et al., ‘Profits and Pandemics: Prevention of Harmful Effects of Tobacco, Alcohol, and Ultra-Processed Food and Drink Industries’ (2013) 381 Lancet 670. R Moodie, ‘What Public Health Practitioners Need to Know About Unhealthy Industry Tactics’ (2017) 107 Am J Public Health 1047. M Nestle, Soda Politics: Taking on Big Soda (and Winning) (Oxford University Press, 2015). N Nic Shuibhne, M. Maci, ‘Proving Public Interest: The Growing Impact of Evidence in Free Movement Case Law’ (2013) 50(4) Comm Market L R 965. S Peeters et al., ‘The Revision of the 2014 European Tobacco Products Directive: An Analysis of the Tobacco Industry’s Attempts to “Break the Health Silo”’ (2016) 25 Tobacco Control 108. K Purnhagen et al., ‘More Competences Than You Knew? The Web of Health Competence for European Union Action in Response to the COVID-19 Outbreak’ (2020) 11 Eur J Risk Regulation 297. P Puska, T Stahl, ‘Health in All Policies—The Finnish Initiative: Background, Principles, and Current Issues’ (2010) 31 Ann R Public Health 315. L Sharma et al., ‘The Food Industry and Self-Regulation: Standards to Promote Success and to Avoid Public Health Failures’ (2010) 100 Am J Public Health 240. SL Steele et al., ‘The Role of Public Law-Based Litigation in Tobacco Companies’ Strategies in High-Income, FCTC Ratifying Countries, 2004–14’ (2016) 38 J Public Health 516. D Studlar, ‘Tobacco Control: The End of Europe’s Love Affair with Smoking’ in S Greer, P Kurzer eds., European Union Public Health Policy: Regional and Global Trends (Routledge 2013). C Sunstein, ‘The Ethics of Nudging’ in A Alemanno, AL Sibony eds., Nudge and the Law: A European Perspective (Hart Publishing 2015). A Taylor, M Jacobson, ‘Carbonating the World: The Marketing and Health Impact of Sugar Drinks in Low- and-Middle Income Countries’ (Center for Science in the Public Interest 2016). B Wansink, ‘Turning Mindless Eating into Health Eating’ in E Shafir ed., The Behavioral Foundations of Public Policy (Princeton 2012) 310–328. D Wyatt, ‘Freedom of Expression in the EU Legal Order and in EU Relations with Third Countries’ in J Beatson, Y Cripps eds., Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams (Oxford University Press 2000).

chapter 8

Socioeconomic Hea lth I n equa lities i n Eu rope The Role of Law and Human Rights Brigit Toebes

1  Setting the Scene It is widely evidenced that there are dramatic health inequalities (i.e., differences in health status) both within and between countries. An important contributor to these health inequalities are the conditions in which we are born, grow, live, work and age: our social determinants of health.1 Important social determinants include one’s education, family income, housing, and employment status. Where such conditions and not ­biological variations or free choice are at the root of inequalities, we call them “socioeconomic health inequalities.” Socioeconomic health inequalities are generally considered to be unnecessary and avoidable. As such, four interrelated concepts play a key role in this chapter: health inequalities, the social determinants, socioeconomic health inequalities, and health inequities. Throughout the chapter, these terms are used to some extent interchangeably, also because they are not always clearly distinguished in the literature. The term “health ­disparities,” which is most similar to “health inequalities,” is more common on the American continent. How does legal scholarship and practice come into play in this context? Given that socioeconomic health inequalities are largely unnecessary and avoidable, it is widely claimed that they lead to health inequities (i.e., avoidable inequalities in health). 1  WHO, Social determinants of health. Available at http://www.who.int/social_determinants/sdh_ definition/en/. Accessed January 2019. See also WHO, 10 facts on health inequities and their causes, https://www.who.int/features/factfiles/health_inequities/en/. Accessed 17 January 2019.

172   Brigit Toebes Addressing health inequities is considered to be an “ethical imperative” and a “matter of social justice.”2 This, first, leads us to the role of human rights. Human rights standards provide a moral and legal framework for assessing matters of social justice, including socioeconomic health inequalities. Given that the European institutions and domestic European government are strongly grounded in the European human rights tradition, this chapter pays attention to how human rights have thus far been framed in relation to socioeconomic health inequalities in Europe. At an operational level, domestic law and policy play a crucial role in connection with social and public health concerns. As also discussed by Wendy  E.  Parmet (“Social Determinants in the United States,” in this volume), law is itself a social determinant by structuring and perpetuating our social conditions.3 The relationship between law and health is very complex, also because laws and policies influence our health both in direct and indirect ways.4 For example, a rule that increases the price of tobacco may boost the health of those who manage to quit directly.5 Laws that are less related to health, also called “incidental laws,” may influence our health indirectly. For example, laws that govern terms and tenure of employment may influence the resources available to people to access tobacco cessation or to pay for cigarettes if they wish to continue smoking.6 Assuming that health inequalities born out of social determinants are unjust, we need to identify how human rights standards and specific laws can be best employed to improve social conditions. This is a very complex and ambitious question that can only lead to partial answers. In an attempt to create some clarity over how this matter plays out in Europe, this chapter looks at how the European institutions and domestic European governments have thus far addressed socioeconomic health inequalities. First, a brief overview is provided of the current data on health inequalities in Europe, with a main emphasis on the European Union (EU) region where most of the existing health research has been carried out. Subsequently, attention is paid to the main European organizations (EU and Council of Europe) and their role in addressing health inequalities. In this context, specific attention will be paid to the role of human rights law as a tool that may give support and priority to improving health and reducing inequities. Next, by way of explaining how social determinants are addressed at the domestic level, attention is paid to socioeconomic health inequalities in the United Kingdom and the Netherlands. 2 Id. See also WHO, Glossary of terms, https://www.who.int/hia/about/glos/en/index1.html. Accessed January 2019. 3  See also S Burris, I Kawachi, A Sarat, “Integrating Law and Social Epidemiology” (2002) J L Med Ethics, 30, 510–521; S Burris, “From Health Care Law to the Social Determinants of Health: A Public Health Law Research Perspective” (2011) U Penn L Rev 159(6), 1649–1667. See also OBK Dingkake, “The Rule of Law as a Social Determinant of Health (letter to editor)” (2017) Health Hum Rights Dec 19(2), 295–298. 4  Burris (n 3). See also the conclusions to his chapter. 5  There are many studies giving evidence of the effectiveness of a tobacco tax. For a recent study, see Prabhat Jha, “The Health, Poverty, and Financial Consequences of a Cigarette Price Increase Among 500 Million Male Smokers in 13 Middle Income Countries: Compartmental Model Study” (2018) BMJ 361, k1162. https://www.bmj.com/content/361/bmj.k1162. Accessed January 2019. 6  In conversation with Scott Burris. For the identification of incidental laws see Burris (n 3), 1663.

Socioeconomic Health Inequalities in Europe   173 By way of a disclaimer, it should be observed that the European legal landscape is very complex, both from a geographic and from a governance perspective. It encompasses two international organizations with a geographic scope and mandate7 in an interaction with a wide range of nation states with distinct domestic jurisdictions.

2  Health Inequalities in Europe While considerable progress has been made when it comes to improving overall population health in the European region, health inequalities have generally widened over the past decennia.8 These inequalities were even witnessed by more affluent countries and have been exacerbated by the economic crisis from 2008 onward.9 Death rates and poorer self-­assessments of health were higher in groups with lower socioeconomic status (SES).10 Less affluent and less educated people in the European Union have worse average levels of health than those with a higher income and ­education.11 These inequalities were larger in some countries than in others: for example, they were very large in most countries in the eastern and Baltic regions and small in some southern European countries.12 7  The European Union (EU) is an economic and political union of 28 European countries. The Council of Europe is an intergovernmental organization whose primary aims are to uphold human rights, democracy, and the rule of law in Europe. With 47 Member States, it has a much wider reach that the EU but less regulatory powers. 8  JP Mackenbach et al., “Changes in Mortality Inequalities over Two Decades: Register Based Study of European Countries” (2016) BMJ, 353, i1732. WHO Europe, Review of Social Determinants and the Health Divide in the WHO European Region: Final Report (Chaired by Michael Marmot). UCL Institute of Health Equity, WHO 2013. Varying from the Netherlands and France to Israel, the Russian Federation, and Tajikistan. Available at http://www.euro.who.int/__data/assets/pdf_file/0004/251878/ Review-­of-­social-­determinants-­and-­the-­health-­divide-­in-­the-­WHO-­European-­Region-­FINAL-­ REPORT.pdf. Accessed January 2019. See also Michael Marmot, Jessica Allen, Ruth Bell, Ellen Bloomer, Peter Goldblatt, on behalf of the Consortium for the European Review of Social Determinants of Health and the Health Divide, “WHO European Review of Social Determinants of Health and the Health Divide” (2012, 8 September) Lancet, 1–19, for the specific remark about the widening gap. 9  WHO Europe (2013 n 8), xiii (although Mackenbach et al., 2018, take a different viewpoint). 10  Three authoritative studies were published by Mackenbach in 1997, 2008, and 2017. See JP Mackenbach, AE Kunst, AEJM Cavelaars, F Groenhof, and JJ Geurts, “Socioeconomic Inequalities in Morbidity and Mortality in Western Europe” (1997) Lancet, 349, 165–169; Johan P Mackenbach, Irina Stirbu, Albert-­Jan Roskam, Maartje M Schaap, Gwenn Menvielle, Mall Leinsalu, and Anton Kunst, “Socioeconomic Inequalities in Health in 22 European Countries” (2008 June) N Engl J Med 358, 23, study for the European Union Working Group on Socioeconomic Inequalities in Health; and Johan P Mackenbach, Matthias Bopp, Patrick Deboorse, Katalin Kovacs, Mall Leinsalu, Pekka Martikainen, Gwenn Menvielle, Enrique Regidor, and Rianne de Gelder, “Determinants of the Magnitude of Socioeconomic Inequalities in Mortality: A Study of 17 Countries” (2017) Health & Place 47, 44–53. 11  European Commission, Commission Staff Working Document, Report on Health Inequalities in the European Union, Brussels, September 2013, SDW(2013), 328, Section 1. Available at https://ec. europa.eu/health/sites/health/files/social_determinants/docs/report_healthinequalities_swd_2013_328_ en.pdf. Accessed January 2019. 12  Mackenbach et al. (2008 n 10), 2468.

174   Brigit Toebes Both behavioural (smoking, alcohol) and structural factors (“poverty, and national economic, policy-­related and structural conditions”) contributed to between-­country variations in socioeconomic inequalities in mortality.13 Several studies confirm that differences in health behaviours related to tobacco consumption, obesity, and harmful use of alcohol contribute significantly to health inequalities.14 Other chapters in this volume address legal responses to behavioural factors. In this chapter, the focus is on the structural factors that cause health disparities, sometimes through their effect on the behavioural factors, as when low income increases the likelihood of tobacco consumption. The existing research shows the vulnerability of certain groups in European society, including minority ethnic groups such as the Roma and (undocumented) migrants.15 Childhood poverty and development also receive particular focus and attention based on the insight that these have a strong influence on health and other outcomes later in life.16 While inequalities persist, they are not as dramatic as in the United States. A recent study suggests that trends in health inequalities have been more favourable in Western Europe than in the United States.17 It found that no Western European country experienced the recent increases in mortality reported in the United States, which have been attributed to greater socioeconomic inequalities, less comprehensive social security arrangements, and widespread availability of dangerous prescription drugs.18 The fact that European healthcare systems are more comprehensive also plays a role in constraining inequalities.19

3  The Role of the European Union and the Council of Europe 3.1  The European Union’s Approach Toward Health Inequalities The European Commission, as the main executive force of the European Union, has explicitly recognized the existence of socioeconomic health inequalities in the EU

13  Mackenbach et al. (2017 n 10), 44–53. 14  European Commission, Commission Staff Working Document, Report on Health Inequalities in the European Union, Brussels, September 2013, SDW(2013), 328, Section 1. Available at https://ec. europa.eu/health/sites/health/files/social_determinants/docs/report_healthinequalities_swd_2013_328_ en.pdf. Accessed January 2019. See also Mackenbach et al. (2017), 2472. 15  For the Roma, see in particular WHO Europe (2013 n 8). 16  WHO Europe (2013 n 8), xxi. 17  Johan P Mackenbach, José Rubio Valverde, Barbara Artnik, MatthiasBopp, Henrik Brønnum-­ Hansen, Patrick Deboosere, Ramune Kalediene, Katalin Kovács, Mall Leinsalu, Pekka Martikainen, Gwenn Menvielle, Enrique Regidor, Jitka Rychtaříková, Maica Rodriguez-­Sanz, Paolo Vineis, Chris White, Bogdan Wojtyniak, Yannan Hu, and Wilma J Nusselder, “Trends in Health Inequalities in 27 European Countries” (2018 June) Proc Natl Acad Sci 115(25) 6440–6445; DOI:10.1073/pnas.1800028115. 18  A Case, and A Deaton, “Rising Morbidity and Mortality in Midlife Among White Non-­Hispanic Americans in the 21st century” (2015) Proc Natl Acad Sci USA 112,15078–15083. 19  Mackenbach et al. (2018 n 17).

Socioeconomic Health Inequalities in Europe   175 region. It has adopted various strategies and policies in the field which will be briefly discussed.20 To start with, what has encouraged the European Union to take action to address the social determinants? Health as an element of a productive economy is a key rationale underpinning the European Union’s policies.21 The European Union’s 2020 agenda for growth and jobs emphasizes the need for health policy as “keeping people healthy and active for longer has a positive impact on productivity and competitiveness.”22 Along similar lines, the European Union sees reducing poverty and social exclusion as a means to enhance sustainable and inclusive growth and as one of the ways to overcome the structural weaknesses in the EU economy.23 In a 2009 Communication, the European Commission expressed concern about the large gaps in health between and within EU Member States.24 In 2013, it followed with an extensive staff working document on health inequalities in the EU region,25 identifying EU policies addressing health inequalities and the social determinants of health. It stressed that achieving greater health equality involves cooperation among a broad range of policy areas, including “health, social affairs, research, education, energy, agriculture, development and regional policies.”26 While it referred to existing strategies in all these areas, the European Union’s cooperation among these dimensions remains somewhat underexplored. All actions remain at a policy level and do not (yet) have any specific implementation in law apart from specific regulation in the areas of tobacco and alcohol. It is worth noting that the European Union’s approach is to some extent grounded in the European human rights tradition encompassing also social rights, including the right to health. The above-­mentioned report took as a starting point for action in relation to the social determinants Article 35 of the Charter of Fundamental Rights of the European Union (CFREU), which recognizes the right of access to preventive healthcare and the right to benefit from medical treatment.27 It also endorsed the UN Convention on the 20  See generally, Website of the European Commission, Social determinants. Available at https:// ec.europa.eu/health/social_determinants/overview_en. Accessed January 2019. 21  In conversation with Tamara Hervey, August 2019. 22  EU 2020 Strategy. Available at https://ec.europa.eu/health/europe_2020_en. Accessed January 2019. 23  EU 2020 Strategy. Available at https://ec.europa.eu/info/business-­economy-­euro/economic-­and-­ fiscal-­policy-­coordination/eu-­economic-­governance-­monitoring-­prevention-­correction/european-­ semester/framework/europe-­2020-­strategy_en#featuresofthetargets. Accessed January 2019. In conversation with Tamara Hervey. 24  Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, “Solidarity in Health: Reducing Health Inequalities in the EU” (2009). Available at http://eur-­lex.europa.eu/legal-­content/EN/ TXT/?uri=CELEX:52009DC0567. Accessed January 2019. 25  European Commission, Commission Staff Working Document, Report on Health Inequalities in the European Union, Brussels, September 2013, SDW(2013), 328, Section 1. Available at https://ec. europa.eu/health/sites/health/files/social_determinants/docs/report_healthinequalities_swd_2013_328_ en.pdf. Accessed January 2019. 26  Id., 10–17. 27  It is thus important to note that this Charter, which constitutes the fundament of the EU when it comes to guaranteeing fundamental rights in the EU, also embraces a right to prevention.

176   Brigit Toebes Rights of Persons with Disabilities (CRPD), to which the European Union became a party in 2010 and which contains the right to the highest attainable standard of health of persons with disabilities.28 By starting from the EU Charter and the UN Convention, the EU approach has focused more on the health effects of social determinants than on the social determinants themselves. The report also paid attention to EU action on chronic diseases with the aim of pointing out the concrete action taken to reduce inequalities in chronic disease control and incidence. In this context, the Commission referred to several strategies aimed at improving treatment as well as at addressing behavioural risk factors including smoking, unhealthy diets, and excessive use of alcohol. To give one example, the European Pact for Mental Health and Well-­Being seeks to address health inequalities by destigmatizing mental health issues and by improving mental health services.29 The report stressed that action on mental health is particularly important in addressing inequalities because “people from less advantaged socioeconomic groups are more vulnerable to mental health problems, and mental health problems may themselves be a reason for weak performance at work or in school, or for social exclusion.”30 As to regulating risk factors, the report signalled that “differences in the prevalence of smoking between advantaged and disadvantaged groups are responsible for a significant proportion of the differences in death and disease rates between these groups.” It indicated that regulation to make tobacco products “less attractive, advertising bans, pricing policy and smoke-­free areas are [therefore] important in addressing social inequalities in tobacco use.”31 This is a field where the European Union has been most proactive from a regulatory perspective. As discussed by Amandine Garde elsewhere in this volume, the European Union has adopted a range of binding and nonbinding measures in relation to tobacco use and other behavioural risk factors. The ability of EU Member States to restrict the marketing of tobacco and unhealthy food and beverages has also been the subject of the case law of the European Court of Justice.32 The preceding discussion shows that the European Union has clearly embraced the social determinants. To my knowledge, no systematic research has thus far been carried out measuring whether these efforts have been effective. Furthermore, much of the action aimed at reducing socioeconomic health inequalities is focused somewhat narrowly on addressing the effects of social determinants, in particular behavioural risk factors including smoking, excessive use of alcohol, and unhealthy diets.

3.2  The Council of Europe With 47 Member States, the Council of Europe is a much larger organization than the European Union, and its mandate concerns the observance of democracy and 28  Id., 10. This refers to report in note 25. 29  European Pact for Mental Health and Well-­being, 2008. Available at http://ec.europa.eu/health/ ph_determinants/life_style/mental/docs/pact_en.pdf. Accessed January 2019. 30  Id., 12. 31  Id., 11 and 12. 32  See the chapter by Garde, in this volume.

Socioeconomic Health Inequalities in Europe   177 human rights. The Council of Europe has expressed interest in social determinants such as education, housing, and employment while recognizing the existence of health inequities.33 Yet, unlike the European Union, it has not developed targeted strategies or programmes in relation to the social determinants of health or reducing health inequalities. The importance of the Council of Europe in relation to social determinants should nonetheless not be underestimated. As the primary intergovernmental human rights organization in Europe, the organization provides an important framework for addressing the social determinants through the lens of human rights. Even though currently the organization does not address the social determinants explicitly, its sophisticated human rights mechanisms provide a moral and legal framework for analyzing socioeconomic health inequalities in Europe. Hence we need to find out how this human rights mechanism has thus far offered protection against health inequalities and how it may be employed more effectively in the future. Before discussing the Council of Europe framework, some general words need to be devoted to how human rights can play a role in addressing the social determinants and health inequalities. According to Marmot et al., “Human rights approaches support giving priority to improving health and reducing inequities.”34 Human rights law makes clear appeals to the governmental authorities at all levels to take action when injustices in health and other social fields occur.35 As suggested earlier, health inequalities arise from the conditions in which people are born, grow, live, and work. If we translate this understanding to human rights law, this suggests that human rights should guarantee not only an entitlement to healthcare but also an adequate physical environment and social conditions.36 Without doubt, the prime right in this field is the right to health as recognized in a wide range of human rights treaties, both at international and regional levels. To cite Hunt, former UN Special Rapporteur on the Right to Health, about the relation between the right to health and the social determinants: “there can be no doubt that the right to the highest attainable standard of health encompasses social determinants. . . . Addressing harmful social determinants is also a legal imperative. Reinforced by law, human rights are equity and ethics with teeth.”37 Both the right to health under the UN International 33  Council of Europe, “Health.” Available at https://www.coe.int/en/web/compass/health. Accessed January 2019. 34  M Marmot, J Allen, R Bell, E Bloomer, and P Goldblat (for the Consortium for the European Review of Social Determinants of Health and the Health Divide), “WHO European Review of Social Determinants of Health and the Health Divide” (2012) Lancet 380, 1011–1029. 35  Brigit Toebes and Karien Stronks, “Closing the Gap: A Human Rights Approach Towards Social Determinants of Health” (2016) Eur J Health L 23, 510–524, 512. See also WHO Europe, eds. Karien Stronks, Brigit Toebes, Aart Hendriks, Umar Ikram, and Sridhar Venkatapuram, Social Justice and Human Rights as a Framework for Addressing Social Determinants of Health: Final Report of the Task Group on Equity, Equality and Human Rights, WHO (2016). Available at http://www.euro.who. int/__data/assets/pdf_file/0006/334356/HR-­task-­report.pdf?ua=1, 29. Accessed January 2019. 36  Toebes and Stronks (2016 n 35), 513. 37  Paul Hunt, “Missed Opportunities: Human Rights and the Commission on Social Determinants of Health: UN Special Rapporteur on the Right to the Highest Attainable Standard of Health (2002–8)” (2009) Global Health Promotion 1757–9759, Supp(1), 36–41, 38.

178   Brigit Toebes Covenant on Economic, Social and Cultural Rights (ICESCR) and the right to protection of health under the (Revised) European Social Charter (ESC) reflect a broad approach to health, including not only access to healthcare but also the “underlying” or social determinants.38 This approach to the right of health is important for the recognition and protection of social determinants. In addition, it is important to connect the right to health to other human rights that relate to and reflect the determinants of health, including the right to life, the rights to information and participation, and the rights to an adequate standard of living, social security, food, housing, property, education, and employment.39 These rights imply legal obligations to ensure access to housing, a safe and clean living environment, education, and social protection. As such, they carry the potential to address the health consequences of poor housing conditions, lack of access to proper education, and occupational health hazards.40 So how have the Council of Europe’s human rights treaty bodies thus far dealt with the social determinants? While the collective complaint procedure under the European Committee of Social Rights, the treaty body of the European Social Charter, has thus far not addressed the social determinants explicitly, this mechanism holds potential for taking on this role in the future. First, from a procedural perspective, the collective complaint procedure allows for the representation of an affected group rather than an individual complainant, for example an indigenous population or a group of workers whose rights have been affected. As such, it may generate accountability for health inequalities in the way they usually affect not just individuals but larger groups of people. Furthermore, the jurisprudence on the substantive rights in the Council of Europe’s European Social Charter currently reflects specific dimensions of the social determinants, including industrial pollution and healthy working conditions; access to housing facilities; and access to food, clothing, and shelter, as well as access to sexual and reproductive health education.41 In the decision in ECCR v. Bulgaria, the European Committee of Social Rights came close to recognizing the social determinants of health and health inequalities by stating that Bulgaria had failed to “take reasonable steps to address the specific problems faced by Roma communities stemming from their often unhealthy living conditions and difficult access to health services.”42 The failure of the authorities to take appropriate measures to address the exclusion, marginalization, and environmental hazards to which Romani communities were exposed, as well as their problems in accessing healthcare services, led the Committee to conclude that Bulgaria 38  Toebes and Stronks (2016 n 35), 514. 39  The interdependence of rights was also affirmed in the Vienna Declaration and Programme of Action, UN Doc. A/Conf.157/23, 12 July 1993, para 5, as adopted during the World Conference on Human Rights, Vienna 1993. 40  Toebes and Stronks (2016 n 35), 514. 41  Relevant decisions of the ESC Committee include Marangopoulos v Greece (30/2005); ATD Fourth World v France (33/2006); INTERIGHTS v Croatia (45/2007); ECCR v Bulgaria (46/2007); DCI v the Netherlands (47/2008); FIDH v Greece (72/2011); and CEC v the Netherlands (90/2013). See also Toebes and Stronks (2016 n 35), 12. 42  ESC Committee, ECCR v Bulgaria (complaint no. 46/2007), para 49.

Socioeconomic Health Inequalities in Europe   179 had violated the right to protection of health (Article 11) and the right to medical assistance (Article 13) in conjunction with the principle of nondiscrimination (Article E) of the European Social Charter.43 The European Court of Human Rights (ECtHR) of the Council of Europe, as the court overseeing the implementation of the European Convention on Human Rights (ECHR), primarily deals with individual complaints and is thus less group-­oriented. In addition, it mainly engages with the civil and political rights in the ECHR, rather than the social rights which are so key to realizing the social determinants. Nonetheless, the importance, influence, and potential of this court in the field of health should not be overlooked. As asserted by Hendriks, the ECtHR has increasingly come out with “far-­reaching judgments in the fields of health and medicine, based on the assumption that healthcare represents an important value in society.”44 While the ECHR does not contain a right to health or healthcare, based on the ECtHR’s case law several (positive) obligations are imposed on contracting States relevant to health-­related concerns.45 These obligations are based on several rights in the ECHR, including Articles 2 (life), 3 (prohibition of torture), 8 (privacy and family life), and 10 (freedom of expression and access to information). Health concerns that the ECtHR has touched on include a denial of access to medical care and dysfunctional health systems, protection of vulnerable persons in healthcare settings and involuntary confinement, access to personal data, abortion, euthanasia, forced sterilization, and environmental health matters.46 Thus far, the ECtHR has not addressed the social determinants or health inequality directly; however, the focus on environmental health matters may reflect one step in that direction.

4  Domestic Approaches: Failures and Successes This section discusses how two European countries (i.e., parts of the United Kingdom and the Netherlands) have framed their strategies and policies in relation to the social determinants of health and to reducing health inequalities. While this country selection is by no means representative of Europe’s overall domestic approach towards the social 43  Id., para 51. 44  Hendriks, “Health and Human Rights and the Council of Europe,” in Toebes et al. (eds.) Health and Human Rights in Europe (Intersentia 2012) 27. 45  Id., 43 and 50.0 46  Examples include dysfunctional health systems: e.g., ECtHR Asiye Genc v Turkey, 27 January 2015, no 24109/07 and Vasileva v Bulgaria, 17 March 2016, no 23796/10; involuntary confinement: ECtHR 16 June 2005, Storck v Germany, no 61603/00 ECtHR 13 January 2009; access to personal data: ECtHR 4 May 2000, Powell v the UK, no 45305/99; abortion: ECtHR 5 September 2002, Boso v Italy, no 50490/99; euthanasia: ECtHR 29 April 2002, Pretty v the UK, no 2346/02; and environmental health protection: ECtHR 9 June 2005, Fadeyeva v Russia, no 55723/00. For a comprehensive overview (updated until 2012) see Hendriks (2012 n 44).

180   Brigit Toebes determinants of health, it gives some insight into how European countries approach the social determinants and health inequalities and what has been achieved so far by two Western European countries that can, to some extent, be qualified as good practice countries. The choice of these two counties is to some extent determined by the availability and accessibility of evidence on outcomes and policies on social determinants. But it also reveals different levels of engagement with the social determinants: while British approaches have been grounded explicitly in the social determinants, in the Netherlands the focus is mostly on addressing health inequalities. In light of these analyses, the concluding section will evaluate Europe’s efforts to address health inequalities and the social determinants of health. This will lead to some more general observations on the role of law and public health law more specifically in relation to the social determinants of health.

4.1  United Kingdom: England and Scotland England in particular and the United Kingdom in general have a long history of engagement with socioeconomic health inequalities. They were front-­runners when it comes to collecting systematic data in the field. The Whitehall studies, carried out in the 1970s and 1980s, investigated social determinants of health among British civil servants in the area of Whitehall in London, with a main focus on cardiovascular disease prevalence and mortality rate. These studies, which were led by Marmot, found a correlation between levels of employment and mortality rates: the lower the grade, the higher the mortality rate.47 In 1980, the Black Report was published by the (former) Department of Health and Social Security, which demonstrated widespread health inequalities and proposed a radical change in social and healthcare policies.48 English strategies and policies to reduce socioeconomic health inequalities have been adopted in stages.49 A 1999 action report, adopted in response to a 1998 inquiry with a range of recommendations, listed a range of new government policies including the introduction of a national minimum wage, higher benefits and pensions, and increased spending on education, housing, and healthcare.50 Subsequently, in 2003, a revised strategy was published by the Department of Health with twin aims: to deliver a national 47  Whitehall studies as conducted by Michael Marmot et al., 1978 and 1991. Published in J Epidemiol Comm Health (1978 Dec), 32(4), 244–249, and Lancet (1991 June 8), 337(8754), 1387–1393. 48  Department of Health and Social Services, Inequalities in Health: Report of a Research Working Group (“Black Repor”). (London: Department of Health and Social Services 1980). Johan P Mackenbach, “Has the English Strategy to Reduce Health Inequalities Failed?” (2010) Soc Sci & Med 71, 1249–1253, 1249. 49  Mackenbach (2010), 1249. (reference to note 48) 50  Department of Health, Reducing Health Inequalities: An Action Report (London: Department of Health 1999). Available at https://webarchive.nationalarchives.gov.uk/20110322214756/http://www. dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_4042496. pdf. Accessed January 2019. As also quoted from Mackenbach (2010 n 48), 1249.

Socioeconomic Health Inequalities in Europe   181 health inequalities target by 2010 (reducing inequalities in infant mortality and life expectancy at birth) and to support a long-­term sustainable reduction in health inequalities. It reiterated the need to tackle the structural determinants but with a stronger emphasis on local policies, including, for example, improved social housing and reduced poverty among vulnerable populations and improved access to public services in disadvantaged communities.51 In 2009, when it became clear that targets to reduce life expectancy and infant mortality by 10% were not being met, Marmot was appointed to chair an independent review to propose evidence-­based strategies.52 The 2010 “Marmot Report” entitled “Fair Society, Healthy Lives,” while reiterating earlier health outcomes about health inequalities, was critical about the previous strategies, indicating that these focused insufficiently on the background causes of ill health by relying increasingly on tackling proximate causes such as smoking.53 Furthermore, target groups were insufficiently reached and local communities were insufficiently involved: more deprived people living outside spearhead areas and local government and other local public sector partners should have been included to a greater extent.54 The Marmot Report contains six policy objectives: (1) Give every child the best start in life; (2) enable all children, young people, and adults to maximize their capabilities and have control over their lives; (3) create fair employment and good work for all; (4) ensure a healthy standard of living for all; (5) create and develop healthy and sustainable places and communities; and (6) strengthen the role and impact of ill health prevention.55 The UK government explicitly embraces this approach to the social determinants of health and identifies action on its governmental website.56 Similar efforts to adopt policies explicitly tackling socioeconomic health inequalities have been made in Scotland, where concern has been raised over “deep-­seated inequalities” despite an overall improvement of health. It is reported that children in the most deprived areas “have significantly worse health compared to children living in the least deprived areas,” being more prone to low birth weight, poorer dental health, higher obesity levels, and higher rates of teenage pregnancy. Especially indicators like healthy life expectancy, mental health, smoking, and alcohol and drug misuse remain significantly worse in the most deprived parts of Scotland.57 On its website on “health and poverty,” 51  UK Department of Health, Tackling Health Inequalities: A Programme for Action, 2003. Available at https://webarchive.nationalarchives.gov.uk/20031221042257/http://www.doh.gov.uk/healthinequalities/ programmeforaction/. Accessed September 2019. 52  Mackenbach (2010 n 48), 1250. 53  Fair Societies and Healthy Lives (The Marmot Review), Strategic Review of Health Inequalities in England post-­2010, 86. Available at http://www.instituteofhealthequity.org/resources-­reports/ fair-­society-­healthy-­lives-­the-­marmot-­review/fair-­society-­healthy-­lives-­full-­report-­pdf.pdf. Accessed April 2018. 54  Marmot Review (n 53), 90; and Mackenbach (2010 n 48), 1251. 55  Marmot Review (n 53), executive summary. 56  Government of the UK, Public Health England. Available at https://www.gov.uk/government/ publications/health-­profile-­for-­england/chapter-­6-­social-­determinants-­of-­health. Accessed January 2019. 57  Audit Scotland, Health Inequalities in Scotland. Available at http://www.audit-­scotland.gov.uk/ docs/health/2012/nr_121213_health_inequalities.pdf. Accessed January 2019.

182   Brigit Toebes the Scottish Government explicitly endorses the Marmot approach, makes explicit links between the social determinants and the human rights framework, and identifies tools for action.58 The Marmot Report has had a significant impact on the design of policies and strategies in the United Kingdom and in other parts of Europe. Statistics nonetheless suggest that the British strategies to reduce socioeconomic health inequalities may not have been that effective. A 2017 governmental report focusing on England specifically establishes that stark inequalities remain. The report indicates that the most recent data on England show “wide inequalities across all indicators related to child health, mental health, smoking, alcohol misuse, and TB, and no trend indicating a clear narrowing of these inequalities.”59 It seems extremely difficult—as suggested by Mackenbach “beyond our means”—to reduce health inequalities, requiring efforts that may be in “excess of what western state machineries can deliver,” such as changing the redistribution of income and wealth, measures which may lack political support.60 Yet there is also some reason for hope that certain targeted actions may have been effective. For example, a systematic review by Bambra et al. carried out between 2007 and 2010 in the United Kingdom suggests that certain categories of intervention have impacted positively on inequalities or on the health of specific disadvantaged groups, particularly interventions in housing and the work environment.61 For example, the review refers to evidence that rent assistance led to improvements in health and health status and that organizational-­level work re­or­ gan­i­za­tion had a positive impact on self-­reported health.62

4.2  The Netherlands A review by Kulhánová et al. reveals that, in the Netherlands, people in lower educated groups have a “higher prevalence of a less than good perceived general health, more 58  Scottish Government, Poverty and Health. Available at http://www.gov.scot/Topics/Statistics/ Browse/Social-­Welfare/IncomePoverty/PovertyHealth. Accessed January 2019. 59  Public Health England, Public Health Outcomes Framework: Health Equity Report (London: PHE Publications 2017). 60  Mackenbach (2010 n 48), 1251. 61  C Bambra, M Gibson, A Snowden, K Wright, M Whitehead, and M Petticrew, “Tackling the Wider Social Determinants of Health and Health Inequalities: Evidence from Systematic Reviews” (2010) J Epidemiol Comm Health, 64, 284–291. 62  Id., referring to LM Anderson, J St Charles, MT Fullilove et al., “Providing Affordable Family Housing and Reducing Residential Segregation by Income: A Systematic Review” (2003) Am J Prev Med 24, 47–67; D Acevedo-­Garcia, TL Osypuk, RE Werbel et al., “Does Housing Mobility Policy Improve Health?” (2004) Housing Policy Debate15, 49–98; and M Egan, C Bambra, S Thomas et al., “The Psychosocial and Health Effects of Workplace Reorganization 1. A Systematic Review of Organisational-­Level Interventions That Aim to Increase Employee Control” (2007) J Epidemiol Comm Health (2007), 61, 945–954.

Socioeconomic Health Inequalities in Europe   183 chronic diseases, higher disability rates and higher risks of all-­cause mortality compared to higher educated groups.”63 While these inequalities were similar to other Northwestern European countries, the authors found larger inequalities in the Netherlands than elsewhere for lung cancer as a cause of death, which led this and several other studies to conclude that the level of education was a strong indicator for lifestyle-­related diseases, in particular lung cancer as caused by smoking.64 Dutch action in relation to socioeconomic health inequalities started in the 1980s, in response to the publication of the above-­mentioned Black Report in England and a report on health inequalities between neighbourhoods in the city of Amsterdam.65 Gradually, the interest among policymakers to address socioeconomic health inequalities rose, further strengthened by the “Health for All by the Year 2000” targets of the World Health Organisation (WHO).66 Governmental policies addressing socioeconomic health inequalities have fluctuated depending on the willingness of governments to take action. Dutch policies and approaches evolved from a broad concern about socially and economically marginalized groups in the 1980s to specific concerns about socioeconomic differences in the 1990s.67 Throughout these years, the emphasis has been on addressing socioeconomic health inequalities rather than on the social determinants. As such, the approach may have been somewhat different from the United Kingdom, where the social determinants approach is clearly embraced, clearly resonating the language from the Marmot Review. From the early 1990s onwards, the Dutch Ministry of Health followed a systematic research-­based approach to tackling socioeconomic inequalities in health. While an initial 5-­year research programme mapped the nature and determinants of socioeconomic inequalities in health,68 a second 6-­year programme was aimed at gaining systematic experience with interventions and policies aimed to reduce health inequalities.69 63  Ivana Kulhánová, Rasmus Hoffmann, Terje A Eikemo, Gwen Menvielle, and Johan P Mackenbach, “Educational Inequalities in Mortality by Cause of Death: First National Data for the Netherlands” (2014) Int J Public Health 59, 687–696. 64  Id., 694. 65  UK Department of Health and Social Services (“Black Report”) (1980); A Lau-­Ijzerman, JDF Habbema, and PJ van der Maas, Vergelijkend buurtonderzoek Amsterdam [comparative neighbourhood investigation Amsterdam] (Amsterdam: GGD 1981); Johan P Mackenback and Karien Stronks, “The Developments of a Strategy for Tackling Health Inequalities in the Netherlands” (2004) Int J Equity Health 3, 11, 1. 66 WHO, Targets for Health for All (Copenhagen: WHO 1985); Mackenbach and Stronks (2004). 67  M Droomers, L den Broeder, L Burdorf, and JP Mackenbach, “The Netherlands,” in C Hogstedt, H Moberg, B Lundgren, M Backhans (eds), Health for All? A Critical Analysis of Public Health Policies in Eight European Countries (Östersund: Swedish National Institute of Public Health 2008). 68  See JP Mackenbach, “Socioeconomic Inequalities in Health in the Netherlands: Impact of a Five Year Research Programme” (1994) BMJ 309, 1487–1491. 69  ZON MW, Programme Committee on Socioeconomic Inequalities in Health—second phase. In Reducing Socioeconomic Inequalities in Health. Final report and policy recommendations from the Dutch Programme Committtee on Socioeconomic Inequalities in Health—Second Phase (Den Haag: ZON MW 2001).

184   Brigit Toebes In response to these efforts, the government adopted a range of policy goals to increase life expectancy of the lowest socioeconomic group by 3 years by 2020.70 Important goals were decreasing inequalities in education and employment and increasing the accessibility of healthcare facilities to persons with low socioeconomic status.71 However, despite pressure from governmental advisory bodies, the attention for this approach dropped at the beginning of this millennium, with governmental policies emphasizing individual responsibility. As a result, the Minister of Health shifted the responsibility to tackle health inequalities to the large cities as part of the Urban Policy Framework. In turn, the cities chose to focus specifically on excess weight in minors.72 With the new 2007 government, political concerns about health inequalities revived with a new governmental approach entitled “Towards an Able-­Bodies Society.”73 However, these plans lacked a clear, coordinated approach, and clear targets were not set.74 The new 2017 governmental agreement contains clear statements on prevention; however, a clear strategy on tackling socioeconomic health inequalities is lacking.75 The subsequent National Prevention Agreement, launched in November 2018, focuses mainly on reducing risk factors and sets a range of targets in this field in collaboration with 70 societal partners.76 A recent report from the Dutch Scientific Council for Government Policy suggests a different direction in the debate by emphasizing the importance of health potential instead of health inequalities. Along these lines, specific attention should go to the first 18 years of life, people with low socioeconomic status, and addressing four behavioural risk factors (i.e., tobacco use, excessive use of alcohol, unhealthy diets and lack of physical exercise).77 This reflects an attempt to redesign new approaches to social determi-

70  Government stand on the final report and policy recommendations of the second Programme Committee on Socioeconomic Health Differences To Reduce Socioeconomic Health Differences and the Public Health Forecast Study Health in the Cities (The Hague: National Institute of Public Health and the Environment 2001). Available at https://www.parlementairemonitor.nl/9353000/1/j9vvij5epmj1ey0/ vi3ak1bzo3vw. Accessed January 2019. 71 Id. 72  Droomers et al. (n 67), 17. 73  Government of the Netherlands, Towards an Able-­bodied Society. A Policy Plan for Dealing with Health Inequalities Related to Socioeconomic Background (The Hague 2008). 74  Droomers et al. (n 67), 17. 75  Regeerakkoord 2017–2021, Vertrouwen in de toekomst [Governmental agreement, Faith in the Future], 10 October 2017. 76  Dutch Government, Nationaal Preventieakkoord [National Prevention Agreement], November 2018. Available at https://www.rijksoverheid.nl/documenten/convenanten/2018/11/23/nationaal-­ preventieakkoord. Accessed January 2019. For a review in English, see RIVM (2018). Available at https://www.rivm.nl/en/news/ambitions-­National-­Prevention-­Agreement-­feasable-­for-­smoking-­more-­ measures-­necessary-­to-­%20reduce-­overweight-­and-­alcohol-­%20use. Accessed January 2019. 77 WRR, Van verschil naar potentieel [from difference to potential], August 2018. Available at https:// www.wrr.nl/publicaties/policy-­briefs/2018/08/27/van-­verschil-­naar-­potentieel.-­een-­realistisch-­ perspectief-­op-­de-­sociaaleconomische-­gezondheidsverschillen. Accessed January 2019.

Socioeconomic Health Inequalities in Europe   185 nants of health: instead of a focus on inequalities, focus on improving the health status of the vulnerable in society.

5 Discussion This chapter has addressed approaches towards socioeconomic health inequalities in Europe, looking both at regional and domestic practices. It is difficult to draw comprehensive conclusions given the variety in health outcomes as well as the varying approaches taken to address inequalities at both regional and domestic levels. Firstly, it is important to establish that, while not as dramatic as in the United States, socioeconomic health inequalities are a reality in Europe. Statistics suggest that health inequalities have widened, both between and within European countries. We may conclude that when it comes to building a healthier European society, reducing health inequalities should be a key priority. The European Union, as well as many domestic European states, has adopted dedicated strategies, policies, and laws to take action to reduce socioeconomic health inequalities. If and to what extent these measures have worked is difficult to measure. Evidence suggests that some targeted interventions have been effective, including the regulation of risk factors such as smoking. Yet given that the causes of inequalities are engrained in the structures of our society, the focus should not only be on regulating risk factors specifically, but also on structural factors, including poverty, lack of income, unemployment, and housing conditions. It remains, however, extremely challenging to adopt more comprehensive governmental policies as well as targeted laws in relation to these broader structural factors and to measure their overall effectiveness. For the purposes of this volume, a specific question arises: What is the potential of law in this context? At a foundational level, human rights law can play a role in addressing social justice and reducing socioeconomic health inequalities. By giving expression to the vulnerability of the individual, human rights law can serve as an overarching framework protecting the rights and interests of those affected by health inequalities and poor social conditions. Taking a human rights approach then also requires looking at the right to health in an interaction with other human rights, including rights to housing and social security, as well as labour rights. This chapter has made it clear that, in Europe, and much more than in the United States, social rights are a strong potential driver for addressing socioeconomic health inequalities. EU policies in this field should be explicitly grounded in social rights, as defined by the human rights framework of the Council of Europe and the EU’s Fundamental Charter. In settings where these social rights are not recognized directly, Wesson suggests that we could also recognize “social condition” as a kind of unlawful discrimination, in addition to other existing kinds such as sex and age. While equality may not lead to an obligation to provide services like healthcare, it would allow courts to decide cases in

186   Brigit Toebes which a group has been excluded from a certain social service or situations where the state curtails social programmes.78 From an accountability perspective, it is important for courts and other accountability mechanisms to have an understanding of health inequities and social determinants and of how human rights come into play in this context. So far, there is little experience with this matter. The European Committee of Social Rights, Council of Europe’s treaty body for the European Social Charter, addresses the social determinants to some extent under the umbrella of the social rights in the Charter. In this context the substantive social rights, including the right to health, housing, social security, and education, are used to address social conditions directly. At a more applied level, we ask what the role of law is in reducing socioeconomic health inequalities. Burris and others have pointed out that law is both a vehicle through which health inequalities are exacerbated and a tool to reduce them and to improve social determinants.79 The role of law in relation to the social determinants of health is incredibly broad-­ranging and extends to many fields, including the regulation of agriculture and food production, education, work environment, income, housing, and the overall socioeconomic and environmental conditions. In this context, it is important not only to focus on the effects of particular laws, but also to investigate the utility of law more generally as a mechanism for advancing health.80 It is obvious that (public) health law cannot play a role in all these areas. It is nonetheless crucial for public health lawyers to have an understanding of how these broader laws, sometimes called “incidental” public health laws, impact on health.81 Academic research on the role of law in relation to the social determinants of health in Europe is scant compared to what has been done in the United States. According to Burris, it is “important to know whether policies intended to improve our health are working.”82 To fully understand how law is implicated in the social determinants, more empirical legal research exploring the impact of law needs to be conducted in Europe. To give just one example of what can be done: in recent research carried out in the United States, increases in the state-­level minimum wage above the federal value were associated with a reduction in heart disease death rates among individuals aged 35–64 years.83 Health inequalities and social determinants are an important phenomenon in our society. They demand action, and human rights and domestic law have a crucial role to 78  Murray Wesson, “Social Condition and Social Rights” (2006) Saskatchewan L Rev 69, 201–216. 79  Burris (n 3), 1649–1667. See also OBK Dingkake, “The Rule of Law as a Social Determinant of Health” (letter to editor)” (2017 Dec) Health Hum Rights 19(2), 295–298; and Parmet, “Social Determinants in the United States,” in this volume. 80  A Wagenaar and S Burris (eds), Public Health Law Research: Theory and Methods (San Francisco: Joseph Wiley and Sons 2013). 81  Burris (n 3), 1663. 82  Id., 1662. 83  ME Van Dyke, KA Komro, MP Shah, MD Livingston, and MR Kramer (2018), “State-­Level Minimum Wage and Heart Disease Death Rates in the United States, 1980–2015: A Novel Application of Marginal Structural Modeling” (2018) Prev Med 112, 97–103. doi: https://doi.org/10.1016/j. ypmed.2018.04.009.

Socioeconomic Health Inequalities in Europe   187 play in framing what needs to be done. This chapter has demonstrated that when it comes to identifying the role of law and human rights in Europe, we are only scratching the surface.

Acknowledgments The author wishes to thank Scott Burris, Karien Stronks, David Orentlicher, and Wendy Parmet for their valuable advice and feedback. All errors and inconsistencies are the responsibility of this author.

chapter 9

Soci a l Deter mi na n ts i n the U n ited States Wendy E. Parmet

1 Introduction The United States spends far more per capita on healthcare than any other nation.1 Despite these high expenditures, the United States fares poorly on a wide range of population health metrics. For example, in 2015, the United States ranked 26th among member nations of the Organisation for Economic Cooperation and Development (OECD) in terms of life expectancy and 35th in terms of infant mortality.2 To make matters worse, even before the COVID-19 pandemic, life expectancy at birth had declined for several years, a trend not seen previously seen since the aftermath of World War I.3 Behind these broad indicia are numerous troubling signs of ill-­health. For example, the country is in the midst of a substance use disorder crisis: in 2017 alone, more than  64,000 Americans died from opioid overdoses.4 That same year, more than 1  Health at a Glance, Health Expenditures per Capita, oecd (Nov. 10, 2017), https://www.oecd-­ ilibrary.org/docserver/health_glance-­2017-­44-­en.pdf?expires=1524577874&id=id&accname=guest&ch eck­sum=00EDFCBA37637E7822EBE0428A727CB2. 2  Comparison with Other Nations, American’s Health Rankings: United Health Foundation, https://www.americashealthrankings.org/learn/reports/2016-­annual-­report/comparison-­with-­other-­ nations (last accessed May 18, 2018). 3  Associated Press, US Life Expectancy Declines Again in Sobering ‘Wake-­Up Call,’ Stat (Nov. 29, 2018), https://www.statnews.com/2018/11/29/u-­s-­life-­expectancy-­declines-­again-­in-­sobering-­ wake-­up-­call/; Anne Case & Angus Deaton, Mortality and Morbidity in the 21st Century, Brooking Papers on Economic Activity, Spring 2017, at https://www.brookings.edu/wp-­content/uploads/2017/08/ casetextsp17bpea.pdf. This Chapter was written during the Trump Administration, before the pandemic began. Although much has changed, the longer-term relationships between US law and social determinants have not. 4  Maya Salam, The Opioid Epidemic: A Crisis Years in the Making, N.Y. Times (Oct. 26, 2017), https://www.nytimes.com/2017/10/26/us/opioid-­crisis-­public-­health-­emergency.html.

190   Wendy E. Parmet 44,000 Americans died by suicide.5 In 2016, 38,000 died from firearms,6 most due to suicide.7 More than a third of the adult population is obese,8 and more than 100,000 American adults live with diabetes or pre-­diabetes.9 All of these conditions add to the costs of the healthcare system.10 These disturbing numbers mask equally troubling disparities (often referred to as “inequalities” in Europe) across populations. For example, although the gap has been falling, life expectancy among African Americans in the United States in 2014 was 75.6 years, as compared to 79.0 years for whites.11 African Americans also have higher rates than whites for eight of the ten leading causes of death prior to the pandemic, as well as higher rates of infant mortality, preterm and low birth weight births, and maternal mortality.12 Other minority groups also experience significant negative disparities. For example, mortality rates for Native Americans are almost 50% higher than for white Americans.13 Likewise, people with disabilities and the LGBTQ community face significant health disparities.14 5  Suicide Rate is Up 1.2 per cent According to Most Recent CDC Data (Year 2016), American Foundation for Suicide Prevention (January 2,2018), https://afsp.org/suicide-­rate-­1-­8-­per cent-­according-­recent-­cdc-­data-­year-­2016/ 6  Maya Rhodan, Gun-­Related Deaths in America Keep Going Up, Time (Nov. 6, 2017), http://time. com/5011599/gun-­deaths-­rate-­america-­cdc-­data/. 7  Margot Sanger-­Katz, Gun Deaths Are Mostly Suicides, N.Y. Times (Oct. 8, 2015), https://www. nytimes.com/2015/10/09/upshot/gun-­deaths-­are-­mostly-­suicides.html. 8  Centers for Disease Control and Prevention, Adult Obesity facts, https://www.cdc. gov/obesity/data/adult.html (last accessed May 18, 2018). 9  Centers for Disease Control and Prevention, New CDC Report: More Than 100 Million Americans Have Diabetes or Prediabetes (July 18, 2017). 10  Amit K. Reddy et al., Survivors of Self-­Inflicted Gunshot Wounds to the Head: Characterization of Ocular Injuries and Health Care Costs, 132 JAMA Ophthalmology, 730, 730 (2014); Linda Gunderson, The Financial Costs of Gun Violence, 131 Annals of Internal Med. 483, 483 (1999); Daniel D. Maeng, Patterns of Health Care Utilization and Cost Before and After Opioid Overdose: Findings from 10-­year Longitudinal Health Plan Claims Data, 8 Substance Abuse and Rehabilitation 57, 57 (2017); David B. Allison et al., The Direct Health Care Costs of Obesity in the United States, 89 Am. J. of Pub. Health 1194, 1194 (1999); Xiaohui Zhuo et al., Lifetime Direct Medical Costs of Treating Type 2 Diabetes and Diabetic Complications, 45 Am. J. Prev. Med. 253, 253 (2013). 11  Centers for Disease Control and Prevention, Morbidity and Mortality Weekly Report (May 5, 2017), https://www.cdc.gov/mmwr/volumes/66/wr/mm6617e1.htm. 12  Id., Centers for Disease Control and Prevention, Morbidity and Mortality Weekly Report (July 8, 2011),; Centers for Disease Control and Prevention National Vital Statistics Reports; Births: Final Data for 2016, 35, Table 13 (Jan. 31, 2018); Centers for Disease Control and Prevention, Morbidity and Mortality Weekly Report, Pregnancy Related Mortality Surveillance, United States, 1991-­1999 (Feb. 21, 2003). In contrast, Hispanics, generally have lower mortality rates than the non-­Hispanic white population. John M. Ruiz et al., Hispanic Mortality Paradox: A Systematic Review and Meta-­Analysis of the Longitudinal Literature, 103 Am. J of Pub. Health. E52, E52 (2012). 13  National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Population Health and Public Health Practice; Committee on Community-­Based Solutions to Promote Health Equity in the United States, Communities in Action: Pathways to Health Equity (Baciu A, Negussie Y, Geller A, et al. eds. 2017), https:// www.ncbi.nlm.nih.gov/books/NBK425844/?report=printable (ebook). 14  Jonathan Mithen et al., Inequalities in Social Capital and Health Between People With and Without Disabilities, 126 Soc. Sci. & Med. 26, 26 (2015).; Vickie L. Harvey, Health Care Disparities and the LGBT Population 77 (2014)

Social Determinants in the United States   191 Disparities also exist based on income and socioeconomic status (SES). For example, between 2001 and 2014, men in the bottom 1%of the income distribution at age 40 had a life expectancy of 72.7 years, while men in the top 1% of the income distribution could expect to live to be 87.3 years.15 Geography and even zip (postal) code, are also associated with differential health outcomes. As the Robert Wood Johnson explains, “Across America, babies born just a few miles apart have dramatic differences in life expectancy.”16 Moreover, although non-­Hispanic whites as a group continue to have lower mortality rates than blacks, both mortality and morbidity at midlife has been increasing among non-­Hispanic whites without a college degree.17 Much of this development can be explained by increases in what economists Anne Case and Angus Deaton have termed “diseases of despair,” which include drug overdoses, suicides, and alcohol-­related liver disease.18 What explains these troubling trends, as well as the continuing disparities between the United States and other high-­income countries and across groups within the United States? Some blame, no doubt, belongs to the US healthcare system and the many gaps and inefficiencies in that system that are discussed elsewhere in this volume. But healthcare itself plays a relatively minor role—some researchers suggest 20%—in determining health outcomes.19 Far more important are the so-­called social determinants of health20 (SDOH). According to the World Health Organization, these are “the conditions in which people are born, grow, live, work and age.”21 They are the background social and environmental factors that help to determine the exposures individuals face and the choices they make, including whether to exercise, smoke, drink alcohol excessively, or practice unsafe sex. As Braveman and colleagues explain, individuals do not choose to make healthy or unhealthy choices in a vacuum; healthy choices “require opportunities and support in the environments where people live, learn, work and play.”22 Numerous (and often interrelated) social determinants affect individual health as well as the differential outcomes experienced by varied populations. Among the culprits 15  Raj Chetty et al, The Association Between Income and Life Expectancy in the United States, 2001-­2014, 315 jama 1750, 1753 (2016). 16  Mapping Life Expectancy, Robert Wood Johnson Foundation (September 11, 2015) https:// www.rwjf.org/en/library/articles-­and-­news/2015/09/city-­maps.html/. 17  Id. at 402. 18  Anne Case & Angus Deaton, Mortality and Morbidity in the 21st Century, Brooking Papers on Economic Activity, 397, 398 (Spring 2017), https://www.brookings.edu/wp-­content/uploads/2017/08/ casetextsp17bpea.pdf. (They also attribute the slowing down of mortality declines to declines in mortality reductions due to heart disease). 19  Samantha Artiga & Elizabeth Hinton, Beyond Health Care: the Role of Social Determinants in Promoting Health and Health Equity, kff (May 10, 2018), https://www.kff.org/disparities-­policy/ issue-­brief/ beyond-­health-­care-­the-­role-­of-­social-­determinants-­in-­promoting-­health-­and-­health-­equity/. 20  The US Burden of Disease Collaborators, The State of US Health, 1990-­2015: Burden of Diseases, Injuries, and Risk Factors Among US States, 319 JAMA 1444 (2018). 21  World Health Organization, Social Determinants of Health, http://www.who.int/ social_determinants/sdh_definition/en/(last visited May 18, 2018). 22  Paula A. Braveman, Susan A. Egerter, & Robin E. Mochkenhaupt, Broadening the Focus, The Need to Address the Social Determinants of Health, 40 Am. J. Prev. Med. S4 (2011).

192   Wendy E. Parmet are poverty, economic inequality, racism, and substandard housing.23 Although all of the pathways by which these factors and others influence health are not fully known, researchers have looked to the accumulated stress that these determinants (and others) create across an individual’s lifetime.24 Adverse experiences in childhood may be especially detrimental to health across the lifespan.25 In their recent work, Case and Deaton point to the distal, root forces that contribute to the epidemic of diseases of despair, which has particularly hit non-­Hispanic whites who lack a bachelor’s degree. As Case and Deaton explain, although the increase in supplies of opioids that occurred as a result of changes in prescription marketing and practices earlier in the century played a role in rising mortality, it merely “add[ed] fuel to the flames.”26 More fundamental are an amalgam of social factors, including declining marriage and employment rates (due to globalization and automation), which has led to a loss of hope for people with less education.27 David M. Cutler summarizes Case and Deaton’s findings by noting, “People despair when their material and social circumstances are below what they had expected. This despair leads people to act in ways that significantly harm their health. This may have a direct impact on death through suicide, or an indirect impact through heavy drinking, smoking, drug abuse, or not taking preventive medications for conditions such as heart disease. At root is economic and social breakdown.”28 The economic, cultural, and political forces creating this breakdown are myriad, but without question the growing levels of inequality in recent decades, compounded by stagnant wage growth, play a role. According to the World Economic Forum, the United States ranks 23 out of 30 developed nations in the World Economic Forum’s “inclusive development index,” which measures distribution of income and wealth and levels of poverty.29 The United States also spends far less than most other developed countries on social programs other than healthcare that can buffer inequality’s bite.30 A full exploration of all of these issues is well beyond the scope of this chapter. Instead, this chapter focuses on one important but often overlooked social determinant—law. The chapter begins by explaining how law influences social determinants and why law 23  Thomas E. Fuller-­Rowell, Poverty and Health: the Mediating Role of Perceived Discrimination, 23 Psych. Sci. 734 (2012); Danny Dorling, The Mother of Underlying Causes: Economic Ranking and Health Inequality; 128 Soc. Sci. & Med. 327 (2015); David R. Williams & Selina A. Mohammed, Racism and Health I: Pathways and Scientific Evidence; 57 Am. Behavioral Scientist 1152 (2013); Virginia A. Rauch et al., Housing and Health: Intersection of Poverty and Environmental Exposures; 1136 Ann. N.Y. Aced. Sci. 276 (2008); Wendy E. Parmet, Lauren A. Smith & Meredith A. Benedict, Social Determinants, Health Disparities and the Role of Law, in Poverty, Health and Law: Readings and Cases for Medical-­Legal Partnerships, 1, 14 (Elizabeth Tobin Tyler et al. eds. 2011). 24  Parmet, Smith & Benedict, supra note 23, at 12. 25  Bravemen et al., supra note 22, at S5, S11. 26  Case & Deaton supra note 3, at 399. 27  Id. at 429. 28 David M. Cutler, Comments and Discussion, Brookings Papers on Economic Activity 444, 445 (Spring 2017). Cutler states that Case and Deaton’s ‘explanation is certainly correct.’ However, he also argues that the data is nuanced and is not as clear as one would hope. 29 Gillian B. White, US Ranks 23d Out of 30 Developed Countries for Inequality, The Atlantic, Jan. 16, 2017, https://www.theatlantic.com/business/archive/2017/01/wef-­davos-­inequality/513185/. 30  Id.

Social Determinants in the United States   193 should itself be viewed as an important social determinant, one that can both magnify or diminish health disparities. The chapter then considers some defining features of US law that may play a role in creating or perpetuating health disparities both within the United States and between the United States and other OECD nations. The chapter then reviews some recent US initiatives, many but not all taken undertaken via law, to address social determinants, and it concludes by considering the barriers that remain to ameliorating social determinants through law, as well as some reasons for optimism.

2  Law as a Social Determinant Discussions of social determinants do not usually cite law as a root cause of poor health. Instead, they point to income inequality, education, discrimination, the natural and built environment, and social supports, among others.31 Yet, as Scott Burris has reminded us, law stands behind all of these forces: it “helps structure and perpetuate the social conditions that we describe as ‘social determinants,’ and . . . acts as a mechanism or mediator through which social structures are transformed into levels and distributions of health.”32 Laws create and limit access to education and healthcare, they regulate or deregulate environmental pollutants, and they help to structure the distribution of income and the robustness of the social safety net. Laws can also condemn or condone discrimination. In effect, law is itself a social force that can create negative health outcomes. It can also be used to reduce health disparities and improve population health. Law can affect population health in numerous ways. Most obviously, laws create, empower, and restrain state, local, and federal public health agencies; regulate the delivery of healthcare; and seek to promote population health by regulating unsafe practices and activities, such as smoking. Although the impact of such health laws on health outcomes is often unknown or disputed (and in some cases unexamined), studies have shown that numerous public health laws, including workplace safety laws, vaccine laws, motorcycle helmet laws, and environmental laws, among others, have improved population health.33 Health laws, however, are not the only laws that affect health. Once we recall the role of social determinants, we can recognize that laws that affect employment, income inequality, housing, the built environment, and education may also impact health. Indeed, although it is often far harder to determine the effect of what Scott Burris calls “incidental public health laws”34 on health because the pathways are multidirectional

31  See, e.g., Center for Disease Control and Prevention, Social Determinants of Health: Know What Affects Health, https://www.cdc.gov/socialdeterminants/faqs/index.htm#faq1. 32  Scott Burris, From Health Care to the Law of Social Determinants of Health: A Public Health Law Research Perspective, 159 U. Penn. L. Rev. 1649, 1655 (2011). 33  Centers for Disease Control and Prevention, https://www.cdc.gov/phlp/about/index.html (last visited May 21, 2018). 34 Burris, supra note 32, at 1663.

194   Wendy E. Parmet and the impacts are often subtle,35 laws that do not relate to health may have a greater influence on population health than do laws that regulate the healthcare system. This is both because social determinants play a more powerful role in determining population health than clinical interventions and because more people may be touched by a law that helps to broadly shape the environment than one that reaches fewer people.36 Relatedly, laws or interventions that “actually change pathological social conditions have enormous potential to improve both the level and the distribution of health because they address fundamental causes expressed in a wide range of ultimate health states that can be reached via a plethora of pathways across the life course.”37 Thus tax or minimum wage laws that help to reduce economic inequality may have a more pronounced positive impact on health across a population than do laws that regulate the provision of clinical care.38 Two additional ways in which law may serve as a social determinant of health deserves mention. First, as Burris explains, law is a social practice that interacts with individuals and communities in disparate ways.39 Thus when we consider law’s impact on health, we need to look not only at formal law, or “law on the books,” but also at what is known as “law on the street,” the way that law is practiced and experienced. For example, police may enforce drug laws more aggressively in minority neighborhoods, leading to mass incarceration among African Americans.40 Likewise “third-­party policing laws, such as nuisance ordinances and crime-­free rental housing laws, have dangerous consequences for victims of crime” and may add to poor health outcomes.41 Even formal legal rights, such as rights against discrimination, may mean very different things to those who have the economic means to enforce those rights than to those who don’t. As Emily Benfer explains, “low-­income and minority individuals are often disadvantaged in legal forums designed to enforce or interpret the law.”42 Law also operates above the street, in the realm of ideology and norms. Law both reflects and reinforces a society’s norms, prejudices, and perspectives. For example, in exploring racism’s role in creating and perpetuating race-­based health disparities, Lisa C. Ikemoto points to “law’s role in defining difference in ways that have continuously placed the health needs of women of color in the shadows.”43 Law thus works not only to structure institutions, but to shape ways of understanding. Scott Burris, Ichiro 35  Scott Burris, et al., Moving from Intersection to Integration: Public Health Law Research and Public Health Systems and Services Research, 90(2) Milbank Q., 375, 378 (2012). 36  Geoffrey Rose, Sick Individuals and Sick Populations, 30 Int’l J. of Epidemiology 427, 428–31 (2001). 37 Burris, supra note 32, at 1660. 38  David Orentlicher, Healthcare, Health, and Income, 46 J.L. Med. & Ethics 567 (2018). 39 Burris, supra note 32, at 1657. 40  Michelle Alexander, The New Jim Crow, 22 Am. Prospect, Jan.-Feb. 2011, at A 19, 19–21. 41 Emily A. Benfer, Health Justice: A Framework (And Call to Action) for the Elimination of Health Inequity and Social Injustice, 65 Am. U. L. Rev. 275, 327–28 (2015). 42  Id. at 307. 43 Lisa C. Ikemoto, In the Shadow of Race: Women of Color in Health Disparities Policy, 39 U.C. Davis. L. Rev. 1023, 1058 (2006).

Social Determinants in the United States   195 Kawachi, and Austin Sarat explain that law helps to determine racial disparities by “constructing certain behaviors as ‘normal’ rather than discriminatory.”44 They add “[l]ike any truly effective system of regulation, this one works through social processes to facilitate the internalization of rules that millions of people follow every day without legal interference or coercion.” Populations thus internalize legal rules and the norms embedded within them and then act upon those rules and norms to shape social institutions that in turn can influence health.

3  The Structure of US Law Once we recognize the myriad pathways, direct and indirect, through which law acts as a SDOH, an overview of all laws that affect health becomes difficult to imagine—and certainly beyond the scope of this chapter because it would entail discussion of an almost unlimited set of laws—from housing codes to regulations of the family—that on the surface have little relationship to health. Still, beyond considering laws that directly relate to health, which is the subject of much of the rest of this volume, it is worth discussing, even if in only with a broad brush, certain key characteristics of US law that may have an especially pronounced effect on other SDOH. One trait worth noting is the highly individualistic nature of US law. Numerous legal scholars have discussed the strong current of individualism found throughout US law and have contrasted it with the greater commitment to solidarity apparent in European law, as well as in international human rights law.45 This individualism, which can be understood as a deeply embedded, though not consistently dominant, chord in American law is reflected in the fact that “rights” in the United States, especially constitutional rights, are generally understood to be negative rights to noninterference from the state, rather than so-­called positive rights to assistance or support.46 Perhaps for this reason, the international human rights laws that are more fully recognized in the United States are, for the most part, those that recognize negative freedoms, such as the rights to freedom of speech and religion that are codified in the International Covenant of Civil

44  Scott Burris, Ichiro Kawachi, & Austin Sarat, Integrating Law and Social Epidemiology, 30 J.L. Med. & Ethics 510, 517 (2002). 45  See, e.g., Salter Storrs Clark, Individualism and Legal Procedure, 14 Yale L. J. 263 (1905)(discussing individualism in legal procedure); Robert Ahdieh, Beyond Individualism in Law & Economics, 91 B.U.L. Rev. 43 (2011)(critiquing the methodological individualism of law and economics while noting its impact on legal doctrine); Tristin K. Green, Insular Individualism: Employment Discrimination Law after Ledbetter v. Goodyear, 43 Harv. C.R.-C.L. L. Rev. 353 (2008)(discussing the individualism in US employment law); Fazal Kahn, Combating Obesity Through the Build Environment: Is there a Clear Path to Success, 39 J. L. Med. & Ethics 387 (2011)(contrasting the individualism in US public health policy to the approach in Britain). 46  See e.g., Frank B. Cross, The Error of Positive Rights, 48 U.C.L.A. L. Rev. 857 (2001)(arguing that constitutional rights are and should be viewed as essentially negative in nature).

196   Wendy E. Parmet and Political Rights.47 In contrast, the United States does not tend to recognize human rights laws that codify positive liberties, such as the “enjoyment of the highest attainable standard of health” or “an adequate standard of living,” which are affirmed in the International Covenant of Economic, Social, and Cultural Rights.48 (Strong strains of isolationism and rejection of international legal regimes have also helped to undermine US support for international human rights laws.) The profound influence of individualism in US law (and society) helps to explain why public health law in the United States is often viewed as presenting a conflict between individual rights (understood often as rights to non-­interference) and the public good. If rights are viewed as means for protecting an individual’s ability be left alone, free from state interference, then public health laws are likely to be seen as abridgements of those rights.49 More subtly, this strong version of individualism prompts Americans and US law to view many health problems, such as obesity or sexually transmitted infections, as questions of individual responsibility.50 By so doing, law itself helps to obscure the role of social determinants and reduce the political pressure to redress them. The individualism in American law may also explain the heavy reliance on informational disclosures and other forms of “libertarian paternalism” relied upon by public health policymakers.51 Rather than limit exposures to hazardous substances or activities, American regulations often take the form of mandatory disclosures that seek to provide individuals with information that they can, at least theoretically, use to decide whether or not to avoid a health risk.52 While this approach can be seen as being more respectful of individual “choice,”53 it is also often less effective at reducing health risks. It can also increase health disparities as highly educated populations may be in a better position to act upon warnings and disclosures than lower SES populations. Perhaps more importantly, the individualism that American law reflects and reinforces may support America’s underinvestment, compared to other wealthy countries, in social welfare programs, such as early childhood education programs, that may

47  G.A. Res. 2200A (XXI), U.N. GAOR, Supp. No. 16, U.N. Doc. A/6316 (Dec. 16, 1966). 48  G.A. Res. 2200A (XXI), UNGAOR Supp. No. 16, U.N. Co. A/6316 (Dec. 16, 1966). 49  This perspective takes the circumstances that an individual faces as given, and does not consider that individual autonomy may be enhanced when individuals face less arduous circumstances. Once it is accepted that individuals may be ‘free’ to exercise more choices if they face fewer health risks, the inherent conflict between public health laws and individual freedom begins to fade. See Wendy E. Parmet, Paternalism, Self-­Governance, and Public Health: The Case of E-­Cigarettes, 70 U. Miami L. Rev. 879, 901-­916 (2016). 50  Adam Benforado & Jon D. Hanson, Seeing Bias: Discrediting and Dismissing Accurate Attributions, in Ideology, Psychology and the Law, 453 (2012). 51  See generally, Richard Thaler & Cass Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness (2008). 52  David Adam Friedman, Public Health Regulation and the Limits of Paternalism, 46 Conn. L. Rev. 1687, 1701 (2014). 53  This approach has also been subject to criticism for violating rights of free speech. E.g., National Institute of Family and Life Advocates v. Bercerra, US, 138 S.Ct. 2361 (2018); see chapter by Parmet on “Noncommunicable Diseases,” in this volume.

Social Determinants in the United States   197 redress the SDOH.54 Indeed, when one compares total spending on social services in the US to that in the E.U., the US’ disproportionate expenditures on health care take on a different light. The US spends far more on health than do EU countries, but far less on other social programs.55 The individualism and emphasis on negative rights to autonomy, rather than on the full panoply of human rights, may also explain a paradox that has critical implications for population health. On the one hand, as noted above, individual rights play an important role in the US legal system, hampering the government’s ability to take actions that aim to redress negative SDOH. For example, the courts have concluded that the First Amendment protects the rights of corporations that seek to advertise and market unsafe products.56 The Supreme Court also reads the Second Amendment as creating an individual right to possess firearms.57 Hence many efforts to use law to improve the SDOH face constitutional challenges. On the other hand, as discussed above, the Constitution does not demand that government act to improve social determinants or reduce health inequities. Thus the non-­recognition of positive rights has led the Supreme Court to conclude that individuals have no constitutional right to government supported health care,58 or even to protection from domestic violence.59 As the Supreme Court has explained, ‘Nothing in the language of the Due Process Clause itself requires the State to protect the life, liberty and property of its citizens against invasion by private actors. The clause is phrased as a limitation on the State’s power to act, not as a guarantee of certain levels of safety and security.’60 Likewise, the Court has made clear that the Constitution does not prohibit racially disparate outcomes, permitting the continuation of policies that adversely affect minorities.61 Rather the Constitution only prohibits intentional acts of racial or other prohibited forms of discrimination. As a result, in a legal system that is heavily focused on individual rights, legal rights may prove a more powerful weapon for those who contest policies aimed at improving the SDOH than a tool for those who seek to reduce disparities and redress social determinants. This tilt toward negative rights extends beyond the Constitution to many realms of US law. For example, justicibility and immunity rules make it extremely difficult for individuals to hold states accountable in court for failing to protect them from health threats.62 Indeed, the civil and criminal litigation stemming from the poisoning of the water in Flint, Michigan may be the exception that proves the rule. In 2011 the City of 54  Howard K. Koh & Anand K. Parekh, Editorial—Toward a United States of Health: Implications of Understanding the US Burden of Disease, 319 JAMA 1438, 1439 (2018) (discussing US underinvestment as compared to other countries in social spending). 55 Id. 56  See National Institute of Family and Life Advocates v. Bercerra, 138 S.Ct. 2361 (2018); Lorillard Tobacco Co. v. Reilly, 533 US 525, 585–89 (2001). 57  See District of Columbia v. Heller, 554 US 570, 591 (2008). 58  See Harris v. McCrae, 448 US 297, 326 (1980). 59  See Castle Rock v. Gonzales, 545 US 748, 757 (2013). 60  DeShaney v. Winnebago City Dep’t Soc. Servs., 489 US 189, 195 (1989). 61  Washington v. Davis, 426 US 229 (1976). 62  Wendy E. Parmet, Populations, Public Health and the Law 65–68 (2009).

198   Wendy E. Parmet Flint, which is largely poor and African-­American, faced a financial emergency due to the loss of its manufacturing base. In response, Michigan Governor Scott Walker appointed an Emergency Manager (EM) who reported to the Governor and was charged with resolving the city’s financial problems. With that charge, the EM decided in 2014 to switch the city’s water supply to the Flint River in order to save money. Almost immediately, residents began to complain about the foul-­smelling water. The EM and state environmental officials, however, offered reassuring statements, even as people began to die from Legionnaire’s disease.63 In 2015, physicians in Flint and outside experts reported on high levels of lead in the water and in the blood of the city’s children. Eventually, it became apparent that the city’s water supply was unsafe, and that many of the city’s residents had been exposed to dangerously high levels of lead. Investigations found that officials had repeatedly misled the public. Civil suits were filed, but they faced numerous challenges. Under Michigan law, the EM was totally immune from civil liability, and state health and environmental officials could only be held accountable if they were found to be grossly negligent.64 Federal constitutional claims also faced challenges because there is no constitutional right to clean water or health protection, and although the race and poverty of the city’s residents almost certainly helped to explain why officials acted as they did, the type of evidence of racial animus that is needed to bring an Equal Protection claim was difficult to produce. In Flint, however, there was sufficient evidence that officials had misled the public and altered test results for prosecutors to bring criminal charges.65 But if officials had simply failed to protect poor minority residents from unsafe water, there likely would have been no path for legal accountability. To the contrary, by allowing the Governor to abridge the political rights of Flint’s residents by appointing an EM, and by providing broad immunities from officials, Michigan created a recipe for the exacerbation of health disparities.

4  Addressing Social Determinants The US lacks a comprehensive, overarching policy to address the SDOH. Nevertheless, public health practitioners and policymakers in the US have issued numerous calls to improve the nation’s health and lower health care costs by addressing the SDOH. For example, the Federal Government’s Healthy People 2020 project, a collaborative effort among federal agencies to set ten year health goals for the nation, underscores the importance of social determinants and includes as one of its four goals, to ‘create social

63  Interim Report of the Flint Water Crisis Investigation, 2, 6–10, https://www.michigan.gov/ documents/ag/Flint+Water+Interim+Report_575711_7.pdf. 64  M.C.L.A. ch. 691.1407 § 7. 65  Amy Haimerl & Abby Goodnough, 6 More State Workers Charged in Flint Water Crisis, New York Times (Jul. 29, 2016), https://www.nytimes.com/2016/07/30/us/flint-­michigan-­water-­crisis.html.

Social Determinants in the United States   199 and physical environments that promote good health for all.’66 Likewise, the federal National Prevention Strategy lists ‘healthy and safe community environments,’ and the ‘elimination of health disparities,’ as among its four ‘strategic directions.’67 State and local health departments, as well as leading national foundations, such as the Robert Wood Johnson Foundation, have emphasized the importance of social determinants and reducing health disparities.68 These calls have led to numerous innovations and interventions, many at the community level, and some financed by federal health care dollars, that aim to improve the SDOH. In particular, the Affordable Care Act (ACA) established new funding models, such as accountable care organizations (ACOs) that were designed to incentivize providers to support an array of non-­clinical factors that affect patient health, as well grants to support partnerships between providers and communities that are designed to improve community health and reduce health disparities.69 Other efforts rely less on the health care system, and seek to improve community health through a variety of non-­health care based initiatives.70 For example, often with grant support, including through the no-­longer extant CDC Healthy Communities Program,71 many localities have established community gardens, farmers’ markets, bike paths, and other initiatives designed to improve the built environment and community health. Although such efforts may well have a positive impact on the health of discrete populations, they face significant limitations. First, they are usually small in scale, and do not even try to serve broad populations. Second, many rely on grants, leaving their long-­term sustainability uncertain.72 Third, many operate relatively ‘downstream,’ for example by providing individuals with chronic health care needs, housing support, access to transportation, or food. Although such efforts are likely beneficial to the individuals they serve, they do not address the broad, upstream issues such as income inequality, wage stagnation, environmental degradation or poor education73 (not to mention the underlying individualistic ideological underpinnings of US law). Redressing such upstream issues would require very different types of policy interventions, most of which lie beyond the competence and jurisdiction of the health and community organizations that focus on the SDOH qua SDOH. As the Robert Wood Johnson Foundation explains, ‘Building a Culture of Health’ requires cross-­sector 66  Office of Disease Prevention & Health Promotion, About Healthy People, HEALTHYPEOPLE. GOV, https://www.healthypeople.gov/2020/About-­Healthy-­People. 67  US Department of Health & Human Services, National Prevention Strategy, Surgeon General of the United States, https://www.surgeongeneral.gov/priorities/prevention/strategy/index.html. 68  Charity Scott, Incorporating Lawyers on the Interprofessional Team to Promote Health and Health Equity, 14 Indiana L. Rev. 54 (2017). 69  For a further discussion, see Peter D. Jacobson & Wendy E. Parmet, Public Health and Health Care: Integration, Disintegration, or Eclipse, 46 J.L. Med. & ethics 940 (2018). 70  These efforts include laws that aim to reduce the incidence of noncommunicable diseases. For a further discussion, see chapter by Parmet on “Noncommunicable Diseases,” in this volume. 71  Centers for Disease Control and Prevention, Healthy Communities Program (2008–2012), https://www.cdc.gov/nccdphp/dch/programs/healthycommunitiesprogram/index.htm. 72  See Jacobson and Parmet, supra note 69. 73 Id.

200   Wendy E. Parmet collaboration that encourage ‘people to see the connection to health and well-­being within their work, whether in education, transportation, community development, law enforcement, business or other fields.’74 To promote such collaborations, and in recognition of the fact that protection of health often requires actions taken outside of the health sector, the Centers for Disease Control and Prevention (CDC) and many US health departments have promoted so-­called Health in All Polices (HiAP). CDC defines HiAP as a ‘collaborative approach that integrates and articulates health considerations into policymaking across sectors.’75 In effect, HiAP seek to involve policymakers outside of health, for example, public works or development agencies, to consider the health impact of their decisions. Although the embrace of HiAP can promote cross-­sectoral collaboration, and recognition of the health impact of non-­health policies, HiAP face political and jurisdictional problems. In many areas, there is a lack of funding and political support.76 In addition, statutes may limit decision-­makers’ ability to weigh health heavily in their decision-­making. For example, when making permitting decisions, local planning boards may be limited by law to consider the impact of traffic, pollution, or aesthetics; the impact of their decisions on chronic diseases and long-­term population health may simply be outside the scope of their lawful authority.77 Moreover, in many situations the health impact of specific policy decisions is unknown. For example, although there is substantial evidence regarding the importance of exercise to health, the health impact of siting a park in a particular location is probably far less certain. Perhaps most critically, once we accept the key role that broad economic and social forces play in determining population health, it becomes apparent that the most important policies for health may be those that deal with macro-­level economic issues that are not readily amenable to the type of micro-­level cross sector collaborations envisioned by HiAP or even a culture of health. Many cities have tried to address such issues, for example, by raising the minimum wage.78 But significant alterations in the broader economic and social environment, and the most upstream roots of diseases of despair, may well require reforms at the national level. Given the importance of wealth and inequality to health, federal tax, minimum wage and labor laws, to give just a few examples, may have a more significant impact on population health than any of the local interventions that have been undertaken to specifically address the SDOH as such. Unfortunately, at the national level, there is little reason to believe that conditions will improve, and disparities will be reduced in the near term. Perhaps most importantly, many political scientists have noted that the US political system, especially at the federal 74  Robert Wood Johnson Foundation, Fostering Cross-­Sector Collaboration to Improve Well-­Being, https://www.rwjf.org/en/cultureofhealth/taking-­action/fostering-­cross-­sector-­collaboration.html (last visited Aug. 9, 2019). 75  Health in All Policies, Centers for Disease Control and Prevention (June 9, 2016), https:// www.cdc.gov/policy/hiap/index.htm. 76 Id. 77  Rene Bowser & Lawrence O. Gostin, Managed Care and the Health of a Nation, 72 S. Cal. L Rev. 1209, 1245–46 (1999). 78  Grace Donnelly, The Minimum Wage will Increase on January 1 for 18 States and 20 Cities, Fortune, Dec. 20, 2017, http://fortune.com/2017/12/20/minimum-­wage-­increases-­jan-­2018/.

Social Determinants in the United States   201 level, faces numerous serious challenges, including high levels of partisan polarization and institutional gridlock that make it difficult for the federal government to enact legislative solutions to address many adverse social determinants.79 In addition, many of the policies of the Trump Administration are likely to exacerbate underlying socio-­economic and racial inequalities. For example, the 2017 tax cuts that the Administration supported tilt heavily toward the upper-­income taxpayers, and may increase income inequality, which is a negative social determinant.80 The tax cuts are also expected to increase the federal deficit, which may leave the government with fewer resources to support services for low-­income populations.81 At the same time, the Administration’s crackdown on immigration has created new barriers to immigrants accessing health care and other social benefit programs, including food support and tax credits,82 raising the risk of exacerbating health disparities. Similarly, proposed changes in environmental,83 housing84 and other policies threaten85 to increase adverse social determinants, or at least, undermine efforts to improve the social and environmental conditions faced by vulnerable populations. Focusing only on health, the Trump Administration has promoted policies that threaten many of the innovations and reforms that are have been attempted to respond to negative social determinants. For example, citing SDOH, the Trump Administration recently invited states to require recipients to work as a condition of receiving Medicaid.86 Should this policy be implemented (an early version was struck down by a court), it will reduce access to health care for many vulnerable individuals.87 Some of the regulatory changes that the Administration has implemented in order to unravel the 79  Cynthia R. Faraina, Congressional Polarization: Terminal Constitutional Dysfunction?, 115 Columb. L. Rev. 1689 (2015). 80  Chuck Marr, Brendan Duke & Chye-­Ching Huang, New Tax Law is Fundamentally Flawed and Will Require Basic Restructuring, Center on Budget and Policy Priorities (Apr. 9, 2018), https://www. cbpp.org/research/federal-­tax/new-­tax-­law-­is-­fundamentally-­flawed-­and-­will-­require-­basic-­ restructuring; Dylan Scott & Alvin Chang, The Republican Tax Bill Will Exacerbate Income Inequality in America, Vox (Dec. 4, 2017), https://www.vox.com/policy-­and-­politics/2017/12/2/16720952/ senate-­tax-­bill-­inequality. 81  Marr, Duke, & Huang, supra note 80. 82  Wendy E. Parmet & Elisabeth Ryan, New Dangers for Immigrants and the Health Care System, Health Affairs (Apr. 20, 2018), https://www.healthaffairs.org/do/10.1377/hblog20180419.892713/full/. 83  Vann R. Newkirk II, Trump’s EPA Concludes Environmental Racism is Real, The Atlantic (Feb. 28, 2018), https://www.theatlantic.com/politics/archive/2018/02/ the-­trump-­administration-­finds-­that-­environmental-­racism-­is-­real/554315/. 84  Glenn Thrush, Under Ben Carson, HUD Scales Back Fair Housing Enforcement, N.Y. Times (Mar. 28, 2018), https://www.nytimes.com/2018/03/28/us/ben-­carson-­hud-­fair-­housing-­discrimination.html; Tracy Jan, Caitlin Dewey, & Jeff Stein, HUD Secretary Ben Carson to Propose Raising Rent for Low-­ Income Americans Receiving Federal Housing Subsidies, Wash. Post (Apr. 25, 2018), https://www. washingtonpost.com/news/wonk/wp/2018/04/25/hud-­secretary-­ben-­carson-­to-­propose-­raising-­rent-­ for-­low-­income-­americans-­receiving-­federal-­housing-­subsidies/?utm_term=.de4b71a5c3cf. 85  Rebecca Moss, Trump’s Labor Department Eviscerates Workplace Panels, ProPublica (Mar. 29, 2018), https://www.propublica.org/article/trumps-­labor-­department-­eviscerates-­workplace-­ safety-­panels. 86  Stewart v. Azar, 2018 WL 3203384 *3–4 (D.D.C. 2018). 87  Id. at *20.

202   Wendy E. Parmet Affordable Care Act may also lead to increases in the number of uninsured, although whether they will do so remains uncertain.88 Although the provision of health care is not the most important SDOH, lack of access for marginalized populations can only increase disparities.

5  Looking Ahead Despite these developments, there are some reasons for optimism, especially at the local, community level. As noted above, there are many initiatives around the country that attempt to reduce health disparities and address the SDOH. There are also some reasons to believe that social cohesion may be growing in productive ways, at least in some communities. For example, in his book about the opioid crisis, Sam Quinnes describes how after years of ignoring the problem, families and communities have begun to come together to raise awareness and combat addiction.89 Whether or not their specific efforts bear fruit, the very act of organizing and working at the community level to counter the crisis represents a step towards rebuilding the once-­frayed communities whose evisceration had helped to spawn the diseases of despair. It also offers the hope that policymakers will now begin, however hesitatingly, to adopt evidence-­based interventions. Thus some communities are now beginning to explore offering safe-­injection sites, despite federal prohibitions against doing so.90 Similar signs of community-­building and engagement are apparent with regard to other issues. Consider gun violence, which kills over 30,000 Americans each year and injuries far more.91 For years, the US Congress has failed to act on the issue, even in the face of polls showing that a large majority of the public supports increased regulation.92 Congressional inaction doesn’t appear to be changing; but following the mass shooting 88  Ricardo Alonso-­Zaldivar, US Clings to Health Coverage Gains Despite Political Drama, Wash. Post (May 22, 2018), https://www.washingtonpost.com/business/us-­clings-­to-­health-­coverage-­gains-­ despite-­political-­drama/2018/05/22/2d35957c-­5d76-­11e8-­b656-­236c6214ef01_story.html?utm_term=. a767fcb02587; Sara R. Collins et al., First Look at Health Insurance Coverage in 2018 Finds ACA Gains Beginning to Reverse, to the Point: Quick Takes on Health Care Policy and Practice, The Commonwealth Fund (May 1, 2018), http://www.commonwealthfund.org/publications/blog/2018/ apr/health-­coverage-­erosion. 89  Sam Quinones, Dreamland: The True Tale of America’s Opiate Epidemic 347–48 (2016). 90 Paige W. Cunningham, The Health 202: Supervised Injection Facilities Are Illegal in the United States. These Cities Want to Open Them Anyway, Wash. Post (Apr. 30, 2018), https://www. washingtonpost.com/news/powerpost/paloma/the-­health-­202/2018/04/30/ the-­health-­202-­supervised-­injection-­facilities-­are-­illegal-­in-­the-­united-­states-­these-­cities-­want-­to-­ open-­them-­anyway/5ae5dbc630fb043711926901/?utm_term=.966a6f4ce190. 91  Centers for Disease Control and Prevention, National Center for Health Statistics, All Injuries, https://www.cdc.gov/nchs/fastats/injury.htm (last visited July 13, 2018). 92  Seung Min Kim et al., Congress Leaves Washington with No Action on Gun Control or School Safety, Chi. Trib. (Mar. 1, 2018), http://www.chicagotribune.com/news/nationworld/politics/ct-­ congress-­gun-­control-­school-­safety-­20180301-­story.html.

Social Determinants in the United States   203 at Marjory Stoneman High School in Parkland, Florida on February 4, 2018, students from that school and others across the nation organized, lobbied, and led mass protests around the country.93 Their actions led directly to new gun regulations in Florida,94 as well as new pressure for gun regulations in many other states.95 Equally important, their movement has emphasized voter registration and political activism; in other words, the type of grass roots efforts and democratic engagement that can build community and lead to legal changes. In this way they are strengthening the social determinants of community, and working towards altering the social determinant of law. For the moment, however, these glimmers of community rebuilding, social mobilization and law reform remain oases in a dessert of deleterious social determinants. Whether they can expand and achieve sufficiently significant changes in US law to improve the SDOH and reduce the shameful health disparities experienced in the US remains to be seen.

93  Peter Jamison, et al., ‘Never again!’ Students Demand Action Against Gun Violence in Nation’s Capital, Wash. Post (Mar. 24, 2018), https://www.washingtonpost.com/local/march-­for-­our-­lives-­ huge-­crowds-­gather-­for-­rally-­against-­gun-­violence-­in-­nations-­capital/2018/03/24/4121b100-­2f7d-­ 11e8-­b0b0-­f706877db618_story.html?utm_term=.cdfa0dac67b2. 94  Maggie Astor, Florida Gun Bill: What’s in It, and What Isn’t, N.Y. Times (Mar. 8, 2018), https:// www.nytimes.com/2018/03/08/us/florida-­gun-­bill.html. 95  Curt Anderson et al., Florida Governor Signs School Safety Bill That Raises Minimum Rifle Age, Creates Handgun Program for School Employees, PBS News Hour (Mar. 9, 2018 4:01 p.m.), https://www. pbs.org/newshour/nation/florida-­governor-­signs-­school-­safety-­bill-­that-­raises-­minimum-­rifle-­age­creates-­handgun-­program-­for-­school-­employees.

pa rt I I I

T R E AT I NG I L L H E A LT H

ACCESS TO THE H E A LT H C A R E SYSTEM

chapter 10

I n troduction to Access to He a lthca r e André den Exter, Mark A. Hall, Allison K. Hoffman, and Keith Syrett

1 Introduction The following two chapters (by Allison K. Hoffman and Mark A. Hall and by Keith Syrett and André den Exter) survey a variety of legal and health policy issues bearing on access to care. Principally, we cover the types and sources of funding for healthcare services and discuss how health insurance programs and national health systems are structured and regulated in ways that may enhance or impede access to medical care for some or all of the population. Also discussed are legal doctrines that bear on patients’ rights to receive treatment and the obligations of physicians and hospitals to provide treatment. Comparison between the United States and European countries on the topic of access to care will inevitably be incomplete and selective, but the stark distinctions readily reveal fundamental philosophical differences in how the matter is conceived, resulting in much larger disparities in access to care in the United States than elsewhere. Considering the much greater extent to which healthcare finance is privatized in the United States, private law doctrine is much more relevant there than in many European countries. Contrariwise, European countries often have constitutional or statutory underpinnings for rights of access to care that are noticeably absent in the United States. Furthermore, they have financing and regulatory systems that seek to create universal access to care, also lacking in the United States. Even with the differences, the two share a healthy dose of public law from regulatory sources, although the particulars differ a great deal based on the structure of healthcare financing and delivery, and the powers of governmental authorities across various countries. To facilitate shared understanding, we begin by specifying some of the terms we use in discussing these topics. Access to care is the primary health policy goal (and problem) we

210   André den Exter et al. address. Access consists of both availability (are providers in reach and can particular services be secured by those who need them?) and affordability. Affordability is determined by how a country helps people to pay for care, including by means of healthcare coverage that entitles or enables a patient to access affordable care. Coverage is commonly provided by some type of insurance—either public (social) or private insurance—that pays for the costs of treatment. But care can be arranged and paid for more directly, without insurance, through a program that simply operates facilities or clinics serving an eligible population, as in the United Kingdom. Typically, these direct-access programs are run by government (e.g., the Veterans Health Administration [VHA] or Indian Health Service [IHS] in the United States), but they may also be privately operated, as for instance the free clinics and ‘safety net’ hospitals that serve the uninsured in the United States. When operated by government, health systems can be referred to as being ‘socialized’. Although that term is used much more in the United States than in Europe, in either setting the question is not whether some aspect of healthcare is socialized or not. Rather, it is which aspect of healthcare is socialized. There is a key distinction between socialized insurance and socialized medicine—which is precisely the distinction just drawn between public insurance versus direct access through government-run hospitals and clinics. Thus, in Europe, as discussed in the subsequent chapter, some countries have social insurance systems that reimburse physicians and clinics, while others, such as the United Kingdom, Italy, and Spain, have socialized medical systems, where most facilities and clinics are operated by the government and funded through global budgets. In the United States, both exist, and neither is universal. A small number of people have access through the government delivery systems referred to previously (VHA and IHS), which can be thought of as isolated pockets of socialized medicine in the United States. A larger, but still limited, share of the population is aided by government financing, which ranges in form from Medicare (for elderly and disabled people) and Medicaid (for those in or near poverty), which are social insurance programs, to governmental subsidies to help low- to middle-income people buy private insurance and access medical care. Such subsidies are a core part of the 2010 reform in the United States, entitled the Patient Protection and Affordable Care Act, but sometimes referred to as ‘Obamacare’ since President Barack Obama signed it into law. Finally, lest naïve readers think that each country falls into a discrete category, recognize also that European countries have fairly extensive development of private insurance or privately paid direct access providers alongside their social coverage systems. In most countries, these systems complement or supplement a public financing system. With these basic categorizations in mind, this section comprises two chapters, the first providing an overview of access to care in the United States and the second addressing access to care in Europe. These chapters diverge in their emphasis, and it is this divergence, perhaps more than any direct comparisons, that is most instructive in comprehending the subject-matter. Based on the issues raised in these two chapters, we offer several key takeaways. First, notions of a positive right to healthcare and the idea of social solidarity are highly influential, although not always decisive, in the discourse concerning health

Introduction to Access to Healthcare   211 systems in Europe. Neither of these ideas is fundamental in the United States, however, which begins with the more libertarian foundations of ‘negative’ liberties (the right to be left alone) and freedom of contract. Second, allocation of limited healthcare resources (aka. ‘rationing’) in Europe happens primarily in relation to which benefits are covered, but in the United States rationing is based more on which people are covered. Because Europe has (more or less) universal coverage, its access issues focus first and foremost on central determinations of the treatments and services that are in and out of coverage and what a ‘right to health’ demands. A discussion about the appropriateness of rationing in European countries takes various forms. The first is through a definition of the ‘basket’ of entitlements, which is the range and scope of services and treatments provided within a system. European systems, for instance, frequently do not cover novel treatments that have not been fully tested, and they often exclude more expensive options that do not work substantially better than less expensive options. These restrictions can be set by statute, by lists, or through explicit agency recommendations based on principles such as costeffectiveness, nondiscrimination, and solidarity. The second resource allocation mech­ an­ism is through legal challenges in instances where benefits are denied, which can sometimes work to expand the treatments and services that must be covered by court order. A further, less explicit form of rationing is waiting time, which can be lengthy in some European countries, even for core covered services. In contrast, in the United States, access issues first and foremost raise questions of who is entitled to coverage instead of what is covered. Because access tends to be fairly broad when coverage is available, problems arise primarily from lack of insurance altogether. Excluded or limited benefits are sometimes part of the conversation, but this is almost never the case at the core of coverage. As a result, rationing is not addressed as explicitly in the United States as in Europe, which perhaps explains why the United States finds it more difficult to establish restraints that make coverage more affordable. Third, the United States has evolved convoluted private law obligations to care because of the fundamental problems of heterogeneity in the financing system, where some people have lesser paying forms of coverage or none at all. The absence of more comprehensive obligations of public healthcare systems has turned the focus more to obligations of private physicians and hospitals, determined under common and statutory law. Thus, for instance, although physicians are free to decline patients even in emergencies, hospitals are not, but for both types of providers, once they begin caring for a patient they must complete treatment unless proper measures are followed to terminate the treatment relationship. Fourth, geography and national boundaries pose particular challenges in Europe, where people regularly move across geopolitical lines to receive healthcare, and countries must balance rights claims with allocation of limited resources. The United States also confronts some geographical issues in that some aspects of healthcare access vary state by state, but these variations are much less dramatic than in Europe and are much less likely to motivate movement in pursuit of access.

212   André den Exter et al. One area of commonality is that in both the United States and Europe, citizenship or some other status of lawful residence often determines eligibility for coverage. Although European countries provide near-universal coverage for legal residents, many, as in the United States, do not do so for undocumented immigrants or those without established residency. A further aspect common to both the United States and Europe (and, for that matter, health systems across the globe) is the challenge of increasing costs. Although the United States spends twice as much per capita as the average in countries of the Organisation for Economic Cooperation and Development (OECD), it, like all other countries, is struggling over the increasing costs of care, especially in light of new prescription drugs and medical technologies with high prices. And all countries are experimenting with similar tools to try to slow the spending—from patient cost sharing, to health technology assessment, to restricted formularies, and to various levels of competition, even though the latter has not proved especially effective. In sum, the comparison between Europe and the United States reveals two very different approaches to dealing with the fundamental question of access to healthcare. Perhaps most significantly, although there is variability across Europe, it starts from a baseline of near-universal access to healthcare that is guaranteed through various legal tools markedly absent in the United States, thus ensuring a greater degree of health equity.

chapter 11

The A m er ica n Pathol ogy of I n equ ita bl e Access to M edica l Ca r e Allison K. Hoffman and Mark A. Hall

1  Overview of Access to Health Care in the United States What most defines access to healthcare in the United States may be its stark inequity. Daily headlines in top newspapers paint the highs and lows. Articles entitled: “We Mapped the Uninsured. You’ll Notice a Pattern: They Tend to Live in the South, and They Tend to be Poor”1 and op-­eds with titles like “Do Poor People Have a Right to Health Care?”2 and “What It’s Like to Be Black and Pregnant when You Know How Dangerous That Can Be”3 run side by side with headlines touting “The Operating Room of the Future”4 and advances in gene therapy that promise cures to everything from vision loss to cancer, accompanied by high six-­figure price tags.5 Americans’ claims that they have access to the best medical care in the world are correct. Equally true are the claims that the system is broken. 1  Quoctrung Bui & Margot Sanger-­Katz, We Mapped the Uninsured. You’ll Notice a Pattern, NY Times (Oct. 30, 2015). 2  The Editorial Board, Do Poor People Have a Right to Health Care? NY Times Editorial (July 7, 2018). 3  Dani McClain, What It’s Like to Be Black and Pregnant When You Know How Dangerous That Can Be, The Nation (March 6, 2017). See also Linda Villarosa, Why America’s Black Mothers and Babies Are in a Life-­or-­Death Crisis, NY Times (April 11, 2018). 4  Laura Landro, The Operating Room of the Future, Wall Street J. (May 28, 2018). 5  Peter Loftus, Drug Firm Spark Therapeutics Will Charge $850,000 for Vision-­Loss Gene Therapy, Wall Street J. (Jan. 3, 2018).

214   Allison K. Hoffman and Mark A. Hall This chapter maps out the complex picture of access to medical care in the United States and reflects on how variable access illustrates, among other things, an American ambivalence about health solidarity. This chapter first considers healthcare financing as one critical element that defines access to medical care and describes the multifaceted US healthcare financing structure, predominated by public insurance programs for select populations and regulated private insurance for others. Second, this chapter describes how access is equally shaped by legal requirements that create treatment obligations for doctors or hospitals, regardless of how someone pays for care. There is no constitutional right to health in the United States, but various layers of statutory and common law have created some guarantees. That said, despite efforts to increase access over the past decades, the United States is still extraordinary, as compared to peer nations, on the unevenness of access to medical care among its population. Even after the passage of the Patient Protection and Affordable Care Act in 2010 (ACA), nearly 9% of the total population remains uninsured and, in many cases, lacks access to basic medical care.6 Even many people with insurance struggle to pay the cost-­ sharing necessary to use medical care. And some cannot get access to doctors, whether insured or not. Despite legal obligations, providers still enjoy considerable autonomy over whom they care for and when. And most laws that create treatment obligations are silent about who will pay for that care, possibly leaving patients who receive life-­saving care with crippling medical debt. Scholars attribute the perpetuation of this uneven picture to a range of factors: or­gan­ ized medicine and other interest groups’ resistance to public insurance, broader ideological skepticism of government or of new taxes, path dependency, and inertia.7 At its root, disparities in access draw from the fact that the notion of a right to healthcare does not have and never has had the same salience in the United States as in its peer nations. In turn, the existing system undermines health solidarity by reproducing a norm where access to healthcare is not universal. Even though US laws have facilitated or guaranteed access to care for some people and in some circumstances, the absence of consensus on any coherent guiding principle is evident in the remaining gaps.

2  Financing and Health Insurance as Point of Entry The United States spends more on medical care than on any other single good or service. In 2017, US medical care spending accounted for nearly 18% of the gross domestic product 6  United States Census, Health Insurance Coverage in the United States: 2017 5 (2018), https://www. census.gov/content/dam/Census/library/publications/2018/demo/p60-­264.pdf (last accessed July 30, 2019). 7  See Timothy Jost, Access to Health Insurance and Health Benefits in Oxford Handbook of US Health Law (I. Glenn Cohen, Allison K. Hoffman, and William M. Sage, eds. 2017).

The American Pathology of Inequitable Access to Medical Care   215 and $3.5 trillion in total spending.8 Relative to its wealth, the US spends well more than other nations of the Organisation for Economic Cooperation and Development (OECD)—twice as much as a comparable country average and 25% more than its closest peer nation, Switzerland.9 Yet most of these other countries outperform the United States on key health outcomes, including life expectancy and infant mortality, as well as subjective patient experience.10 A major part of the problem in the United States is outsized prices that impede affordability and thus access. Although utilization in the United States is generally not more than elsewhere, prices for medical services and goods in the United States are many time those in peer nations.11 For the most part, the United States has no central, regulatory price controls. Some states, like Maryland, have experimented with price controls. Each insurance program, public and private, must separately negotiate rates with every provider. And the largest public insurance program, Medicare, has statutory restrictions that prevent the government from regulating prices on prescription drugs. The US strategy has been to look to the market and competition to check prices, but this strategy has not worked well.12 Another part of the problem can be explained by how people access, and do not access, medical care in the United States. A century ago, people who needed medical care simply paid for it. After a century of medical advances and the growth of a medical industry of hospitals, drugs, and devices, medical care had become so expensive that very few people could afford to pay for it out of pocket. For most people, having good health insurance has become a necessary gateway to accessing good medical care,13 and, for many doctors and hospitals, it is the key to being paid. For most Americans, the first request when they go to a doctor or hospital is for their insurance card. In fact, some hospitals will not admit someone for care without proof that they can pay. The average price of normal childbirth in the United States is more than $10,000 and coronary bypass surgery costs on average more than $78,000.14 The

8  Centers for Medicare & Medicaid Services, NHE Fact Sheet Historical NHE 2017, https://www. cms.gov/research-­statistics-­data-­and-­systems/statistics-­trends-­and-­reports/nationalhealthexpenddata/ nhe-­fact-­sheet.html. 9  Bradley Sawyer and Cynthia Cox, Peterson-­Kaiser Health System Tracker, How Does Health Spending in the US Compare to Other Countries? (2018), https://www.healthsystemtracker.org/ chart-­collection/health-­spending-­u-­s-­compare-­countries/#item-­start. 10  Eric C. Schneider et al., The Commonwealth Fund, Mirror, Mirror 2017: International Comparison Reflects Flaws and Opportunities for Better US Health Care 8–9 (2017). Marian F. MacDorman et al., Is the United States Maternal Mortality Rate Increasing? Disentangling Trends from Measurement Issues, 128 Obstet. Gynecol. 447, 453 (2016). 11  Gerald F. Anderson et al., It’s Still The Prices, Stupid: Why The US Spends So Much on Health Care, And a Tribute to Uwe Reinhardt, 38 Health Aff. 87 (2019). 12 Allison K. Hoffman, Health Care’s Market Bureaucracy, 66 UCLA L. Rev. 4(2019). 13 John A. Nyman, The Value of Health Insurance: The Access Motive, 18 J. Health Econ 141 (1999). 14  Sarah Kliff & Soo Oh, America’s Health Care Prices are Out of Control. These 11 Charts Prove it, Vox (2018), https://www.vox.com/a/health-­prices (last accessed July 25, 2018).

216   Allison K. Hoffman and Mark A. Hall annual per person spending on medical care in 2017 was greater than $10,000.15 Compare this amount to the median household income, which before taxes was $57,652 in 2017.16 The lowest quartile of earners had a mean income of just over $13,000 and the second quartile of just over $35,000.17 For this majority of Americans, paying for medical care out of pocket is not an option. This section will thus primarily focus on how insurance—public and private—defines access to care. Section 2.1 briefly describes what people without health insurance do to pay for care. Section 2.2 offers an overview of private and public insurance as a primary gateway to access medical care, including how it developed, how it evolved with the passage of the ACA, and how it has continued to evolve after the ACA’s passage. Section 2.3 explains how the heterogeneity in health insurance in the United States exacerbates inequitable access, even among the insured. The final Section 2.4 considers why this inequitable picture remains and whether it will persist in the future.

2.1  The Uninsured and Access to Medical Care Even with the primacy of health insurance, there are a few other ways that people can access care without health insurance to pay for it. First, some people receive charity care through free clinics, volunteer services offered by providers, fee waivers from hospitals’ charity care programs, or drug assistance programs. All of these types of charity care, however, are contingent and extremely limited in scope. Second, for select populations and some low-­income patients, the government provides care directly, such as through the Indian Health Service (IHS) for more than 2.5 million federally recognized American Indians and Alaska Natives18; the Veterans Health Administration (VHA) for more than 9 million people who served for a designated period of time in the active military, naval, or air service and were not dishonorably discharged19; and through federally qualified community health centers (FQHC). The IHS and VHA provide a full range of free or highly subsidized inpatient and outpatient healthcare services in designated facilities. The VHA has, for example, 170 medical centers and more than 1,000 outpatient sites.20 FQHCs, in contrast, are federally funded safety nets that offer mostly outpatient care on a sliding scale basis.

15  Centers for Medicare & Medicaid Services, NHE Fact Sheet Historical NHE 2017, https://www. cms.gov/research-­statistics-­data-­and-­systems/statistics-­trends-­and-­reports/nationalhealthexpenddata/ nhe-­fact-­sheet.html (last accessed July 30, 2019). 16  United States Census Bureau Quick facts, https://www.census.gov/quickfacts/fact/table/US/ PST045217 (last accessed July 25, 2018). 17  United States Census Bureau, Historical Income Tables: Households, Table H-­3 Mean Income Received by Each Fifth and Top 5 Percent, https://www.census.gov/data/tables/time-­series/demo/ income-­poverty/historical-­income-­households.html (last accessed July 30, 2019). 18  IHS Profile, Indian Health Services, https://www.ihs.gov/newsroom/factsheets/ihsprofile/ 19  About VHS, Veterans Health Administration, https://www.va.gov/health/aboutvha.asp. 20 Id.

The American Pathology of Inequitable Access to Medical Care   217 Public provision of care used to be more common in the United States. In the early 20th century, a majority of hospitals were operated by state or local governments. But public provision has receded due in part to a constellation of political interests.21 The very term “socialized medicine” has been co-­opted in the United States as the rallying cry to reduce or contain the government’s role in the provision of medical care even in areas where the government is already involved, such as the regulation of private insurance. Third and finally, in some cases, people can and still do pay for care out of pocket. They do so in a variety of circumstances, ranging from people with financial resources paying for elective or concierge care to people without, or with poor, insurance coverage patching together funding for care in any way possible.

2.2  Map of Public and Private Financing The US health insurance system is a convoluted map of different sources of public and private healthcare financing. Unlike in most other countries, the public system in the United States is neither universally available nor primary. The maze of coverage sources is confusing even for students of the US system, as evinced by the fact that it will take several pages just to describe its basics.

2.2.1  The Evolution of a Fragmented System Although all countries have a mix of public and private financing for healthcare, the United States is the only nation among its peers that lacks a universal public system to create access to basic care and, not surprisingly, the only one among peers that tolerates high rates of uninsurance.22 Most other peer nations finance care with universal social insurance through taxes (either earmarked or general revenue), as in France, Canada, or Australia. Alternately, some provide care directly through government-­owned hospitals and employed physicians, as in the UK’s socialized medical systems. The United States does neither of these things. Even the countries that on first blush might look more like the US system because of the centrality of the private market and options, such as Switzerland, are fundamentally different. After the passage of the ACA with its mandate that most people have health insurance and its creation of regulated marketplaces for purchase, many people attempted to analogize the US system to the Swiss one. The Swiss system is similar only to the extent that each person in Switzerland must purchase a health insurance policy directly from an insurer and each person has a choice among dozens of insurers.23 Beyond that, the two diverge. The Swiss government mandates 21 Jost, supra note 7, at 7. 22  Even in Switzerland, where people buy insurance from private insurance companies, the uninsurance rate is usually only 1–2%. If you google, “uninsurance rate Switzerland,” in fact, Google reports back on the unemployment rate. 23  Nikola Biller-­Andorno & Thomas Zeltner, Individual Responsibility and Community Solidarity— The Swiss Health Care System, 373 New Eng. J. Med. 2193, 2195 (2015).

218   Allison K. Hoffman and Mark A. Hall basic benefits for all plans, sets physician rates through a national fee schedules and maximum pricing for hospitals and pharmaceuticals, and approves the community-­ rated insurance premiums.24 The United States does regulate benefits in the exchange plans but does not regulate fees or approve premiums. Unlike in the United States, Swiss insurers are all not-­for-­profit organizations that must return surplus to policyholders.25 Coverage is highly subsidized and nearly universal, guaranteed by a strongly enforced mandate that, in a very un-­American way, garnishes wages up to the amount of premiums if someone does not have insurance. Access to care does not vary significantly, except with respect to what are considered “amenities,” like single-­room occupancy in hospitals, or services like dental care.26 The Swiss system is in effect a social insurance program, just with an element of choice built in. The Dutch and German systems, which also offer some choice of health plan, are also social insurance.27 The US system, even if it shares some surface characteristics, like choice of private insurance plans in a marketplace, is fundamentally different in critical philosophical and practical ways. At the time that other countries were developing social insurance systems, the United States faltered. Despite repeated attempts over the 20th century to create a national system for healthcare financing in the United States, none succeeded and different forms of healthcare financing developed incrementally and over time.28 Over the course of the 20th century, public insurance programs were created to pay for care for populations considered especially vulnerable and private insurance filled in from there. President Lyndon Johnson signed the bedrock programs of public financing, Medicare and Medicaid, into law in 1965 as part of his Great Society, and the programs have since been considered the greatest success of 20th century US health reform.29 Even so, Medicare and Medicaid were a failed attempt at universalism.30 They set out to pay for healthcare for the elderly, the “deserving poor”—focusing on those who are young or pregnant—and for people with disabilities. In other words, Medicare and Medicaid selected out more vulnerable populations for public insurance. This cabining out of more vulnerable populations is in direct contrast with European models based on solidarity, where the healthy and wealthy cross-­subsidize others in insurance pools. Medicare, which pays for medical care for 57 million older Americans and some people with disabilities, is itself a constellation of many parts and many rules. Medicare Part A 24 Id. 25  Id. at 2195. 26 2196. 27  Timothy Stoltzfus Jost, The Experience of Switzerland the Netherlands with Individual Health Insurance Mandates: A Model for the United States?, http://law2.wlu.edu/deptimages/Faculty/ Jost%20The%20Experience%20of%20Switzerland%20and%20the%20Netherlands.pdf (last accessed May 22, 2018). 28  See, e.g., Stuart Altman & David Schactman, Power, Politics, and Universal Health Care: The Inside Story of a Century-­Long Battle (2011); David Blumenthal & James A. Morone, The Heart of Power: Health and Politics in the Oval Office (2009); Paul Starr, Remedy and Reaction: The Peculiar American Struggle Over Health Care Reform (2011). 29  Social Security Amendments of 1965, Pub. L. No. 89–97, 79 Stat. 286 (1965) (codified as amended at 42 USC. §§ 1395, 1396 (2012)). 30  Theodore R. Marmor, The Politics of Medicare 95–96 (1970).

The American Pathology of Inequitable Access to Medical Care   219 pays for hospital benefits, and enrollment is free and automatic for all persons who ­contributed sufficient payroll taxes during their working years. Medicare Part B pays for outpatient care and is optional, although nearly all eligible people enroll. It is financed by general tax revenue and monthly premiums, which are income adjusted. Both of these parts have cost-­sharing requirements when people use care, which leave enrollees vulnerable to potentially high out-­of-­pocket spending for using medical care. For example, in 2019, Part A included a deductible of $1,364 per hospitalization.31 For hospital stays longer than 60 days, beneficiaries incurred cost sharing of $341 per day for days 61–90 and $682 per day beyond that.32 For outpatient care, beneficiaries owed a $185 deductible in 2019 and 20% coinsurance for most services.33 Ninety percent of all retirees obtain private supplemental insurance coverage, through various forms, to help fill in these gaps.34 Medicaid has become the largest single insurer in the United States through a series of incremental expansions.35 Medicaid is funded jointly by the state governments and through federal “matching funds,” which are at least as much as what the states spend and more in many states.36 It began as a program to cover poor people who were receiving cash welfare, including children and their parents and the elderly, blind, and disabled. Over time, additional categories of people were added, including pregnant women and children in families with slightly higher incomes.37 With the ACA, Medicaid eligibility was extended to all non-­elderly poor and the income threshold was lifted up to 138% of the federal poverty level (just over US$35,000 for a family of four in 2019). In a legal challenge to this provision, the Supreme Court found it unconstitutional, beyond Congress’s authority. The Court, in this case called NFIB v. Sebelius, held that the conditions of the expansion, which required states to expand eligibility to this entire population or else lose their current federal funding, were like a “gun to the head” of the states and thus a coercive use of federal spending power.38 The effect of the decision in this case was to make the expansion optional. States could expand and, if they did, get additional federal matching funds, or not and maintain their current programs. As of the end of January 2020, 36 states and Washington DC had expanded; 14 had not yet.39 Even with the decision in NFIB v. Sebelius, Medicaid has 31  Medicare.gov, 2019 Medicare Costs, https://www.medicare.gov/pubs/pdf/11579-­medicare-­costs. pdf (last visited July 30, 2019). 32 Id. 33 Id. 34  Juliette Cubanski et al., Henry J. Kaiser Family Found., Medicare Chartbook 60 (4th ed. 2010). 35  Cindy Mann & Deborah Bachrach, Medicaid as Health Insurer: Evolution and Implications, Commonwealth Fund to the Point (July 23, 2015). 36  Laura Snyder & Robin Rudowitz, Medicaid Financing: How Does it Work and What Are the Implications?, Kaiser Family Foundation Issue Brief (May 20, 2015). 37  Sara Rosenbaum & Benjamin D. Sommers, Rethinking Medicaid in the New Normal, 5 St. Louis J. Health L. & Pol’y 127, 138 (2011). 38  NFIB v. Sebelius, 567 US 519 (2012). 39  Kaiser Family Foundation, Status of State Action on the Medicaid Expansion Decision as of January 10, 2020, https://www.kff.org/health-­reform/state-­indicator/state-­activity-­around-­expanding-­ medicaid-­under-­the-­affordable-­care-­act/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Loc ation%22,%22sort%22:%22asc%22%7D (last accessed February 17, 2020).

220   Allison K. Hoffman and Mark A. Hall become an increasingly universal program for the poor, regardless of their welfare status, and the number-­one financing source for long-­term care. The creation of Medicare and Medicaid to cover limited populations catalyzed a half-­ century of efforts to fill in the gaps. The Children’s Health Insurance Program (CHIP) was created in 1997 for children just above Medicaid income thresholds, Medicare Part D was added in 2003 to provide prescription drug benefits to elders, and, most recently, the ACA expanded Medicaid and filled in coverage gaps in the Medicare Part D program. Now, just over one-­third of the population has public health insurance, mostly with Medicare and Medicaid and a small number with military or veteran’s coverage.40 Private financing—in the forms of employer-­sponsored health insurance (ESI), individual health insurance, out-­of-­pocket payments, and charity care—fills in remaining gaps. Health insurance developed as an employment benefit throughout the 20th century and has become deeply entrenched as such. While employers are not required to offer their employees health insurance, many do in large part because expenditures on health benefits are excludable from taxes for both the employer and employee. The beneficial tax treatment means that health benefits have become a less expensive way to compensate employees.41 More than half of Americans have private health insurance,42 most of whom have ESI.43 Without access to public insurance or ESI, people who want health insurance are left to seek it out on their own, directly from a private insurance company. Six to seven percent of the non-­elderly (about 15 million people) had health insurance through this “individual market” prior to the passage of the ACA.44 Historically, individual-­market coverage was relatively more expensive than ESI, in part because administrative costs were as high as 15–20% of total costs. This cost arose in part because, before the ACA, insurers could underwrite and exclude people based on their health, with few legal restrictions.45 As many as three in five people who applied for individual health insurance before the ACA could not afford the high premium prices or were denied coverage, and many people remained uninsured or underinsured.46 40  Jessica C. Barnett & Edward R. Berchick, Current Population Reports, Health Insurance Coverage in the United States: 2016 2 (2017), https://www.census.gov/content/dam/Census/library/ publications/2017/demo/p60-­260.pdf. 41  By one estimate, health insurance costs employers approximately $.65 per dollar of benefits. Jonathan Gruber, Covering the Uninsured in the United States, 46 J. Econ. Literature 571, 574 (2008). 42  Jessica C. Barnett & Edward R. Berchick, Current Population Reports, Health Insurance Coverage in the United States: 2016 2 (2017), https://www.census.gov/content/dam/Census/library/ publications/2017/demo/p60-­260.pdf. 43 Id. 44  Paul Fronstin, Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 2009 Current Population Survey, 334 EBRI Issue Brief (Sept. 2009), at 5. 45  See, e.g., Melinda Beeuwkes Buntin et al., The Role of the Individual Health Insurance Market and Prospects for Change, 23 Health Aff. Nov.–Dec. 2004 at 79, 81; Sara R. Collins et al., The Commonwealth Fund, Squeezed: Why Rising Exposure to Health Care Costs Threatens the Health and Financial Well-­Being of American Families 3–4 (2006). 46  Michelle M. Doty et al., Failure to Protect: Why the Individual Insurance Market Is Not a Viable Option for Most US Families, 1300 Commonwealth Fund Pub. 1–3 (2009).

The American Pathology of Inequitable Access to Medical Care   221

2.2.2  The ACA and Blurring of Public and Private Insurance The lines between public health insurance and private insurance have blurred over time.47 Public health insurance is increasingly outsourced to private insurers as administrators and is reliant on private-­market tools to attempt to control spending. As one example, programs like Medicaid have increasingly looked to private managed care companies to administer benefits since permitted by law in 1997.48 In 2016, more than three-­quarters of the 80 million beneficiaries were enrolled in managed care plans.49 Medicare reforms also have integrated private industry, in their case with the goal that competition would offer a means to control spending. In 1997, Congress created a new program in Medicare called Medicare+Choice (now Medicare Advantage), which allowed people to select to get their Medicare coverage from private managed care plans. These plans cover the traditional benefits under Parts A and B, and sometimes also Part D drug benefits. They also provide supplemental coverage to fill in the out-­of-­pocket costs not covered by traditional Medicare. The theory, in establishing these plans, was that competition among private insurers to gain Medicare beneficiary enrollment would drive up quality and down costs. This theory has faced many problems in practice. For many years these plans were excessively overpaid. Especially in the initial years, the private plans attempted to select out healthier populations to maximize profit. Even without these problems, enrollees have difficulty differentiating among plans and often make poor choices.50 As public insurance is shaped by private industry and market-­based strategies, private health insurance has been increasingly regulated, and, after the ACA, individual market coverage in particular has begun to take on qualities of social insurance. The ACA transformed the individual market, regulating what coverage insurers could offer, to whom, and at what cost. Prior to the ACA, apart from a few federal laws,51 most of the regulation of health insurance, especially in the individual market, was reserved for the states by a federal law called the McCarran-­Ferguson Act.52 The ACA in effect federalized much of the regulation of health insurance to make individual-­market coverage more affordable and 47  Amy Monahan, The Interaction Between Public and Private Health Insurance in The Oxford Handbook of US Health Law (I. Glenn Cohen, Allison K. Hoffman, and William M. Sage, eds. 2017). 48  Balanced Budget Act of 1997, Pub. L. No. 105–133. 49  Center for Medicare & Medicaid Services, Medicaid Managed Care Enrollment and Program Characteristics 11 (2016). 50  Allison K. Hoffman, Health Care's Market Bureaucracy, 66 UCLA L. Rev. 4 (2019). 51  One example is the Health Insurance Portability and Accountability Act of 1996, Pub. L. No. 104–191, 110 Stat. 1936 (1996). Other examples include the Employee Retirement Income Security Act of 1974, which set minimum standards for employee health plans, and the Genetic Information Nondiscrimination Act of 2008, which restricted the use of genetic information in health underwriting. Employee Retirement Income Security Act of 1974, Pub. L. No. 93–406, 88 Stat. 829 (1974); Genetic Information Nondiscrimination Act of 2008, Pub. L. No. 110–233, 122 Stat. 881 (2008). 52  McCarran-­Ferguson Act, ch. 20, 59 Stat. 33 (1945) (codified as amended at 15 USC. §§ 1011–1015 (2012)).

222   Allison K. Hoffman and Mark A. Hall meaningful. The ACA required insurers to take all applicants despite pre-­existing ­conditions, a requirement known as “guaranteed issue.”53 It eliminated underwriting and allows premium variation based on only four factors: age, geography, family size, and tobacco use status.54 Even if an insurer devised a way to cherry-­pick out healthier applicants, the law intended to disgorge any resulting profits through reinsurance and risk-­adjustment arrangements.55 Policies were required to include a range of mandated benefits similar to those offered in most employer plans, in 10 categories of “essential health benefits,” including preventive care without cost sharing, and were no longer allowed to impose annual or lifetime coverage limits on these mandated benefits.56 Finally, the ACA’s individual mandate required that most Americans carry health insurance that offers “minimum essential coverage” or else pay a penalty, which proved one of the most contentious parts of the law.57 Some people referred to the mandate as the linchpin of the individual market because it would resist adverse selection, where people would wait until they knew they needed medical care to buy coverage. The mandate intended to promote heterogeneous risk pools by requiring everyone to buy in, regardless of immediate medical need. Over time it appears less critical to the functioning of the ACA than initially imagined, as the mandate has fallen away with little impact on the individual market In addition, the ACA provided subsidies to help lower income individuals comply with this mandate. Anyone who earns from 100% to 400% of the federal poverty level in and does not have another adequate source of insurance, such as through an employer or Medicaid, is eligible for subsidized premiums. Lower earners (100–250% of the federal poverty level) are also eligible for cost-­sharing reductions to help them pay for the deductibles or other cost-­sharing when they use medical care.58 The ACA established exchanges for the sale of the newly regulated policies where, in theory, private insurers would compete to offer plans to new enrollees. Nearly 200 pages of the ACA are devoted to the endeavor of regulating private insurance and, in particular, revamping the individual market into a quasi-­public enterprise. The ACA left the majority of employer plans largely untouched, with several exceptions. The regulation that proved most contentious was one that all plans, including employer plans, cover contraceptive products and services as part of a general requirement to cover preventive care without cost sharing. The law exempted houses of worship from this requirement, and regulations created an accommodation for religious nonprofit organizations. Other employers, including Hobby Lobby—a closely held company, challenged this requirement as in violation of their religious freedom and prevailed in the Supreme Court.59 53  Patient Protection and Affordable Care Act § 1201, 42 USC. § 300gg (2012). 54 Id. 55  Id. at § 1341–1342, 42 USC. § 18061–62. 56  Id. at § 1302, 42 USC. § 18022; Id. at § 1001, 42 USC. § 300gg-­11 (2012), 42 USC. § 300gg-­13 (2012). 57  Id. at § 1501, 26 USC. § 5000A (2012). Some people are exempted from the penalty for reasons including religious objection or affordability, defined as when premiums cost more than 8% of household income. Id. 58  Id. at § 1401, 26 USC. § 36B (providing for “premium tax credits”); Id. at § 1402, 42 USC § 18071 (providing for “cost-­sharing reductions”). 59  Burwell v. Hobby Lobby, 573 US (2014).

The American Pathology of Inequitable Access to Medical Care   223 The ACA’s complex path to a simple goal of more universal coverage is unusual when viewed comparatively, but it offered an incremental way to pursue transformative policy goals. Optically, the ACA preserved preexisting private market structures and values. Practically, however, it bent and reshaped them to create more universal and meaningful access to healthcare in the United States. Yet, in large part because of the multifaced regulatory efforts needed to achieve their goals, the ACA policies to expand insurance coverage have been vulnerable.

2.2.3  ACA Challenges Even in the best-­case scenario, the ACA’s hoped-­for gains were hard to achieve. In addition to the legal challenges just discussed and numerous others, the ACA has faced a series of technical and political challenges in its efforts to expand access. For example, the initial roll-­out of the exchanges were an infamous disaster, with technical problems hampering enrollment. Even once the exchanges were running, people have struggled to choose health plans, often selecting plans in a way that seems contrary to their own interests. Others have struggled to afford plans on the exchanges, especially if they are receiving no or little premium subsidy support. Reinsurance and risk-­adjustment payments have been underfunded and underpaid.60 In part because of these problems, some states have faced low insurer participation and some states even had regions with no participating insurers. As discussed earlier, the Supreme Court decision in NFIB v. Sebelius dealt a blow to the Medicaid expansion. Even under the best conditions, the ACA’s piecemeal approach to expanding coverage was a laborious process. Even worse, when the Trump Administration succeeded the Obama Administration, it worked to undermine the ACA’s insurance expansion efforts, revealing simultaneously the durability of some of its policies and its greatest fault lines as well. Following President Donald Trump’s election in November 2016, the Republican majority in the US Congress pursued a wholesale repeal of the ACA. These efforts failed in part because of popular resistance, especially to the repeal of the Medicaid expansion, which was enabling more low-­income people to access medical are. The one area where Congress did succeed was to zero out the individual mandate penalty, as a part of tax reform, carving out one of the many interlocking pieces the ACA put in place to reform the individual market. Akin to the zeroing out of the individual mandate penalty by Congress, the Trump   Administration attempted a slow dismantling of the ACA reforms to the individual market through a series of technocratic rules that made it harder for insurers to participate in these markets and harder for people to afford coverage and  to access care even when they have it. For example, the Administration issued  several rules that allow for the sale of plans with less comprehensive coverage than the ACA required. Although the Secretary of the Department of Health and Human Services in the Trump Administration claimed that these plans would 60  Timothy Jost, Risk Corridor Claims by Insurers Far Exceed Contributions, Health Affairs Blog (Oct.1,2015),http://healthaffairs.org/blog/2015/10/01/implementing-­health-­reform-­risk-­corridor-­claims-­by-­ insurers-­far-­exceed-­contributions/.

224   Allison K. Hoffman and Mark A. Hall create more affordable coverage options,61 the result of these efforts would have been that many people have health insurance inadequate to pay for the care they need. These options for “bare-­bones” coverage also encourage adverse selection on the exchanges, where healthier people buy minimal coverage and sicker people stay in more comprehensive plans, whose premiums will become more expensive over time. Another round of administrative efforts focused on scaling back the Medicaid expansion and undermining Medicaid program enrollment altogether. The primary means for doing so was through the use of the Medicaid waiver program, known as Section 1115 waivers, which allow states to experiment with new benefits and program designs. In a letter to state Medicaid directors issued in January 2018, the Administration encouraged states to use Section 1115 waivers to experiment with work and community-­engagement requirements for beneficiaries. No previous administration would allow states to adopt these requirements, which are impossible for many of the vulnerable Medicaid population to meet and which create paperwork burdens and hurdles that result in disenrollment. In Arkansas, which started implementation in June 2018, 29% of the targeted population failed to meet requirements in the first month of implementation.62 The loss of coverage made these programs vulnerable to legal challenges that they were inconsistent with the objective of Medicaid. The Trump Administration also took aim at the contraceptive coverage requirement. In October 2017, the Administration issued new regulations that significantly expanded the employers who were exempted from the requirement. Newly exempted were all nonprofit and closely held organizations with objections based on religious or moral beliefs and all publicly traded for-­profit corporations with religiously based objections. This new rule was contentious both because of its substance, including its arguably discriminatory effects against women employees and their health needs, and because it was issued without an opportunity for public comment, as is required under the Administrative Procedure Act (APA). All of these actions incrementally scale back insurance coverage and could limit access to affordable medical care. All of these administrative actions have also been challenged in court, based on procedural and substantive grounds, and challengers largely prevailed in early decisions. These new rules have been vulnerable in large part because of the Administration’s short-­cutting procedural requirements under the APA. For example, initial rulings struck down the approval of work requirements in the states of Kentucky, Arkansas, and New Hampshire and the new contraceptive coverage rules in courts in California and Pennsylvania because the courts found they were promulgated in violation of the APA—without justification. It is impossible to know exactly where this struggle between the ACA’s efforts to expand insurance coverage and subsequent efforts to undermine it will land. What is clear is 61 Alex M. Azar, Obamacare Forgot About You. But Trump Didn’t, Wall St. J. (August 15, 2018). 62  Erin Brantley & Leighton Ku, A First Glance at Medicaid Work Requirements in Arkansas: More than One-­Quarter Did Not Meet Requirement, Health Affairs Blog (August 13, 2018), https://www. healthaffairs.org/do/10.1377/hblog20180812.221535/full/.

The American Pathology of Inequitable Access to Medical Care   225 that it illustrates the deep disagreement in the United States about who should be guaranteed access to what medical care. It is also clear that the gradual chipping away at the ACA since its passage risks further embedding the inequity of access among Americans.

2.3  Implications of Inequitable Coverage 2.3.1  Heterogeneity of Insurance and Access Even as the overall number of Americans with health insurance increases, the heterogeneity of how people get insurance preserves meaningful differences. Being insured does not guarantee the same access for everyone. A number of attributes make it easier or harder for people with insurance to access and to afford medical care, including what type of insurance they have, where they live, and how much disposable income they have to pay for cost-­sharing. It would be impossible to touch on all of the ways access differs among insured people in this chapter, but here we highlight a few critical aspects. One way that the many sources of coverage detailed earlier can differ is in their “provider networks.” Insurers typically contract with a limited number of physicians, hospitals, and other providers to obtain discounted rates for their customers. When patients receive care from “in-­network” providers, they pay much less in out-­of-­pocket costs than if they receive care from “out-­of-­network” providers. As a practical matter, then, an individual’s choice of provider can be restricted by the insurer’s network. In addition, as discussed in Section 3, providers and hospitals have significant discretion over which patients they accept for medical care, especially for non-­emergency care. Some providers choose to accept only patients with private insurance, and an increasing number of providers, especially those in high-­demand practice areas and cities with more demand than supply, do not accept insurance at all or charge retainer fees for a patient to join a practice. One leading national hospital, the Mayo Clinic, recently announced that it will prioritize privately insured patients over those on Medicare or Medicaid.63 Others do so whether they make that policy explicit or not. Providers and facilities that accept public insurance are more likely to accept Medicare than Medicaid, since Medicare’s reimbursement rates are higher. Rates of provider participation are highest in employer plans, followed by ACA marketplace plans, and lowest in Medicaid.64 Among participating providers, availability of appointments was again highest for ESI and lowest for Medicaid. Variability in which benefits plans cover likewise can be significant and is defined in ways that allow more unpredictability and less transparency than how benefits are defined in Europe. For example, employer plans have discretion for the most part to cover—or not cover—whatever benefits they would like. Although most large employers 63  Casey Ross & Andrew Joseph, Mayo Clinic: Privately Insured Patients to Get Priority over Medicaid, Medicare Patients, Stat (March 15, 2017). 64  Daniel Polsky et al., Scope of Primary Care Physicians’ Participation in the Health Insurance Marketplaces, 37 Health Affairs 1252 (2018).

226   Allison K. Hoffman and Mark A. Hall cover a wide range of services, employers could choose to offer a very limited set of benefits. Furthermore, even though the ACA set a baseline by requiring all individual-­ market plans to cover the 10 categories of essential health benefits, it gave states considerable discretion in deciding what exactly would fall into these categories. Furthermore, it did not require employer plans to cover these benefits. Even though most do, they are not legally required to do so. Plans also vary on what kind of legal recourse is available when medical claims are denied. Although ESI is superior to other coverage in many ways, beneficiaries have limited legal recourse and remedies when medical claims are denied by health plans. These plans, and the remedies for violations of plan terms, are governed by a federal law called the Employee Retirement Income Security Act of 1974 (ERISA). The Supreme Court has interpreted this law to shield health insurers from liability under state law for denials of coverage under employer plans.65 If members of an employer plan have their claims denied, their only recourse is to pay for the medical care denied and seek reimbursement through federal courts under the terms of ERISA’s remedy provisions. If they are physically harmed by the plan’s denial of care, there is generally no legal remedy for this harm. For non-­employer plans, such as Medicaid plans or ACA exchange plans, beneficiaries still have access to state courts for coverage denials and a fuller range of remedies. Finally, someone’s resources—financial and intellectual—can have a significant impact on their access to medical care. Can they afford to pay the cost-­sharing under their plans? Can they navigate the complexity of the system and their insurance plans? Can they take time off work to get to doctors and other appointments, especially if their plans have a more restricted network that requires them to travel farther for care? The multifaceted approach to healthcare financing in the United States not only creates inequitable results, it also undermines solidarity, confuses patients, and creates tremendous overhead costs and burdens for providers. It obscures inequities in the system, perhaps intentionally. This structure has been built through incremental attempts to move toward more universal coverage, but the end result still leaves significant gaps, both in who is covered and in how meaningfully the coverage provides access to care.

2.4  The Future of Health Insurance in the United States This fragmented system is a result of various factors that will continue to hamper efforts at creating a more uniform structure for financing care in the United States. That said, the shortcomings of the ACA and its belabored efforts to build out private insurance, followed by efforts to dismantle ACA, could create opportunities. When the ACA health insurance exchanges failed to produce high enrollement politicians, policy experts, and pundits began to discuss more seriously solutions that would expand public health insurance and move the United States financing system closer to its European counterparts. 65  Aetna v. Davila, 542 US 200 (2004).

The American Pathology of Inequitable Access to Medical Care   227 These ideas build on the existing Medicaid and Medicare programs, extending public health insurance to more people. One popular idea is a Medicaid buy-­in to replace the ACA marketplaces. Since most states have well-­established Medicaid programs and some have had trouble ensuring sufficient options in their exchanges, Medicaid buy-­in could ensure access to more affordable options for exchange enrollees. Other proposals build on the existing Medicare program, either allowing buy-­in by younger populations or, more transformatively, creating a Medicare-­for-­All program, which many progressive candidates running for office, including US Senator Bernie Sanders, have taken up as their mantle. These types of expansion of public insurance will continue to meet significant resistance by insurers and hospitals whose livelihood could be threatened by them. And they are certainly not a silver bullet. These programs have their own problems that would become even more detrimental if they expand. And efforts to finance them through new tax revenue will inevitably be highly unpopular. Yet it is possible that the light that the ACA has shone on problems in healthcare access in the United States and the difficulty in remedying them through private market reforms that require constant technocratic tinkering may shift the political balance.

3  The Treatment Relationship Of course, more than the financing structure affects access to healthcare. Both medical practices and legal requirements can expand or contract access to care. This section discusses the legal obligations that have developed through common and statutory law that provide access to care under certain conditions. It focuses in particular on the doctor– patient relationship, specifically, the legal rules that govern the structure of the treatment relationship: the duty to treat and the formation, modification, and termination of the relationship.66

3.1  Constitutional Rights of Access Unlike in many other countries, in the United States there is no constitutional obligation for government to fund or provide healthcare. The Supreme Court has firmly stated that “the Constitution imposes no obligation on the States to pay . . . any of the medical expenses of indigents.”67 This is because the Due Process Clause traditionally has been seen as protecting individuals from improper government interference (“negative”

66  This portion of the chapter is based on materials previously published as Mark A. Hall, et al., Health Care Law and Ethics in a Nutshell (3rd ed. 2011). 67  Maher v. Roe, 432 US 464, 469 (1977).

228   Allison K. Hoffman and Mark A. Hall liberties), rather than generating entitlements to state-­conferred assistance or benefits (“positive” liberties).68 Nevertheless, violations of negative liberties can be argued in many contexts of healthcare regulation. One court found a generalized constitutional right to be free of poorly justified state restrictions on medical decision-­making.69 The court struck down a law that allowed only licensed physicians to practice acupuncture as an infringement of patients’ right to “obtain or reject medical treatment,” which the court found was encompassed by the right to privacy identified in Roe v. Wade. Most courts, however, require only a “rational” justification for such restrictions,70 and, in any case, they view the state’s concern to protect health as a “compelling” interest,71 so these types of argument rarely succeed. For instance, courts have upheld state bans on alternative cancer therapies that are probably harmless but thought to be ineffective.72 And the US Food and Drug Administration (FDA) may restrict patients’ access to unapproved drugs even if they are a dying patient’s only hope.73 Even though there is no general constitutional right to healthcare, “when a State does decide to alleviate some of the hardships of poverty by providing medical care, the manner in which it dispenses benefits is subject to constitutional limitations” imposed by the Due Process and Equal Protection Clauses of the Constitution.74 Thus, a publicly funded hospital cannot arbitrarily discriminate in the patients it treats or in the services it provides.75

3.2  Common Law and Statutory Duties to Treat Patients Although US healthcare law is a largely “constitution-­free zone,” common law and statutory law have a great deal to say about a patient’s rights to be treated. A patient’s right to receive treatment differs between paying and indigent patients. Paying patients, of course, have greater access. But certain obligations to treat are independent of the ability to pay, and for paying patients there remain some barriers based on race, disability, and other factors. Thus, the ability to pay is not the primary issue in the following discussion. Nevertheless, access to medical care by the indigent is the subject of both explicit and implied attention in much of what follows. 68  See also Deshaney v. Winnebago County Department of Social Services, 489 US 1989 (echoing the foregoing). 69  Andrews v. Ballard, 498 F.Supp. 1038 (S.D. Tex. 1980). 70  That is, as long as there is a conceivable justification for the policy, it is permitted. 71  Even for fundamental rights, the state can override the right in order to promote a very important public interest. 72  Cf. Rutherford v. United States, 442 US 544 (1979). 73  Abigail Alliance for Better Access to Developmental Drugs v. von Eschenbach, 495 F.3d 695 (D.C. Cir. 2008) (en banc). 74  Maher, supra, 432 US at 470. 75  See Memorial Hospital v. Maricopa County, 415 US 250 (1974) (unconstitutional to refuse county health services to temporary residents).

The American Pathology of Inequitable Access to Medical Care   229

3.2.1 Doctors 3.2.1.1  The “No-­D uty” Rule A doctor is generally under no duty to accept patients, regardless of the seriousness of their condition, their ability to pay, or the physician’s basis for refusing. In the seminal decision, Hurley v. Eddingfield, which is still regarded as stating “good” (i.e., prevailing) law, the court affirmed the dismissal of a suit for damages on behalf of a would-­be patient who died when a physician refused to treat her—notwithstanding that the doctor had been her family physician in the past, was available to render care (and aware that other physicians were not), was told she was now seriously ill and relying on an expectation of treatment, gave no reason for the refusal, and was offered payment.76 The court reasoned that a physician’s traditional freedom to select patients remained unaltered by the advent of state licensure law, which imposed no obligation on a physician “to practice at all or on other terms than he may choose to accept.” This “no-­duty” rule tracks the historical absence, in American tort law, of any legal obligation to aid strangers in distress. Professional medical ethics reflect a similar policy: “Even the Hippocratic Oath, by which every doctor is morally bound, assumes a pre‑existing relationship of patient and physician, which relationship in its inception is basically contractual and wholly voluntary,”77 and the American Medical Association (AMA) Principles of Medical Ethics leave a physician free “to choose whom to serve” (though they recognize an exception for emergencies). Implicit in these articulations of the “no duty” rule is the axiom that, where a physician–patient (or hospital/patient) relationship does exist, there is a legal obligation to treat. This duty to treat is fiduciary in nature and persists until the relationship is properly terminated (sec. 3.5). Since the formation of the treatment relationship is foundational to the entire range of issues that make up law and medicine (including malpractice and most of bioethics), what constitutes “formation” is important. 3.2.1.2  Formation of the Treatment Relationship The court in Hurley v. Eddingfield, supra, absolved Dr. Eddingfield despite his having been the deceased’s “family doctor” and the fact that the medical condition was active labor. This reflects the general rule that an established custom of past treatment does not oblige a doctor to treat a patient’s future illnesses; doctor–patient relationships are specific to a “spell of illness” and must be established, or renewed, accordingly. Within a given “spell,” however, the law often requires very slight involvement before finding that a treatment relationship between patient and doctor (or hospital) has been formed. A patient’s description of symptoms over the phone followed by a physician’s brief instructions, a telephone call to a physician’s office for the purpose of initiating treatment, or scheduling an appointment to treat a particular medical problem have all sufficed to support a fact-­finder’s inference that a doctor or hospital had undertaken to 76  Hurley v. Eddingfield, 156 Ind. 416, 59 N.E. 1058 (1901). 77  Agnew v. Parks, 343 P.2d 118, 123 (Cal.App.1959).

230   Allison K. Hoffman and Mark A. Hall provide care. While little is generally required, the decisions are not uniform: courts have also found that no relationship arose where the call to a physician’s office to schedule an appointment did not itself seek or generate medical advice and that, where a patient interpreted the physician’s response to her telephone contact as a refusal to undertake care, the requisite “consensual” characteristic of the relationship was missing—irrespective of the objective content of their communication. Physicians’ informal “curbside” consultations with colleagues normally will not establish a relationship between the patient and the consultee-­physician. Courts fear that implying a treatment relationship out of limited, routine consultative contacts (of which the patient, incidentally, is often unaware) would chill a useful medical practice to the detriment of patients and physicians alike. Of course, more formal physician referrals likely will result in legal recognition of the treatment relationship. Finally, no treatment relationship customarily arises where physicians examine patients for the benefit of third parties. Thus, physicians conducting physical exams for insurance eligibility or for employment-­related purposes generally are not held liable to the examinee for failure to treat or for other medical errors or nondisclosures. Exceptions have arisen, however. In the employment context, courts have implied a limited relationship, imposing a duty that extends only to disclosure of any test results that “pose an imminent danger to the examinee’s physical or mental well-­being” or have implied the relationship where the physician affirmatively undertook treatment or gave advice. At least one case held that an employer itself (in contrast to the examining physician) may be liable to the examinee for negligent failure to disclose a serious medical problem discovered in a pre-­employment exam.78 In contrast, another case found that a life insurer has no duty to disclose positive HIV test results to a policy applicant.79 However, the court, seemingly unaware of the “no duty” rule usually applied in such situations, suggested in dictum that if a physician (rather than the company) had been “directly involved,” the court might find a duty to disclose such information, based on patients’ expectations, professional ethics, and physician expertise in health matters. Of course, to the extent that these policy rationales are persuasive, they undercut the no-­ duty rule that normally applies in these cases.

3.2.2 Hospitals It is sometimes stated that the “no duty” rule just documented for physicians applies with equal force to hospitals. While this might have been true at one time (the older case law seems to say so, but it is not without ambiguity), this general “no duty” rule unquestionably is not the law now. Hospitals and other healthcare institutions, in contrast with physicians, operate under numerous sources of law (both statutory and court created) that prohibit the arbitrary refusal to admit patients. As noted earlier, uninsured patients traditionally have relied on free care rendered by public or private hospitals. Most larger cities maintain a municipal hospital that is 78  Dornak v. Lafayette General Hospital, 399 So.2d 168 (La. 1981). 79  Deramus v. Jackson Nat. Life Ins. Co., 92 F.3d 274 (5th Cir. 1996).

The American Pathology of Inequitable Access to Medical Care   231 obliged to treat all patients regardless of the ability to pay, and many smaller localities historically provided for the uninsured by compensating private hospitals for treating the poor. This local largesse has become overtaxed, however, and so many municipalities have greatly scaled back or eliminated their support for medically indigent residents. Private hospitals have a long tradition of caring for the poor, but their capacity for charity care has been stretched thin by reimbursement constraints that eliminate the ample revenues previously received from insured patients. As a consequence of these various social forces, some private hospitals turned desperately ill patients away from their emergency roemergencyoms, usually by transferring them to public municipal facilities. This practice of “patient dumping” led both to litigation over a private hospital’s obligation to render emergency care to indigent patients and to a federal law addressing the practice (Section 3.2.2.2). Because of this history and the persistence of access disparities, the following two subsections, while exploring hospital treatment duties generally, have particular importance for access to care by the indigent. 3.2.2.1  The General Duty to Provide Care Wilmington General Hospital v. Manlove80 is the seminal decision that finds in the common law a duty on the part of hospitals to act reasonably in their patient selection decisions. Manlove involved a hospital emergency room that refused to treat a severely ill infant because he was under the care of another physician who was not a member of the hospital’s medical staff. As a result, the infant died. By analogy to the tort of negligent termination of gratuitous services, the court reasoned that in cases of “unmistakable emergency,” a hospital that maintains an emergency room which by “established custom” has been open can properly be held responsible for refusing to treat a patient whose condition “worsens” as a consequence of time lost pursuing the unforthcoming treatment. Detrimental reliance is thus at the core of the case. Accordingly, Manlove applies only to emergency care, and even then its scope is rather limited (Section 3.2.2.2, infra). A potentially more powerful and sweeping common law theory—one that would cover all forms of hospital treatment—asserts that private hospitals owe duties to the public at large on the ground that they are “quasi-­public” by virtue of the importance of their services, the funding they receive from public sources, their licensure, and their tendency to enjoy monopoly status in a community. The Manlove court rejected this view, however, and it has not in fact been widely adopted, although a few courts have been receptive to it.81 3.2.2.2  Access to Emergency Care Notwithstanding the absence of a general duty to rescue, in certain areas US law has been slowly (perhaps even ambivalently) but perceptibly responsive to the moral 80  174 A.2d 135 (Del.1961). 81  See, e.g., Leach v. Drummond Med. Group, 192 Cal. Rptr. 650 (Cal.App.1983) (reasoning applied to the only physician group practice in town).

232   Allison K. Hoffman and Mark A. Hall c­ hallenge of taking action to reduce acute, visible, and avoidable suffering. One of the areas in which this trend may be at work is in the legal recognition of a hospital duty to care for emergency patients irrespective of their ability to pay. Common Law and Statutory Rights. Manlove was the first case to fashion a theory of relief for patients denied hospital emergency care. Its principal impact has been in securing access to emergency care by uninsured patients. In addition, many states have laws expressly requiring hospitals to treat emergency patients without regard to their ability to pay. Federal law imposes the same duty on hospitals that maintain charitable tax exemption. These established legal protections are limited, though, by their narrow definitions of what constitutes an emergency and of the extent of treatment required in an emergency. For instance, the Manlove theory, applied in a number of states, addresses refusals to treat only in cases of “unmistakable” emergency, only where the patient’s condition worsens due to the delay in finding an alternative source of care, and only where the delay is caused by reliance on an ER’s open-­door custom. Many of the state statutes define an emergency as a situation requiring immediate treatment in order to prevent loss of life or limb—which can exclude a broad range of serious, albeit less extreme, medical conditions. Two Arizona cases point the way toward a more expansive duty to provide emergency care. Eschewing Manlove’s reliance-­based approach, the Arizona Supreme Court has implied a sweeping duty “to provide emergency care to all persons presenting themselves for such aid” (emphasis in original), ostensibly based upon the state regulatory requirement that all general hospitals maintain emergency facilities as a condition of licensure.82 A decade later the court relied on private hospital accreditation standards that prohibit discrimination based on the “source of payment,” incorporated by reference into the state’s hospital licensing statute, to conclude that hospitals may never transfer emergency patients on economic grounds.83 These two Arizona cases are important for two reasons. First, they are best understood as based in common law public policy—essentially, an emergency room application of the “quasi-­public status” theory discussed earlier—rather than on idiosyncrasies of state regulatory law and are thus of general rather than parochial interest. Second, they allow courts to redefine the nature of an emergency and the extent of the treatment required. Hospitals are obliged to treat any patient with a “need for immediate attention” and to provide such patients all care that is “medically indicated.” The duty to treat thus encompasses far more than care necessary to prevent the patient’s condition from deteriorating: “The relevant inquir[y] . . . d[oes] not relate to ‘stabilization’ and ‘transferability,’ but rather to the nature and duration of the emergency.”84 This broadened theory has not been adopted by other states, perhaps due to the subsequent federal enactment of and

82  Guerrero v. Copper Queen Hospital, 537 P.2d 1329 (Ariz.1975). 83  Thompson v. Sun City Community Hospital, 688 P.2d 605 (Ariz.1984). 84  688 P.2d at 611.

The American Pathology of Inequitable Access to Medical Care   233 widespread reliance on the Emergency Medical Treatment and Active Labor Act (EMTALA)—notwithstanding EMTALA’s own arguable limitations in this connection. The Emergency Medical Treatment and Active Labor Act. The Emergency Medical Treatment and Active Labor Act of 1985 (“EMTALA”), 42 USC. 1395dd, has become the single most important legal tool governing access to emergency care, due principally to its uniform national applicability and its remedies. Hospitals that receive Medicare payment must comply with EMTALA’s terms for all their patients. EMTALA creates a private right of action for damages for violation of its terms by such hospitals, though there is no comparable action against physicians. It also authorizes civil money penalties of up to $50,000 for negligent noncompliance by both hospitals and physicians. EMTALA was enacted in the belief that state law was too weak to prevent the widespread “dumping” of indigent and uninsured patients. Its protections, however, go further: they are triggered by the refusal to properly examine or treat “any individual” who comes to a hospital emergency department seeking care, irrespective of whether the patient can pay for care or the hospital’s reasons for refusing care. EMTALA requires, first, that the hospital provide for an “appropriate medical screening examination within the capability of the hospital’s emergency department” to determine whether there is a medical emergency. If so, treatment must then be provided to the point of “stabilization.” Specific analogous provisions also apply to women in labor. Screening. A moment’s reflection on the statutory language just quoted suggests that a diagnostic screening might be “[in]appropriate” in varying ways and for different reasons. Uncertainty has thus arisen over just what hospital conduct the statutory phrase reaches, as well as what standard of performance it imposes. Initially, there was debate over whether EMTALA requires or assumes an improper motive for refusing treatment based on factors such as economics, demographics, or personal characteristics. This has been criticized on the grounds that there is no statutory support for such a reading, that it is so inclusive as to be virtually without limit (and therefore meaning), and that it is nonetheless sufficiently difficult to prove that it would defeat virtually all EMTALA claims. In Roberts v. Galen of Virginia,85 the Supreme Court put this issue to rest, holding that EMTALA’s stabilization requirement (and probably also its screening requirement) applies regardless of motive. Instead of a subjective motivation test, many courts have applied a more objective test that looks to whether the patient was harmed by “disparate” or “non-­uniform” treatment—some purposeful variation from the medical practices that the hospital would otherwise apply to similarly situated persons. This approach focuses solely on whether a hospital complied with its own standard procedures and not whether those practices are themselves reasonable. For instance, a hospital that failed to do an expensive diagnostic test on a patient with chest pains or a headache would not be liable under EMTALA for an undetected heart attack or brain aneurysm simply because other hospitals would have done the test; instead, a challenger would need to show that this particular hospital normally handled cases like this differently. Courts have adopted this “disparate treatment” 85  525 US 249 (1999).

234   Allison K. Hoffman and Mark A. Hall analysis largely in an effort to avoid making EMTALA a federal malpractice law that would displace ordinary state-­law negligence claims, a result which they believe Congress did not intend.86 Treatment and Stabilization. If the required screening reveals an “emergency medical condition,” the hospital must undertake treatment. EMTALA’s definition of “emergency” is a condition reasonably likely, without “immediate” treatment, to create “serious jeopardy” to the person’s health. This definition is at least as inclusive as many of those found in state common law and state statutes, and the range of conditions it covers is thus reasonably broad. The more problematic issue is how far treatment must proceed under the statutory mandate to “stabilize” the condition. Stabilization is defined as a level of treatment likely to prevent “material deterioration” of the condition during transfer. (Transfers of unstabilized patients are permitted in limited, specified circumstances involving a written request or expected medical benefit). The facts of a pre-­EMTALA case are instructive.87 An auto accident victim came to the emergency room of a private hospital with “multiple deep facial lacerations, a possible head injury, traumatic damage to the teeth and multiple bruises and contusions of the body, resulting in considerable loss of blood.” The hospital merely bandaged him, took X-­rays, monitored for shock, and administered IV fluids to stabilize his blood pressure before transferring him to a Veteran’s Administration hospital for further treatment. This course of action is probably entirely consistent with EMTALA: thus, in many situations, EMTALA actually may not require more extensive care than was due under state common and statutory law. Ironically, the expansive treatment obligations in a few common law precedents (see the two Arizona cases, supra) may actually exceed the “stabilization” requirement of EMTALA. Given EMTALA’s dominance of the field, however, their further application seems unlikely, even though EMTALA does not actually preempt such common law claims. “Preventive” Dumping. EMTALA requires screening and stabilization of anyone who “comes to” an emergency department. Under this language courts have rejected EMTALA claims by patients who do not, literally, show up at the hospital. By regulation, the Secretary of Health and Human Services has confirmed that “comes to” requires physical presence on hospital property; however, a hospital’s own ambulance is deemed hospital property, and, arguably, non-­hospital ambulances cannot be rerouted except for reasons of lack of hospital capacity or staff. EMTALA has not been without controversy. Some have criticized it for responding to a problem that was never as widespread as claimed or that no longer exists, and for imposing an awkward and poorly drafted solution. Others believe that EMTALA has been effective; still others that patient dumping persists at unacceptable levels, notwithstanding EMTALA. Under any view, however, EMTALA has become central to the law of access to emergency medical care, and there is little reason to believe its role will decline. 86  Vickers v. Nash. General Hosp., 78 F.3d 139, 142–143 (4th Cir. 1996) (reviewing cases). 87  Joyner v. Alton Ochsner Medical Foundation, 230 So.2d 913 (La. App. 1970).

The American Pathology of Inequitable Access to Medical Care   235

3.2.3  Doctors Within Health Care Organizations How can hospitals, dependent on doctors to deliver care, comply with their institutional duty to treat (under the various legal theories explored above) if Hurley leaves physicians free to refuse patients? One solution is regulatory: since EMTALA was enacted, physicians are no longer completely free to refuse emergency patients with impunity because they may face civil fines for negligent noncompliance with EMTALA’s terms. A second solution (pre-­EMTALA) is contractual. Hospitals may require as a condition of medical staff membership that physicians assist in treating emergency and indigent patients. If physicians accept this condition by joining the medical staff or working in the emergency room, then this contractual obligation may extend to the patient as a third-­party beneficiary.88 A similar solution applies to managed care, where a health plan may contractually bind participating physicians to see individuals it has a contractual duty to treat. In Hand v. Tavera,89 the court relied on the applicable contracts (characteristic of health plans) to find a treatment relationship with the plan’s on-­call physician. The physician refused to authorize the patient’s admission to the plan’s hospital based on the symptoms and history conveyed by a telephone consult from the ER. As a consequence, the patient had a stroke at home. The court reasoned that the enrollee paid premiums to the plan to purchase medical care in advance of need; the plan arranged to meet its obligation to provide care by paying physicians; those physicians, in return, agreed to treat the plan’s members. The identity of the physician who happened to be on call for emergency admissions was immaterial: the plan brought the patient and physician together “just as surely as though they had met directly and entered the physician-­patient relationship.”90

3.3  Wrongful Denials: Antidiscrimination Law and Refusal to Treat As discussed thus far, physicians (and, to a considerably lesser extent, hospitals) enjoy substantial legal discretion to refuse patients for “good” reasons, “bad” reasons, or no stated reason at all. In a limited number of areas, which are the subject of this section, federal law specifically disapproves certain bases for treatment refusals. In addition, states often have counterpart regulatory laws, generally applicable to “public accommodations” (which covers hospitals but often not medical offices). Private accreditation standards also prohibit discriminatory practices by hospitals on the basis of race and other characteristics, including source of payment, and the federal charitable tax exemption for hospitals carries with it certain obligations to provide care on a nondiscriminatory basis to paying patients.

88  Hiser v. Randolph, 617 P.2d 774 (Ariz.1980). 90  864 S.W.2d at 679.

89  864 S.W.2d 678 (Tex. App. 1993).

236   Allison K. Hoffman and Mark A. Hall

3.3.1  Title VI: Race, Ethnicity, and Gender. Title VI of the federal civil rights law, enacted in 1964, prohibits any “program or activity receiving federal financial assistance” from discriminating against, excluding, or denying benefits to individuals on the grounds of race, color, or national origin.91 Thus, overt discrimination by hospitals participating in the federal Medicare or Medicaid programs (which were enacted in 1965) or receiving other financial support is barred. Despite Title VI, subtler forms of racial (as well as gender) discrimination by healthcare providers doubtless persist. For instance, Title VI has been invoked in a few cases challenging decisions to relocate or to close hospitals serving predominantly minority populations. Until 2010, however, there was no clear prohibition of discrimination that applied to physicians. That is because, prior to the ACA, courts and regulators interpreted Title VI as not applying to physicians at all. (Even though physicians receive federal funds through Medicare and Medicaid, courts and regulators interpreted those programs in a strained way that avoided triggering Title VI.) The ACA changed this legal landscape by broadening the definition of federally funded activities to include funding of physician services and also by extending the prohibition to discrimination based on gender.

3.3.2  Disability Discrimination Two closely related laws, both of which apply to a wide range of activities beyond healthcare, have become important in this field: Section 504 of the Rehabilitation Act of 1973 (29 USC. 794), and the Americans with Disabilities Act of 1990 (42 USC 12101). HIV/ AIDS first drew attention to the application of disability law to healthcare, and HIV cases continue to arise and receive coverage, but the application of disability discrimination law to healthcare is considerably broader. The most obvious difference between the two laws is the reach of their regulation. Section 504 applies to federally funded “programs and activities” (which includes hospitals that receive Medicare reimbursement, but probably not doctors). The ADA, by contrast, reaches various entities irrespective of whether they receive federal financial assistance, including state and local governments (Title II) and public accommodations (Title III). Of greatest importance here, the latter include the “professional office of a healthcare provider, hospital, or other service establishment,” so doctors’ offices are covered. While there is more case law developed under sec. 504 because it has been in effect for so much longer, the greater reach of the ADA suggests that it will ultimately supersede section 504 in importance. 3.3.2.1  Protected Class The first question in most disability discrimination cases is whether the individual falls within the protected class. Section 504 protects a “handicapped individual,” defined as someone with a “physical or mental impairment which substantially limits one or more of such person’s major life activities” or someone with either a “record of,” or who is 91  42 USC. 2000d.

The American Pathology of Inequitable Access to Medical Care   237 “regarded as having,” such an impairment. The ADA’s definition of “disability,” except for the choice of the operative word, is almost verbatim.92 These terms are quite broad, reflecting Congress’ intent to protect people against discrimination arising not only from prejudice but also from fear and myth.93 Echoing those policies, the Supreme Court held that non-­symptomatic HIV infection constitutes a disability under the ADA. In Bragdon v. Abbott,94 an HIV-­positive patient alleged that her dentist violated the ADA when he refused to fill a cavity for her in his office (offering, instead, to do so at a hospital, though there was no evidence that the hospital would be safer or even that he had privileges to practice there). The Court held that non-­ symptomatic HIV infection constitutes (in the statutory terms) a “physical impairment” from the moment of infection onward, and that, by interfering with the plaintiff ’s reproductive capacity, the infection “affected a major life activity” because of the centrality to life of reproduction and sexual relations. The open-­ended nature of this “major life activity” category is suggested by the Court’s intimation that other plaintiffs might persuasively assert that HIV impacts other life activities as well. Finally, the Court concluded that HIV infection was a “substantial limit” on the plaintiff ’s reproductive activity, noting that her status would impose significant risks of infection on male sexual partners (20–25%) and on any child during gestation and childbirth (8–25%). The Court emphasized that this third requirement is met “even if the difficulties [generated by the disability for the life activity in question] are not insurmountable.” 3.3.2.2  Core Provisions Finding a handicap or disability is only the first inquiry. Section 504 prohibits regulated programs or activities from excluding, denying benefits to, or discriminating against any “otherwise qualified handicapped individual . . . solely by reason of his handicap” (29 USC. 794). “Otherwise qualified” means able to meet program requirements “in spite of ” the handicap, as established through an individualized, factually specific inquiry. So, for example, a quadriplegic would not be otherwise qualified to work on construction sites but would be otherwise qualified to work at desk jobs. Even where people cannot initially meet all program or activity requirements, they may nonetheless be “otherwise qualified” if the sponsor of the program or activity can make “reasonable accommodation” (i.e., take steps, short of incurring “undue financial and administrative burdens” or making “a fundamental alteration in the nature of the program” (Arline, supra), that would enable the person to meet the requirements), in which case the sponsor must do so. The ADA rules, though not identical, are similar. A key difficulty in disability law is determining whether a person is “otherwise qualified” for the benefit or service. Classical applications of this standard arise in cases involving access to education and employment, in which the analysis has two salient characteristics: (1) it impliedly assumes that the benefit or service is generally available to a qualified class of people, under eligibility requirements established by its sponsor; 92  42 USC. 12102(2). 94  524 US 624 (1998).

93  School Bd. of Nassau County v. Arline, 480 US 273, 279, 284 (1987).

238   Allison K. Hoffman and Mark A. Hall and (2) because the disability is not the reason for which the person seeks the benefit or service, it is coherent to ask whether the person can meet the eligibility terms notwithstanding (or “in spite of ”) the disability, with any needed reasonable accommodation. In healthcare, this analysis applies logically enough where an individual is seeking access to care for a problem that is unrelated to his disability, as where a physician who is treating a patient for an ear infection refuses to perform medically indicated surgery after learning that the patient is HIV positive. The benefit (ear surgery) is generally available on certain terms to those who need it, and, since the patient is seeking care for a condition other than the disability (HIV status), it is coherent to ask whether, with reasonable accommodation (here, relating to the patient’s infectiousness and immunocompromise), the patient qualifies for the benefit “in spite of ” that disability. But in cases in which it is precisely the disability that gives rise to the need for access to healthcare, this is not a very coherent approach to “otherwise qualified.” A major case in this regard arose when a hospital wanted to deny artificial ventilation to an infant with anencephaly whose respiratory distress was a result of the anencephaly. With some struggle, a few courts have found a meaningful approach in these cases. They preclude a healthcare provider from using disability alone as the basis for withholding medical benefits. A person is “otherwise qualified” for a particular medical benefit if “there is no factor apart from the mere existence of disability that renders the participant unqualified for the benefit.”95 To withhold care, the provider would need to have a “bona fide medical reason” for the denial. This would allow a physician to refuse care based on the claim that the physicians felt underqualified to treat the patient’s particular condition. Or, if the patient’s condition would preclude a successful outcome from treatment, as with a kidney transplant in a patient with lung cancer, care could be denied on account of the condition.

3.4  Modifying the Terms of the Treatment Relationship Although the formation of the treatment relationship is essentially contractual in nature, once established, tort and fiduciary law generally govern the parties’ conduct and their mutual obligations through doctrines such as medical malpractice and informed consent. This section explores the extent to which patients and physicians are free to modify those rules and define the terms of their relationship in accordance with their own preferences. Courts have generally refused to enforce agreements with patients by which healthcare providers try to waive their liability for negligence. The leading case is Tunkl v. Regents of the University of California,96 in which the California Supreme Court concluded

95  Woolfolk v. Duncan, 872 F. Supp. 1381, 1389–1390 (E.D. Pa. 1995) (seeking care for, and alleging discrimination based on, HIV status). 96  383 P.2d 441 (Cal. 1963).

The American Pathology of Inequitable Access to Medical Care   239 that a hospital’s exculpatory agreement with a patient, signed at admission, bore all the indicia of an unconscionable adhesion contract. Tunkl and similar cases do not, by their terms, preclude liability waivers that fall short of full exculpation, and in fact courts are likely to enforce releases from liability where the care provided departs from standard medical practice for good reason, as when a patient leaves the hospital early against medical advice or refuses recommended medical treatment. In this connection, consider Shorter v. Drury,97 holding that a document signed by a Jehovah’s Witness surgical patient releasing providers from responsibility for consequences “due to my refusal to permit the use of blood” constituted an enforceable assumption of the risk of an otherwise-­avoidable death by the patient, rather than an unenforceable exculpation of negligence for the physician. Courts have also responded favorably to agreements changing the forum or mech­an­ ism of dispute resolution. For example, health maintenance organizations (HMOs) may require their members to arbitrate rather than litigate medical negligence claims, at least where this is agreed to through bargaining by a powerful representative (a large employer) and there is a choice of alternative plans.98 On the other hand, “point of treatment” arbitration agreements, presented for signature upon hospital admission or at the doctor’s office, are less likely to receive judicial approval because of concerns about their fairness, though these decisions, too, are not uniform. Agreements to alter the prevailing standard of care, rather than to waive it entirely, as in Tunkl, are more difficult. If notice is adequate and there is some choice, should HMOs (for example) be allowed to contractually bind enrollees to accept a lower-­than-­normal standard of care (e.g., anything above “gross negligence”) by plan providers as a cost-­ containment mechanism that would benefit enrollees by reducing premiums? The law on such questions is not well-­developed but is likely to be a focal point of controversy as medical standards are increasingly subsumed within managed care contractual arrangements.99

3.5  Terminating the Treatment Relationship Professional duties arise on the formation of a treatment relationship and continue until it is properly terminated. Patient “abandonment” is the term applied to an improper termination of treatment that is intentional, in contrast with termination that is due to a mistake in medical judgment. The latter is a matter for ordinary malpractice law, but this distinction is frequently confused.

97  695 P.2d 116 (Wash. 1985). 98  Madden v. Kaiser Foundation Hospital, 552 P.2d 1178 (Cal. 1976). 99  For competing arguments, see Clark Havighurst, Health Care Choices: Private Contracts as Instruments of Health Reform (1996) (endorsing such a contractarian approach); Maxwell Mehlman, Fiduciary Contracting: Limitations on Bargaining Between Patients and Health Care Providers, 51 U. Pitt. L. Rev. 365 (1990) (criticizing it).

240   Allison K. Hoffman and Mark A. Hall Where a treatment relationship exists, the law of abandonment requires that the physician (or hospital) provide all necessary care unless the relationship is terminated (1) by the patient or (2) by the provider after giving the patient proper notice and an opportunity to secure an alternate source of care. Abandonment law is thus much more forgiving than is usually recognized. The only explicit restraint on a doctor’s (or hospital’s) freedom to abandon a patient is the procedural one of notice. As classically conceived, there is no real substantive content to abandonment law because the law does not scrutinize the reasons for abandonment: so far as abandonment law is concerned, doctors may, with proper notice, stop treatment because they want to retire, or go on vacation, or simply because they dislike the patient. To ensure that there is no question about the adequacy of notice to the patient and opportunity to obtain substitute care, prudent physicians as a practice usually take af­fi rm­a­tive steps themselves to arrange for substitute care. Vacationing doctors will usually have associates cover their cases, and retiring doctors will ordinarily tell patients that a designated physician has agreed to take their cases. Prudent hospitals, likewise, will never simply discharge an ill patient, even after ample notice; instead, they will locate an alternative facility to which a patient can be transferred. These pragmatic accommodations have created a degree of uncertainty in abandonment law. Because this body of law is based on an implied contractual undertaking and on notions of fiduciary responsibility, its precise limits are not firmly set. Consequently, the case law in different states offers conflicting indications of whether simple notice of treatment termination is sufficient or whether instead the law requires healthcare providers to arrange for a substitute source of care. This point becomes a critical issue in the modern context where doctors and hospitals face increasingly severe constraints in healthcare reimbursement. A provider might seek to terminate care because the patient’s insurance runs out or won’t cover the treatment. If patients are given “notice,” and perhaps appeal rights, will this suffice under the usual procedural requirements of abandonment doctrine—or might a creative and sympathetic plaintiff persuade a court to read a substantive element into the doctrine and prohibit the termination of treatment based on inability to pay? Three cases shed conflicting but ambiguous light on the legality of “economic abandonment.” In Ricks v. Budge,100 the court allowed the plaintiff to maintain an action alleging the following facts: After ordering Mr. Ricks to the hospital for a seriously infected hand, Dr. Budge refused treatment and walked out because Mr. Ricks would not immediately catch up on his past due accounts. This decision is frequently cited by commentators for the proposition that it is illegal to abandon a patient who cannot pay. However, these facts do not support a general prohibition of economic abandonment. Instead, the holding is perfectly consistent with purely “procedural” abandonment law, which only requires the doctor “to give the patient sufficient notice so the patient can procure other medical attention if he desires,” (Id.) and prohibits patient abandonment only at a critical stage in the course of treatment. 100  64 P.2d 208 (Utah 1937).

The American Pathology of Inequitable Access to Medical Care   241 Additional support for this view is suggested in a leading abandonment decision, Payton v. Weaver.101 There, the court allowed a physician to stop treating an uncooperative patient despite the unavailability of any substitute care. This case concerned a renal dialysis patient who “frequently appear[ed] for treatment late or at unscheduled times in a drugged or alcoholic condition, used profane and vulgar language, and on occasion engaged in disruptive behavior, such as . . . cursing staff members with obscenities.” Although Dr. Weaver’s attempts to find alternative treatment centers were unsuccessful, the court held that he “gave sufficient notice of [his intent to cease treatment] and discharged all his obligations.” It is difficult to determine, though, whether Payton v. Weaver invokes a purely procedural abandonment rule because its compelling facts might also provide substantive justification for the decision to discontinue treatment. Finally, in Muse v. Charter Hosp. Winston-­Salem,102 the court ruled that a psychiatric hospital illegally interfered with the physician’s medical judgment when it encouraged the discharge of an adolescent patient whose insurance ran out, which led to the patient’s suicide three weeks later. Although the opinion never mentions the abandonment doctrine, it is based on the hospital’s alleged “policy or practice” of discharging patients when their insurance runs out. Even so, the decision is subject to the same uncertainty as Ricks and Payton: we don’t know whether the hospital is liable because the discharge policy is per se wrong or because the procedure it followed failed to sufficiently notify the parents of their son’s fragile condition so they would be sure to find alternative care. Even with such notice, however, a patient who is discharged for financial reasons might be unable to find a suitable alternative source of care, thus making it difficult for the initial provider to terminate care in a legally safe manner.

4 Conclusion What is evident from the preceding discussion of access to healthcare in the United States is that there is no singular, coherent set of guidelines or policies that determine who has access to what and when. The lack of constitutional rights and of any centralized system of financing or regulating access has created great legal and policy complexity. Although the basic moral questions that inform policy and legal development are the same in the United States as in Europe, the United States deals with these questions on a case-­by-­case basis and in opaque ways that often simply avoid answering them altogether. The result is that the most defining characteristic of access to medical care in the United States is its inequity.

101  182 Cal. Rptr. 225 (1982). 102  452 S.E.2d 589 (N.C. App. 1995), aff ’d mem., 464 S.E.2d 44 (N.C. 1995).

chapter 12

Access to He a lthca r e i n Eu rope André den Exter and Keith Syrett

1 Introduction The manner in which healthcare systems in Europe are organised stems from specific historical, political, and cultural traditions, which are reflected in distinct arrangements in each European state, with only a very limited ‘pan-­European’ approach. Traditionally, we differentiate between Bismarck and Beveridge-­like models; that is, social health insurance (SHI) versus national health systems (NHS). While both purchase and provide necessary healthcare services for those in need,1 a key difference (as noted in the next section) is the way these types of system are funded, either by (income-­related) premiums or through taxation. To be sure, this typology of healthcare systems is, in reality, too simplistic given the heterogeneity of SHI systems and NHS-­like models (United Kingdom, Italy, Spain, etc.)2 and the fact that it overlooks other (hybrid) systems as in Scandinavia.3 Nevertheless, it will be adopted here to keep the analysis to a manageable length. Describing the main features of European healthcare systems remains therefore an effort to identify key characteristics of these systems in terms of organisation, financing, and delivery of health services, and the main actors. Structured by questioning what the system covers, who are eligible to receive healthcare. when patients receive healthcare, and the physician’s duty to provide care, this chapter highlights the applicable regulatory framework and describes some trends. 1  For more detail on this classification see, e.g., H E Sigerist, ‘From Bismarck to Beveridge: Developments and Trends in Social Security Legislation’ (1999) 4 J Public Health Pol 474–496. 2  These three systems will be used to illustrate the key features of the NHS model in this chapter. 3  A more systematic review of countries’ healthcare systems can be found at the European Observatory on Health Systems and Policies (‘Health Systems in Transition, HiT’), www.euro.who.int

244   André den Exter and Keith Syrett Differences in the organisation and financing of healthcare systems make a meaningful comparison difficult. Nonetheless, despite the patchwork of systems there are some common concerns, such as the struggle in defining the healthcare basket, problems of rationing services in order to cut costs, and the need for more transparent decision-­ making, as well as the impact of human rights law on guaranteeing access to high-­quality care for all. The two models also share a consensus on abstract common values like equal access for all, solidarity, effectiveness and efficiency in healthcare, good-­quality care, and access based on medical need, as will be outlined in the next section.

2  Key Characteristics of European Health Systems 2.1  Social Health Insurance Systems In Europe, SHI systems provide for coverage of a large part—or the entire—population. These—nearly—universal SHI schemes are generally part of a broader national social security system and highly regulated by law (organisational structure, scope and nature of healthcare entitlements covered, [private contractual] relationship between insurer and providers, regulated prices or tariffs, cost-­sharing measures, etc.). Depending on the legal system, the health insurance entities (sickness funds, mutualities, state funds, etc.) may have limited or extensive self-­regulatory powers to compete on quality and/or prices.4 In addition, voluntary health insurance (VHI) has been introduced, covering health services excluded from or partially covered by the social health insurance scheme (e.g., amenity care, access to private hospitals, dental care, physiotherapy, user charges [known as ‘cost-­sharing’ in the United States], etc.). The VHI market (both supplementary and complementary) gives policyholders greater choice of provider and level of amenity, and it contributes to financial protection where there are gaps in the scope and depth of social insurance.5 Despite the heterogeneity of SHI systems in Europe, they share a wide range of services covered, such as what is generally considered as medical necessary and effective, although the exact scope of healthcare services differs by country, largely depending on the available (financial) resources.6

4  E.g., the Dutch SHI scheme is generally considered as one of the most ‘competitive’ systems introducing the concept of managed competition derived from Enthoven’s managed care model: A Enthoven, ‘The History and Principles of Managed Competition’ (1993) 12(Suppl 1) Health Affairs 24–48, whereas other health insurance schemes have less or limited self-­regulatory powers and are more usually considered as part of the executive branch (e.g., French mutualities), see K Chevreul, et al, ‘France: Health System Review’ (2015) 17(3) Health Syst Transition (HiT), 29. 5  A Sagan, S Thomson, Voluntary Health Insurance in Europe: Role and Regulation (WHO 2016) 29. 6  Countries like Romania and Bulgaria have a particularly large coverage gap, see Organisation for Economic Cooperation and Development (OECD), Health Statistics 2016.

Access to Healthcare in Europe   245 Since the 1990s, several Central and Eastern European countries have introduced similar social health insurance systems based on the Bismarckian approach, some of them even experimenting with elements of regulated competition on quality (Czech Republic, Bulgaria, Poland, etc.).7 Based on European core values such as nondiscrimination, collective responsibility, and risk and income solidarity (i.e., low- and high-­income groups, individuals with families, elderly and young, healthy and sick persons), SHI systems are constructed to redistribute financial risks of ill-­health in order to guarantee universal healthcare access. These underlying values and characteristics make SHI more than a simple insurance based on actuarial principles, but rather a ‘way of life’.8 In contrast with NHS and other tax-­based systems, SHI systems largely depend upon members’ contributions or premiums to fund healthcare services. These premiums are traditionally income-­related (risk-­independent) and sometimes up to a designated ceiling.9 The financial flows within the SHI system are those among the population, purchasers, and providers. Sometimes sickness funds are responsible for collecting contributions (as in Austria), or they can be a state-­run fund (Estonia) using these revenues to fund health services for their members by contracting with both individual and institutional health providers. Traditionally, institutional providers (hospitals) are public or private not-­for-­profit entities, whereas ambulatory health providers may work on a for-­profit basis.10 Since the 1990s, SHI systems have been confronted with increased expenditures in healthcare. Countries vary significantly in their methods for controlling healthcare costs, including delisting health services from the benefit packages, increasing premiums, introducing co-­payments and deductibles for health services paid by consumers, the use of cost-­effectiveness criteria, and different contract types and payment mechanisms. Several systems have even introduced market elements, triggering competition among providers and/or purchasers and aiming at more efficiency and cost-­containment measures (Switzerland, the Netherlands, Germany).11 The effects of what is called ‘regulated competition’ on equal access remain a matter of heated debate among various scholars.12 For instance, Den Exter and Guy argue that the Dutch experience reveals serious concerns about whether a system of regulated competition and emerging private health arrangements respects the basic human right of equal access to healthcare services. From a human rights perspective, combining competition and private initiatives in

7  A den Exter, Health Care Law-­Making in Central and Eastern Europe: Review of a Legal-­ Theoretical Model (Intersentia 2002); R Saltman, R Busse, and J Figueras (eds.), Social Health Insurance Systems in Western Europe (Open University Press 2004). 8  Saltman, Busse, and Figueras, 5. 9  ibid., 6. 10  Over the past 10 years, there has been a noticeable growth in the capacity of the for-­profit hospital sector: OECD Health Data 2016, quoted by P Jeurisson, A Duran, and R Saltman, ‘Uncomfortable Realities: The Challenge of Creating Real Change in Europe’s Consolidating Hospital Sector’ (2016) 16(Suppl 2) BMC Health Services Res 168. 11  Based on the concept of ‘managed competition’; see Enthoven (n. 4). 12  E.g., M Sandel, What Money Can’t Buy: The Moral Limits of Markets (Allen Lane 2012).

246   André den Exter and Keith Syrett healthcare markets with restrictive measures inspired by social values (e.g., solidarity and equity) appears to be an extremely difficult exercise.13 Individual choice is another key element of SHI. Consumers can usually choose from all ambulatory providers and hospitals. This choice may also include treatment options. In several SHI systems there is even choice of insurer and health plan. The general idea is that consumer choice (or ‘exit’) will trigger competition among providers and/or insurers to deliver the best quality healthcare for a reasonable price. But often free choice of provider is limited by the administrative requirement of referral to visit a medical specialist (especially where geographical accessibility is problematic). In addition, selective contracting of providers has been introduced in some countries to contain hospital costs at the price of free choice (e.g., the Netherlands). In the end, there is little evidence of choice and competition acting as drivers of performance in areas including access and equity.14 Finally, contemporary SHI systems use risk pooling mechanisms among competing funds to reallocate financial resources to the individual funds. This will compensate sickness funds with relatively large numbers of ‘high-­risk’ insured people. These ‘levelling systems’ use different criteria (e.g., age, gender, disability, region), but risk equalisation remains imperfect, meaning that the intended risk solidarity for certain categories of insured parties will diminish.15

2.2  National Health Systems NHSs are financed primarily from public funds. The UK National Health Service, often regarded as the archetype of this type of system, is financed primarily from general taxation revenues and additionally from ‘national insurance’ (a mandatory tax on earnings and self-­employed profits which contributes to a fund from which certain state benefits are paid). Only a very small percentage of its funding—just over 1% in 2015–2016—is accounted for by user charges, such as those for prescriptions, dental treatment, and other patient fees; some revenue is also generated by land sales and parking charges. Presently, the United Kingdom spends a little less than 9.4% of its gross domestic product (GDP) on health.16 Italy’s Servizio Sanitario National (SSN) is primarily funded through national and regional taxes, with some co-­payments for pharmaceuticals and outpatient care. Some 8.9% of GDP is spent on healthcare.17 Likewise, the vast majority of the funding of Spain’s Sistema Nacional de Salud (SNS) is accounted for by general taxation, with some 13  A den Exter and M Guy, ‘Market Competition in Health Care Markets in the Netherlands: Some Lessons for England?’ (2014) 22 Med L Rev 255–273. 14  S Gregory, A Dixon, and C Ham (eds.), Health Policy Under the Coalition Government: A Mid-­Term Assessment (King’s Fund 2012), quoted by den Exter and Guy, 267. 15  R van Kleef, et al., ‘Improving Risk Equalization with Constrained Regression’ (2016) 18 Eur J Health Econ 1137–1156. 16 OECD, Health at a Glance 2017, section 7.3. 17 Id.

Access to Healthcare in Europe   247 co-­payments for outpatient pharmaceutical prescriptions and orthopaedic and prosthetic equipment. Spain also possesses two other statutory schemes: mutual funds catering to civil servants, the armed forces, and the judiciary, which are funded by a mixture of employee contributions and taxation, and the mutualities focused on assistance for accidents and occupational diseases, which are funded through employer and employee contributions. In 2016, Spain spent 9% of its GDP on healthcare.18 While these are national health systems, there are significant local and regional dimensions to the organisation and delivery of healthcare in each instance. In England,19 the budget for the NHS is administered by the NHS Commissioning Board (known as ‘NHS England’), which retains some one-­third of the total funds for the services for which it arranges provision, such as primary care and services for rare and very rare conditions, and distributes the remainder to those bodies—currently, clinical commissioning groups (CCGs)—which arrange provision for the local populations that they serve. The SSN in Italy is regionally based. The national government sets general objectives and establishes fundamental principles, determines the ‘basket’ (i.e., the health services included), and allocates funds. The 19 regions and 2 autonomous provinces undertake planning, authorise and accredit providers, monitor the quality of their care, and allocate resources to locally based Aziende Sanitari Locali, which deliver care either directly or through public or private providers. As for Spain, the national government has strategic planning and monitoring responsibilities, with purchasing and provision of care carried out at regional level by the country’s 17 Autonomous Communities. Notably, however, in response to the global financial crisis and the intervention of the European Union, a degree of recentralisation was brought about by law RDL 16/2012, which limited regional autonomy on expenditure and determination of benefit packages. Each of these systems is founded on particular values which are enshrined in law. The Constitution of the Republic of Italy ‘safeguards health as a fundamental right of the individual and as a collective interest’,20 and the founding law of the SSN proclaims public responsibility for health protection, universal and equitable access to healthcare, global coverage of the care needs of all citizens, public financing of healthcare through general taxation, and portability of rights across the country and reciprocal care between regions.21 In Spain, the Constitution of 1978 recognises the right to health protection,22 and the General Health Law (Ley General de Sanidad) establishes the fundamental characteristics of the SNS as being universality, comprehensiveness, and financing through 18 Id. 19  The situation in the United Kingdom is complicated by different forms of organisation of healthcare in the four constituent nations and distinct legal systems in Scotland and Northern Ireland from that which applies to England and Wales. The focus of discussion in this chapter will be on England, as (by far) the largest of the four nations. For a discussion of the differences in health systems, see G Bevan et al., The Four Health Systems of the United Kingdom: How Do They Compare? (Health Foundation/Nuffield Trust, 2014). 20  Article 32. 21  Law No. 833/1978. 22  Article 43.

248   André den Exter and Keith Syrett public resources (with provision for contributions and fees for the provision of certain services), as well as requiring adequate organisation to provide comprehensive healthcare, coordination, and, where necessary, integration of all public health resources into a single system.23 The position in England is somewhat different, as might be expected of a state that does not possess a codified constitution. The values upon which the NHS was founded in 1948—comprehensiveness, universal availability, and access based on clinical need rather than ability to pay—were reflected in the governing legislation of the Service.24 but in a manner which did not serve particularly to elevate them above more mundane provisions relating to the minutiae of the day-­to-­day functioning of the NHS. This changed in 2009, when an NHS Constitution for England was first published, the quasi-­ legal document being designed to preserve the founding principles of the Service from political interference and to clarify the rights and responsibilities of staff and patients. It should be noted, however, that the ‘rights’ protected by the Constitution are mainly declaratory of the existing legal or administrative position and that there is no direct legal obligation placed upon NHS bodies to uphold the rights; rather, they are merely required to ‘have regard’ to them when performing health service functions.25 Accordingly these ‘rights’ are best viewed as considerations pertinent to the exercise of administrative discretion by bodies making decisions on matters of access within the NHS and are thus consonant with the traditional position adopted in English law: that realisation of such access comes about through organisational arrangements within the NHS rather than as a matter of human rights. We return to this issue in Section 3.2.2.

3  What Is Covered? The Struggle to Define the Basket 3.1  Lists of Entitlements and Coverage Decision-­Making 3.1.1  Social Health Insurance Systems The key characteristics of SHI systems have largely been set by statute and derived from constitutional rights such as the right to healthcare, right to life, the social state principle, or a combination of these rights. SHI systems therefore reflect and realise the state obligation to guarantee access to healthcare facilities for the entire population in terms of health insurance entitlements, set by law and organised according to national traditions: public (administrative) law (Germany, France, Austria, etc.) or civil law (Netherlands)— based systems. 23  Law 14/1986, Article 46. 24  Initially the National Health Service Act 1946, now the National Health Service Act 2006 (as substantially amended by the Health and Social Care Act 2012). 25  Health Act 2009, s 2(1).

Access to Healthcare in Europe   249 SHI systems define a statutory catalogue of health services to which the insured is entitled: the benefit package. Some countries opt for a rather detailed list of entitlements (e.g., including various preventive and primary care services; hospital services like obstetric and neonatal care services in the Czech Republic, Romania),26 whereas others identify more general categories of care (ambulatory and hospital care, as in the Netherlands) elaborated by health professionals (guidelines) and/or in insurance policies or health plans. The range of services differs by country depending on the criteria applied for defining the services covered and on the (financial) resources available. Most countries use ‘medical necessity’, ‘effectiveness’, and ‘cost-­effectiveness’ to define health benefits, although these are not always specified in detail by law.27 What is considered medically necessary changes over time, depending on developments and innovations in medical science. Scientific research and medical practice may question the clinical effectiveness of particular interventions or technologies, whereas cost effectiveness studies may urge the exclusion or inclusion of existing and new therapies on economic grounds. Periodic review of the current necessity and appropriateness of services listed is therefore crucial to guarantee access to good-­quality care. Apart from these criteria, SHI countries differ in listing either excluded or covered services, or a combination of both, in defining the benefit catalogue. Within these systems, coverage decision-­making has been institutionalised by federal or national bodies (e.g., the Federal Joint Committee in Germany, the High Authority for Health in France, or the Health Care Institute in the Netherlands), with a wide range of regulatory powers. These decisions, ‘listing or delisting’ services from the benefit catalogue, are based on evaluation of evidence-­based reports by the Institute for Quality and Efficiency in Health Care (IQWiG, Germany), and similar authoritative bodies (Health Care Institute, Netherlands, etc.). So far, these evaluation studies have focused primarily on the cost effectiveness of new medicines. Initiatives at European level, such as establishing a European Union (EU)-wide network on health technology assessment (HTA) and the European Commission’s proposal for an EU Regulation on HTA, if adopted, will help to improve the evaluation process while increasing transparency in the appraisal decision-­making process.28 But an explicit rationing mechanism or cost-­effectiveness threshold is absent in most health insurance systems. Although recommended, a ceiling value has not yet been accepted by the Dutch government.29 Setting a (flexible) limit at policy level remains extremely difficult. Instead, health insurance systems are experimenting with more 26  By using so-­called framework contracts set by the Ministry of Health, listing the statutory entitlements as well as the terms of contracting providers: see C Vlãdescu, ‘Romania: Health System Review’ (2016) 18(4) HiT 29–30. 27  B Gibis et al., ‘Shifting Criteria for Benefit Decisions in Social Health Insurance Systems’ in Saltman, Busse, and Figueras (eds.) (n. 7), 189–190. 28  See the HTA Core Model of EUnetHTA (www.EUnetHTA.eu) and the Proposal for a Regulation on health technology assessment and amending Directive 2011/24/EU, 31 January 2018, COM(2018)51 final. 29  It was suggested to apply a bandwidth with a median value of €40,000/per added life-­year (QALY), CPB Document no. 152, 10 (in Dutch), www.zorginstituutnederland.nl.

250   André den Exter and Keith Syrett ­ exible price reimbursement mechanisms for new and expensive medicines, including fl ‘conditional admission’. In the Netherlands, for instance, high financial risk medicines will be put ‘on hold’ and therefore excluded from coverage by health insurance until there is positive advice from the advisory body, the Health Care Institute.30 This Institute reviews the clinical and cost effectiveness of the new medicine. Thus, in cases of proven clinical and cost effectiveness and positive advice, the Minister of Health may decide to grant reimbursement. Candidates for conditional admission include different types of oncology treatments, with expected overall costs per indication exceeding €40m, or more than €50,000 per treatment. On several occasions, the Minister has triggered the conditional admission procedure. Depending on the assessment’s outcome, the Ministry applies a case-­by-­case approach to conclude a ‘reasonable price’ arrangement with the pharmaceutical company for each medicine individually. In total, approximately 25 financial arrangements have been made. Throughout 2015–2016, this policy resulted in a substantial cost reduction of medicines of €160m.31 Still, such agreements have been criticised due to the price nondisclosure clause as stipulated by the pharmaceutical companies.32 Parliament will only be informed about the overall cost savings, and confidentiality of the financial details to third parties remains a cornerstone of these agreements. Other countries (Belgium, France, Germany, Czech Republic, etc.) have also concluded so-­called managed entry agreements for high-­price medicines and other technologies.33 Certainly, these managed entry agreements have increased patients’ access to new medicines that would not be reimbursed otherwise, but at the cost of transparency on the financial conditions.

3.1.2  National Health Systems The technique of defining the ‘basket’ of entitlements through lists also characterises NHSs. Hence, in Italy, although only from 2001 onward (except in relation to pharmaceuticals, in respect of which a National Pharmaceutical Formulary has operated since the institution of the SSN),34 SSN benefits have been categorised via two lists determined on the basis of medical necessity, effectiveness, human dignity, appropriateness, and 30  Parliamentary Decision Stb 131, 2018, 23 April 2018. It mostly concerns so-­called off-­label medicines that have received a market authorization for a particular (cancer) treatment, but which appear to be effective for other indications, though not yet evidence-­based. 31  Parliamentary proceedings 2017–2018, 29477 nr. 442, Pharmaceutical policy, Letter from the Minister of Health to Parliament (in Dutch), 25 September 2017. 32  E.g., the Dutch Health Care Institute has called on the pharmaceutical company Vertex to reveal the price to allow public scrutiny (www.zorginstituutnederland.nl, orkambi). 33  See, e.g., T Morel et al., ‘Reconciling Uncertainty of Costs and Outcomes with the Need for Access to Orphan Medicinal Products: A Comparative Study of Managed Entry Agreements Across Seven European Countries’ (2013) 8(198) Orphanet J Rare Dis 1–15; K Pauwels et al., ‘Managed Entry Agreements for Oncology Drugs: Lessons from the European Experience to Inform the Future’ (2017) 8(171) Front Pharmacol 1–15; A Ferrario et al., ‘The Implementation of Managed Entry Agreements in Central and Eastern Europe: Findings and Implications’ (2017) 35 Pharmacoeconomics 1271–1285. 34  See G France, F Taroni, and A Donatini, ‘The Italian Health-­Care System’ (2005) 14 Health Econ S187–S202, S189.

Access to Healthcare in Europe   251 efficiency in delivery. Regions can provide treatments and services beyond those defined on the positive list, but must use their own finances to do so. The negative list comprises services of proven clinical ineffectiveness, those considered beyond the scope of the health system (such as ritual circumcision), certain services which may be judged appropriate but only on a case-­by-­case basis (e.g., orthodontic services), and services, such as cataract treatment, for which hospital admission (as distinct from, say, day surgery) is likely to be inappropriate. Whereas the scope of coverage of the Italian health system is founded upon law, definition of the basket of treatments and services which are provided by the NHS in England is a matter which rests primarily with health service managers, albeit that, in respect of new technologies (especially pharmaceuticals), an important role is played by the process of health technology assessment, as outlined subsequently. Although Parliament has drawn up two lists—the ‘black’ and ‘grey’ lists—which respectively specify treatments which may never be prescribed by general practitioners (such as cranberry juice, soaps, shampoos, and toothpastes!) and those which may only be prescribed in certain circumstances for reasons connected to clinical safety or cost effectiveness, there is no positive list of the treatments and services which should be made available, and there are frequent media reports of restrictions on access being imposed by NHS England, CCGs, or their predecessor institutions. Importantly, such a position is not unlawful provided always that it does not amount to a ‘blanket ban’ on access which is incapable of being displaced by proof of exceptional circumstances on the part of the patient.35 A rather different approach to listing has recently been adopted in Spain. RDL 16/2012 regulated the package of benefits, not by explicitly excluding particular treatments and services (some, such as psychoanalysis, hypnosis, and spa treatments, were already proscribed),36 but rather by listing those for which co-­payments are required. Core services, which comprise prevention, diagnosis, treatment and rehabilitation services, and emergency medical transportation, remain free. However, co-­payments apply to the supplementary package, which covers pharmaceutical prescriptions and orthoprosthetic devices, and to accessory services, such as occupational therapy, which ‘have been vaguely described as all activities, services or techniques, without character of benefit, that are not considered essential and/or are used as aid-­devices for chronic care improvement’.37 These common packages, which apply nationwide, can be supplemented at a regional level, with the Autonomous Communities providing additional services, treatments, and procedures from their own resources.

35 See R v North West Lancashire Health Authority, ex parte A, D and G [1999] EWCA Civ 2022. 36  See Royal Decree for Services Provision 63/1995. 37  E Bernal-­Delgado et al., ‘Spain: Health System Review’ (2018) 20(2) HiT 56.

252   André den Exter and Keith Syrett As in SHI systems, processes and institutions of health technology assessment also play an important role in determining the scope of coverage of national health systems.38 The United Kingdom has been a global pioneer in this regard. What is now called the National Institute for Health and Care Excellence (NICE) was established in 1999, with a remit to address the perceived inequity of ‘postcode prescribing’, whereby treatments and services might be available in one locality but not in an adjacent one. This was to be done through two main mechanisms: production of clinical guidelines which set out ‘best practice’ in the management of a particular condition, and appraisals of new health technologies (primarily, though not exclusively, pharmaceuticals) with a view to recommending (or not) their use on the NHS. It is the latter which is most pertinent to the definition of the basket of treatments in the NHS in England: such appraisals carry a mandatory funding requirement,39 meaning that if NICE recommends use of a treatment or service on the NHS, funds must be allocated to enable access to it (generally, within 3 months); although there remains no obligation upon the treating clinician to use the technology in question if it is not considered appropriate for the patient. This is further underlined by a provision in the NHS Constitution which ­confers a ‘right to drugs and treatments that have been recommended by NICE for use in the NHS’.40 NICE therefore effectively functions as a gate-keeper to access to many new treatments and services on the NHS. This necessarily makes its decisions controversial, especially given its utilisation of the Quality Adjusted Life Year (QALY) methodology—which has never formally been validated by the legislature—which frequently attracts adverse media commentary as being tantamount to ‘putting a price on life’.41 There is greater explicitness than in the Dutch case as to the threshold below which new technologies will be deemed cost effective for use on the NHS, this being £20,000–30,000 per QALY in the majority of cases, but with a much more generous threshold (starting at £100,000 per QALY) for drugs for rare diseases.42 However, comparably to the approach on conditional reimbursement taken in the Netherlands, access can be provided to technologies which exceed the threshold through patient access agreements concluded between the pharmaceutical company and NHS England. Special arrangements for managed 38  However, as in SHI systems, there is variability in HTA institutions and approaches between systems. For example, unlike the UK, neither Italy nor Spain possess a singular national body whose responsibility is exclusively to make HTA recommendations, and while HTA must be performed in relation to decision-­making on pharmaceuticals in Italy, it is only advisory (in respect of all technologies) in Spain. For discussion, see J Chamova, Mapping of HTA National Organisations, Programmes and Processes in EU and Norway (European Commission 2017). 39  National Institute for Health and Care Excellence (Constitution and Functions) Regulations and the Health and Social Care Information Centre (Functions) Regulations 2013, SI 2013/259, regs 7(6) and 8(6). 40  Department of Health, NHS Constitution for England (2015), section 3a. 41  See, e.g., BBC Television, The Price of Life (29 June 2009); C Cook. ‘How Much Is a Year of Life Worth?’, www.bbc.co.uk/news/health-­28983924; J Rigby, Why the NHS Thinks a Healthy Year of Life Is Worth £20,000’, www.channel4.com/news/drugs-­life-­breast-­cancer-­nice-­20-­000-­a-­year-­of-­life-­nhs. 42  See www.nice.org.uk/news/feature/changes-­to-­nice-­drug-­appraisals-­what-­you-­need-­to-­know (4 April 2017).

Access to Healthcare in Europe   253 access are also made for certain cancer drugs where there is significant remaining clinical uncertainty.43 Furthermore, in light of the commitment in the NHS Constitution to ‘the most effective, fair and sustainable use of finite resources’,44 a budget impact test was introduced in 2017, under which a commercial discussion between NHS England and the company is automatically triggered where a product has a net budget impact of £20m or more per year in any of the first 3 years of its use on the NHS.45 The outcomes of these discussions are made publicly available, suggesting a greater degree of transparency than is apparent in the SHI systems previously noted.46

3.2  Challenging Entitlements in Court 3.2.1  Social Health Insurance Systems Although lists of entitlements are most significant in defining the nature and scope of healthcare benefits under an SHI scheme, the judiciary has also struggled with the insured’s health insurance claims. The list is not a mere declaratory statement. Rather, individuals may challenge their benefit entitlements in SHI countries as set by law (and individual health insurance plans), obliging states to comply with their statutory and constitutional obligations. Failure to guarantee these contracted services makes health funds, in principle, liable to legal claims. But the judiciary in SHI countries has not accepted an absolute right to healthcare for the insured and has thus permitted certain restrictions made by competent administrative bodies. The judiciary—as under the NHS model—has largely accepted the wide discretionary powers of policy-­making in defining the type and scope of the SHI benefit package, but has imposed more rational decision-­making based on objective, evidence-­based criteria and excluding discrimination.47 Particularly in life-­threatening situations, domestic courts have been confronted with harsh decisions on the reimbursement of a non-­listed medicine or treatment. Although the judiciary generally respects such administrative decisions (restricting itself to procedural review), on several occasions national courts have concluded such decisions to be unconstitutional or to violate basic human rights (substantive review). The most prominent example is the Nikolaus case, in which the German Constitutional Court lifted the ban on reimbursement of an experimental treatment method.48 This 43  NHS England, ‘Appraisal and Funding of Cancer Drugs from July 2016 (Including the New Cancer Drugs Fund) A New Deal for Patients, Taxpayers and Industry’ (8 July 2016). 44  Department of Health (n. 40), Principle 6. 45  See n. 42. For criticism, see V Charlton et al., ‘Cost Effective but Unaffordable: An Emerging Challenge for Health Systems’ (2017) 356 Br Med J j1402. 46  See NHS England, ‘NHS England Strikes Deal on New NICE Recommended Lung Cancer Immunotherapy Drug’ (6 June 2018), www.england.nhs.uk/2018/06/nhs-­england-­strikes-­deal-­on-­new-­ nice-­recommended-­lung-­cancer-­immunotherapy-­drug/. 47  E.g., Dutch Supreme Court 30 March 2018, ECLI:NL:PHR:2017:1609; Sofia Appeal Court Decision 209, 1 August 2007; Latvian Supreme Court, Case No. A42561808, SKA—13/2010, 11 March 2010; Swiss Federal Court, BGE 129 V 167f [2003]. 48  Case BvR 347/98, 6 December 2005, also known as the ‘Nikolausbeschluss’.

254   André den Exter and Keith Syrett case involved a young patient suffering from Duchenne muscular dystrophy, a progressive and lethal illness. At present, there is no effective therapy for this disease. Reimbursement of the costs of a new treatment method, the so-­called immune biological therapy, was rejected by the social insurance fund on the ground that it was not evidence-­based (the wirksamkeit criterion). The Court ruled, however, that statutory criteria for limiting health benefits (i.e., ‘ausreichend, zweckmässig, wirtschaftlich’) should be interpreted in line with constitutional values such as the right to life, bodily integrity, and the welfare (or social) state principle.49 More specifically, in case of life-­threatening diseases for which medical treatment is lacking according to general medical standards, an experimental treatment with curative or positive effect (‘spürbare positive Einwirkung’) on the disease course cannot be denied in the absence of scientific evidence. The alternative’s effectiveness could be based on other evidence, for instance expert opinions and medical practice.50 With this ruling the Court has, albeit in exceptional cases, extended healthcare access to newly developed, and in most cases extremely expensive, diagnostic and treatment methods that are likely to have a positive effect.51 It means that when scientific evidence is absent, the required probability standard of effectiveness is rather flexible: the more severe and hopeless the situation, the less stringent the likeliness standard.52 And although the Court recognized the need for cost or cost-­ benefit considerations (‘Wirtschaftlichkeitsgebot’, Art. 12 SGB V),53 these criteria were not decisive. The Nikolaus ruling stirred feelings in German legal doctrine.54 In essence, it shows that despite the legislature’s (here the Federal Joint Committee, G-­BA) discretionary powers to formulate binding guidelines on evidence-­based medicine and applied selection criteria, standards should ultimately comply with constitutional values. Similar examples have been litigated in the Netherlands, inspired by the Nikolaus case, although not based on constitutional rights but rather on the civil law principle of ‘reasonableness’.55 The permissive approach of the German Constitutional Court seems contrary to the Swiss Supreme Court’s ruling in the Myozyme I case.56 On appeal, a Swiss health insurance fund challenged the court order of the Insurance Tribunal to continue reimbursement of an experimental treatment for Pompe’s disease, a rare and life-­threatening condition. 49  ibid, 55. 50  ibid, 66. 51  See also Art. 12(3) SGB V incorporating the Nikolaus ruling. Examples accepted under this provision concern an experimental combined therapy for ovarian cancer (€15,000 p.m.) BvR 2045/12, 26 February 2013; experimental stem cell transplantation LSG Baden Würtemberg, 13. November 2012, L11 KR 2254/10. 52  Still, the life-­threatening element has been interpreted narrowly, i.e., an immediate threat and not a potential lethal outcome: BVerfG, 1 BvR 452/17, May 11, 2017, paras 26–27. 53  At 57–59. 54  J Huster, ‘Anmerkung’ (2009) 9 JuristenZeitung 466–468; G Dannecker and A Streng, ‘Die Bedeutung des Nikolaus-­Beschlusses für die Priorisierungsdebatte’ in B. Schmitz-­Luhn and A. Bohmeier (eds.), Priorisierung in der Medizin. Kriterien im Dialog (Springer 2013) 135–146. 55  E.g., Dutch Supreme Court, HR 19 December 2014; denied in Supreme Court Latvia, Case No. A42561808, SKA—13/2010, 11 March 2010 (Gaucher disease and the constitutional right to health). 56  Judgment of the Federal Supreme Court of Switzerland of 23 November 2010 (BGE 136 V 395).

Access to Healthcare in Europe   255 The Supreme Court annulled the Tribunal’s ruling by reasons based both on lacking clinical effectiveness (‘Wirksamkeit’) and cost-­effectiveness. The costs of treatment were calculated at CHF 700,000 per year (€565,000). Because general criteria to assess cost effectiveness were absent, the Court applied a—controversial—cost-­benefit analysis, concluding that the excessive costs of treatment would be disproportionate compared to the benefit (i.e., only relieving the symptoms of the disease, not postponing or preventing its fatal outcome). Moreover, approval would violate the equality principle in that a disproportionate amount of scarce resources would be allocated to a certain individual but not to others who are in the same position.57 This line of reasoning has been criticized by legal scholars.58 Although cost-­benefit/effectiveness analysis is relevant at the macro level (benefit package decision-­making), it seems less appropriate at the individual doctor–patient level since it will ultimately force the judiciary to decide on society’s willingness to pay for rare diseases, a question that can only be answered by the legislature. Distinctly from the Nikolaus case, the Swiss Supreme Court declined to review the constitutionality of denial of access under the right to life, personal freedom, and the right to assistance when in need.59 As these rights were not invoked in the Supreme Court, it could abstain from such a human rights assessment. Ultimately, this case triggered public deliberation which resulted in a federal by-­law providing a legal basis and guiding principles of cost considerations in coverage decision-­making, but without setting a threshold.60 Instead, health insurance funds are supposed to review reimbursement of expensive interventions on a case-­by-­case basis, applying cost-­effectiveness evidence (i.e., in case of non-­listed medicines). Several years later, in Myozyme II, the Swiss health insurer again refused the reimbursement of a myozyme treatment in case of morbus pompe.61 Different from Myozyme I, from 1 November 2011, Myozyme was included in the so-­called ‘Spezialitätenliste’ (special list of medicines, SL), followed by a 50% price reduction (approximately CHF 370,000 per year).62 Listed medicines comply with the statutory conditions of clinical and cost effectiveness and efficiency (Art. 32 Swiss HIA) and should therefore be reimbursed. Refusal of cost reimbursement can only be justified in case of noncompliance with the 57  ibid, 7.7–7.8. 58  E.g., F Kesselring, ‘First Fundamental Decision of the Federal Supreme Court of Switzerland on Cost-­Effectiveness in the Area of Human Health Care’ (2011) 3 Eur J Risk Regulation 442–446; S Huster and A Bohmaier, ‘Die Myozyme-­Entscheidung des Sweizerischen Bundesgerichts aus der Perspektive des deutschen Verfassungs- und Krankenversicherungsrecht’ (2012) 106 Zeitschrift für Evidenz, Fortbildung und Qualität im Gesundheitswesen 443–448. 59  Articles 10 and 12 of the Swiss Federal Constitution. 60  Federal By-­law on Health Insurance AS 2011 654 (Explanatory note), Art. 71a(3) KVV, reading: ‘Die zu übernehmenden Kosten müssen in einem angemessenen Verhältnis zum therapeutischen Nutzen stehen (. . .)’, which can be interpreted as an implicit cost-­benefit assessment, idem Art 71 b (3) KVV; confirmed by the government‘s reply on Parliamentary question no 11.3154 (6 June 2011), in particular question no 4. 61  BGE 142 V 478, 16 September 2016. 62  But still far more than the Myozyme I threshold of CHF 100,000 per QALY.

256   André den Exter and Keith Syrett SL restrictions (e.g., medical indication, provided by a qualified physician, etc.). Taking Myozyme’s cost effectiveness for granted, the court concluded that the insurer’s duty of care required cost reimbursement of the claimed treatment.

3.2.2  National Health Systems The limited purchase which legally enforceable ‘rights’ possess in relation to access to healthcare in the English NHS is perhaps best demonstrated in the context of resource allocation. Although the NHS has been forced to limit access to treatments and services on financial grounds from the very early days of its existence, judicial engagement with ‘rationing’ decisions only started to gather momentum in the mid-­1990s, in the context of a shift (common to many health systems at the time) from ‘implicit’ mechanisms— where allocative decisions were masked by clinical judgement—to ‘explicit’ forms where the existence of financial constraints to access was visible to those affected. This greater visibility appears to have engendered instability, manifest in part by increasing recourse to the courts.63 As a starting point, the law draws a clear distinction between a right to refuse treatment—grounded in patient autonomy—and a right to demand treatment: in the case of the latter, ‘autonomy and the right of self-­determination do not entitle the patient to insist on receiving a particular medical treatment’,64 in part at least because this would disrupt the rational allocation of finite resources. Consonant with the founding principle of the NHS, as also set out in the NHS Constitution,65 the law thus envisages establishment of clinical need as determining whether resources should, in principle, be allocated for a particular treatment or service. Nonetheless, the existence of such need is not decisive. The courts have ruled that it is lawful for an NHS decision-­maker to prioritise certain conditions, or treatments for such conditions, over others66 thus, in effect, generating a list of priorities, some of which may not be met at all if resources do not allow. English courts thus take cognisance of the opportunity costs generated by the finite nature of resources for healthcare. However, even though, as a consequence, there is no right to access a particular treatment or service, patients have been accorded rights of a procedural character. Procedural protection is afforded to the individual by obliging NHS decision-­makers to take account of factors which may render their case exceptional and which thus necessitate a departure from a general policy not to fund a service or treatment.67 The individual also benefits from an explanatory obligation based upon

63  For discussion, see K Syrett, ‘Impotence or Importance? Judicial Review in an Era of Explicit NHS Rationing’ (2004) 67 Mod L Rev 289–304. 64  R (Burke) v The General Medical Council [2005] EWCA Civ 1003, [31] (Lord Phillips). 65  Department of Health (n. 40), Principle 2. 66 See R v North West Lancashire Health Authority, ex parte A, D and G (n. 35). 67  See, e.g., R (Murphy) v Salford Primary Care Trust [2008] EWHC 1908 (Admin); and further National Health Service Commissioning Board and Clinical Commissioning Groups (Responsibilities and Standing Rules) Regulations 2012, SI 2012/2996, reg. 34(2)(ii).

Access to Healthcare in Europe   257 judicial decisions in the field68 and set out thus by the NHS Constitution: ‘if the local NHS decides not to fund a drug or treatment you and your doctor feel would be right for you, they will explain that decision to you’.69 This latter procedural right also has a broader community dimension in so far as it encourages transparency in resource allocation which is likely both to enhance the quality of decision-­making (e.g., by requiring that a decision-­maker be properly attentive to all relevant factors and can justify the choice made in relation to these, thus ensuring rationality) and to enhance public understanding of the need for ‘hard choices’ in healthcare and the criteria upon which these are based. This, in turn, can serve to facilitate realisation of the ‘accountability for reasonableness’ model of procedural justice, which has been widely promoted as a means of securing legitimacy for allocative decisions in the absence of substantive consensus on the ethical basis for these.70 By contrast, human rights of the type protected by the European Convention on Human Rights (ECHR) and given domestic effect in the United Kingdom by the Human Rights Act 1998, have had little impact in cases on access to healthcare in English courts. For example, in a case in which an individual was denied gastric band surgery since his body mass index fell below the threshold for funding, the court rejected an argument for access based upon interference with his right to respect for private life under Article 8 ECHR, pointing to Strasbourg jurisprudence on the matter (discussed in Section 3.2.3).71 With one quite specific exception,72 arguments based upon Convention rights have merely triggered an obligation for a decision-­maker to justify its choice in order to convince the court of its rationality (and thus, lawfulness), rather than providing bases for a substantive judicial reordering of priorities in order to give direct effect to the right(s) at play. It might be expected that the position would be different in systems, such as Spain and Italy, where health-­related rights are afforded a degree of protection through constitutional law. The presumptive legal (and political) weight attached to a claim formulated as a right would seem to afford less margin for arguments that community interests in the efficient and equitable distribution of scarce public funds necessitate denial of access to treatment to a particular patient. Certainly, in contrast to the position in England, rights-­based claims have formed the basis for successful litigation (e.g., to gain reimbursement for expenses incurred in utilising the private health sector in cases of ‘emergency’ access to treatments which are not funded publicly or which cannot be accessed

68  See especially R v Cambridge Health Authority, ex parte B [1995] 1 FLR 1055, 1065 (Laws J). 69  Department of Health (n. 40), section 3a. See also National Health Service Commissioning Board and Clinical Commissioning Groups (Responsibilities and Standing Rules) Regulations (n. 67), reg. 35. 70  See further K Syrett, Law, Legitimacy and the Rationing of Health Care: A Contextual and Comparative Perspective (Cambridge University Press 2007). 71  R (Condliff) v North Staffordshire Primary Care Trust [2011] EWCA Civ 910. 72  R v North & East Devon Health Authority, ex parte Coughlan [1999] EWCA Civ 1871 (unlawful to close care facility where residents had been promised a ‘home for life’: Article 8 ECHR [right to respect for home] engaged).

258   André den Exter and Keith Syrett in a suitably timely manner).73 However, it is notable that courts in these jurisdictions have, in other contexts, also exhibited sensitivity to the problem of scarcity,74 suggesting that they are attentive to the dangers of disrupting the collective activity of resource allocation through ‘judicialisation’ of the decision-­making process, a phenomenon which has proved highly problematic across Latin America.75

3.2.3  Europe-­wide While some European health systems have been more generous than others, all of them, however funded, are struggling with the phenomenon of citizens alleging entitlements to access healthcare treatments and services.76 In addition to cases brought in domestic courts, European central courts have sought to settle disputes relying on international human rights (ECHR, European Social Charter, European Social Security Code, and the Convention on the Rights of the Child). But except in relation to categories of vulnerable claimants, these courts have been largely reluctant to review what are considered to be national health policy issues.77 Their position is reflective of the still controversial status of the ‘human right’ to access healthcare, coupled with a reluctance to disturb allocative choices made by national authorities. Illustrative is the Sentges case, requesting a highly expensive medical device (a robotic arm) that was neither approved nor listed as a health insurance entitlement. Under those circumstances, the European Court of Human Rights (ECtHR) refused to interfere in the state’s space for manoeuvre in determining the scope of the health insurance entitlement.78 This position was restated in the Wiater case, as follows: ‘The allocation of public funds in the area of healthcare . . . is not a matter on which the Court should take a stand. It is for the Member States to consider and decide how their limited resources

73  See, e.g., STSJCL 660/2009 (treatment for eating disorder, personality disorder and alcoholism) (Spain). 74  See, e.g., Cost. 104/2013 (Italy); STS 16-­11-­2009 App no 4426/2008 (Spain). 75  For discussion, see, e.g., E Lamprea, ‘The Judicialization of Health Care: A Global South Perspective’ (2017) 13 Ann Rev Soc Sci 431–449, and articles in a special section on ‘Judicial Enforcement of Health Rights: Focus on Latin America’ (2018) 20 Health Human Rights 67–197. 76  See, e.g., C Flood and A Gross (eds.), The Right to Health at the Public/Private Divide: A Global Comparative Study (Cambridge University Press 2014). 77  See for instance Nitecki v Poland App no 65653/01 (ECtHR, 21 March 2002); Sentges v Netherlands, App no 27677/02 (ECtHR, 8 July 2003), 30909/06 (ECtHR, 10 April 2012); Hristozov v Bulgaria Apps nos 47039/11 and 358/12 (ECtHR, 21 February 2012) 13 November 2012); Wiater v Poland, App no 42290/08 (ECtHR, 15 May 2012). An exception is Panaitescu v Romania, App no 30909/06 (ECtHR, 10 April 2012), where the Romanian health Insurance fund refused to enforce a domestic court order requiring provision of the necessary cancer treatment for free. In this exceptional case of unreasonable bureaucratic unwillingness, the Court ruled that the state had not adequately protected the patient’s right to life, holding a breach of the state’s procedural obligation under Article 2 ECHR. More successful have been health-­related complaints concerning vulnerable categories of patients (prisoners, women, children). See, e.g., D v United Kingdom App no 30240/96 (ECtHR, 2 May 1997) (discussed in Section 4.3), Pantea v Romania App no 33343/96 (ECtHR, 3 June 2003), Wenner v. Germany, App no 62303/13 (ECtHR, 1 September 2016), etc. 78  Sentges, see n. 77.

Access to Healthcare in Europe   259 should be allocated’.79 More recently, in McDonald v United Kingdom, both national courts and the ECtHR were asked whether local authorities were allowed to withdraw or amend care support where the recipient’s circumstances are unchanged but where a cheaper alternative is available. Again, the ECtHR reaffirmed that states have a wide margin of appreciation in issues of healthcare policies and that this margin is particularly wide when the issues involve an assessment of priorities in the context of the allocation of limited state resources.80 More potentially successful is the reliance on free movement principles under EU law, which may entitle the patient to receive healthcare abroad. With the landmark cases Decker81 and Kohll,82 the Court of Justice of the European Union (CJEU) has gradually extended EU patients’ right to seek cross-­border healthcare. Absence of treatment options caused by undue delay has been generally accepted for reimbursement of treatment abroad.83 The central issue in these rulings is that the so-­called prior authorisation requirement, which is conditional for reimbursement of healthcare provided abroad, restricts free movement of patients and health services. Instead of a general ban on prior authorisation, the Court accepted such a barrier of free movement for reasons of public interest (i.e., financial sustainability of the healthcare system), but only in case of inpatient healthcare services since it is assumed that consuming most outpatient healthcare abroad will not disrupt the financial balance of national healthcare systems.84 The Court’s rulings opened a fierce debate touching the heart of healthcare policy-­ making; namely, the organisation and financing of healthcare. This resulted in the cross-­ border care Directive, officially the Directive on the application of patients’ rights in cross-­border healthcare (Directive 2011/24/EU) that came into force on 25 October 2013.85 Under the Directive, Member States remain responsible for providing safe, high-­ quality, efficient, and quantitatively adequate healthcare to citizens on their territory (Article 4). Simultaneously, they have to respect basic legal principles. In the case of cross-­border care this means applying objective, nondiscriminatory criteria—which should be known in advance—as well as providing access to a judicial review procedure if granting cross-­border healthcare is refused while taking into account all relevant circumstances. It follows that, when there is a delay, it is not permissible to use waiting lists merely as a necessary planning argument to refuse authorisation without taking into account the patient’s medical history, needs, and degree of pain (see further later). 79  Wiater v Poland, ibid para 39., 80  McDonald v UK App no 4241/12 (ECtHR, 20 August 2014), paras 54–55. 81  Case C-­120/95, Decker v. Caisse de maladie des employés privés, ECLI:EU:C:1998:167. 82  Case C-­158/96, Kohll v. Union des caisses de maladie, ECLI:EU:C:1998:171. 83  E.g., Case C-­368/98 Vanbraekel [2001] ECR I-­5363; Case C-­385/99 Müller-­Fauré and Van Riet [2003] ECR I-­4509; Case C-­372/04 Yvonne Watts [2006] ECR I-­4325; Case C-­444/05 Stamatelaki [2007] ECR I-­3185; Case C-­512/08 European Commission v French Republic [2011] ECR I-­8833; Case C-­173/09 Elchinov [2010] ECR I-­8889; and Case C268/13 Petru ECLI:EU:C:2014:2271. 84  With the exception of some cost-­effective invasive outpatient healthcare facilities like computed tomography (CT) scans. 85  Directive 2011/24/EU of the European Parliament and of the Council of 9 March 2011 on the application of patients’ rights in cross-­border healthcare OJ L 88/45, 4 April 2011.

260   André den Exter and Keith Syrett Second, prior authorisation can be justified for reasons of public interest, but only if the patient needs hospital and high-­technology care (Article 8).86 This restriction to inpatient healthcare is the direct result of previous Court of Justice rulings. Third, the obligation to reimburse cross-­border healthcare is limited to services to which the insured person is entitled. As such, the Directive respects the fundamental ethical choices of Member States. If controversial medical interventions (e.g., abortion, stem cell therapy) are excluded from the benefit package at home, one cannot claim reimbursement when such an intervention is performed abroad (Article 7(1)). Furthermore, the reimbursement level of healthcare costs abroad is limited to the assumed costs in the home Member State/state of affiliation. Although the Directive appears to reinforce the apparent individualistic trend of the Court’s case law on patient mobility, there is no evidence of an increase in patients searching for healthcare abroad since the Directive came into force.87 One of the reasons for the limited flow of cross-­border care cases is the difference in maximum reimbursement rate by country. The patient’s required direct payment can be considered a (significant) barrier to cross-­border care, particularly for low-­income patients,88 whereas language barriers may also discourage patients from seeking healthcare abroad. These, and other reasons that may clarify the limited financial impact of cross-­border care so far, raise a bigger concern of potential inequalities in access to cross-­border care access since one cannot ignore that certain strata (i.e., those who are healthy, wealthy, have access to information, speak languages, and are able to travel) may access EU-­based health rights, whereas others are left behind.89 Despite its laudable motive, the Directive has limitations. These can largely be explained by Member States’ unwillingness (legitimate or not) to regulate key issues of cross-­border healthcare at a pan-­European level.90 Such limitations require alternative action in order to realise an internal market for patient care while simultaneously respecting national values and traditions on how to organise and finance healthcare systems. At best, aiming at (indirect) convergence of national health systems by means of ‘soft-­law’ mechanisms seems the highest attainable objective so far. 86  Other reasons that justify prior authorisation are treatments presenting a particular risk for the patient or the population (art. 8(2)(b)) and healthcare provided by a provider that, on a case-­by-­case basis, could give rise to serious and specific concerns relating to the quality or safety of the care (art.8(2)(b)–(c)). 87  According to a recent study, the overall numbers of treatments abroad that require prior authorisation are too small to draw significant conclusions. For instance, in 2015, fewer than 4,000 requests for prior authorisation treatment abroad were granted (€24m), whereas the total amount of treatments without prior authorisation was €41m, although the study did not cover all EU member states: European Commission, ‘Member States Data on cross-­border patient healthcare following Directive 2011/24/EU, Year 2016’, 18 July 2018. 88  While the reimbursement rates may substantially differ by country (East-­West, South-­North). 89  Taking into account the Directive’s upfront payment requirement (Art.7(4)). See also T Hervey, ‘Patient Mobility, Solidarity, and Equal Access’ in A den Exter (ed.), Cross-­border Health Care and European Union Law (Erasmus University Press 2017) 19–34. 90  See D Sindbjerg Martinsen, ‘The Politics of the Cross-­Border Care Directive’ in den Exter, ibid. 5–17.

Access to Healthcare in Europe   261 Inspired by the Patient Mobility Directive, Member State authorities and health insurance funds have concluded bilateral and multilateral agreements with foreign health providers, allowing their patients to receive treatment or diagnostics abroad under similar conditions as in the state of affiliation.91 A recent study shows a diverse picture of collaboration initiatives addressing gaps in availability of healthcare services across Europe.92 With some exceptions, most cross-­border healthcare initiatives occur between countries or regions with similar welfare traditions and geographical proximity or connected via specific historical ties. Such initiatives trigger numerous questions on differences in patient safety and quality standards and procedures, patients’ rights, reimbursement issues, and liability norms to be addressed in the agreements made and whether or not they are based on international (private law) principles. It has been argued that these institution-­led (as opposed to patient-­led) border-­crossing initiatives promote patients’ healthcare access by reducing waiting times and simultaneously facilitate the exchange of information, thus enhancing quality and safety standards for those services provided.93

4  Who Can Access Treatment? 4.1  Membership as a Precondition for Social Health Insurance Eligibility Established by law, membership in SHI systems is mandatory and covers the entire or a large part of the population (workers and their relatives, pensioners, unemployed, self-­ employed, etc.), whether or not excluding or opting-­out categories of persons above a high-­income threshold (Austria, Belgium, and Germany). Here, complementary (private) insurance is the alternative. In case of several insurance funds operating in an SHI, the insured have a choice of insurer to increase efficiency through limited competition among insurers (Czech Republic, Germany, Netherlands, and Switzerland).94 Essential to membership is that both the insured as well as employers pay a contribution, creating a risk pool for the groups covered and reallocated according to various 91  Well-­known examples are the interregional partnerships on the Franco–Belgian border; German–France–Swiss border project (Trisan); and the Euregio Maas–Rhine, see European Commission, European Cross-­Border Cooperation on Health; Theory and Practice (EU 2017); I Glinos and R Baeten, ‘Dreams vs Reality: Seven Case Studies on the Desirability and Feasibility of Cross-­ Border Hospital Collaboration in Europe’ (2014) 117 Soc Sci & Med 18–24. 92  European Commission, Study on Cross-­border Cooperation in Health Care. Capitalising on Existing Initiatives for Cooperation in Cross-­Border Regions, Final Report (European Commission 2018). 93  ibid, 236–239. 94  Unless choice is excluded since the insured are affiliated to a specific fund determined by professional occupation and region (Austria, Belgium, France): see OECD, Health Care Coverage in OECD countries in 2012, OECD working paper no 88 (OECD 2012) 12.

262   André den Exter and Keith Syrett principles. These contributions are traditionally wage-­related, with the government subsidising low-­income categories unable to pay (e.g., unemployed). The formal role of the insured is therefore to pay a premium for receiving free access to a comprehensive package of covered health services. Membership is generally related to legal residency (i.e., those who lawfully stay in the country [legal status]). As a consequence, ‘migrants with irregular status’ or ‘irregular migrants’ are excluded from membership of a SHI. Instead, all SHI countries offer access to some level of healthcare services, including emergency, primary, and specialist secondary care, although the scope of services provided varies significantly. Except for France and the Netherlands, out-­of-­pocket payment for primary and secondary care (full cost) is the norm, thus seriously eroding ‘the entitlement’.95 In theory, private health insurance is open to irregular migrants, but in many cases the high costs make this option illusory in practice.96 But even when provided for free, as in the Netherlands, administrative requirements limit access to a restricted number of contracted health providers only and for a limited time. In Germany, the requirement on public servants to inform the immigration authorities is a further barrier for irregular migrants to visit a health professional.97

4.2  Eligibility for National Health Systems Care In NHS systems, eligibility is defined not by membership in plans, but rather by citizenship and/or residency. This can present a challenge to foundational principles of universality such as those expressed in the NHS Constitution, which proclaims that ‘the NHS is available to all’.98 However, NHS systems have taken steps to limit the access accorded to certain classes of the population, thus reducing the pressure upon finite healthcare resources. In part, the disquiet is over the practice of ‘medical tourism’, a long-­standing phenomenon (which, as noted previously, is facilitated by the free movement principles of EU law), but which creates political and ethical concern in systems where contributions by taxpayers fund (largely) free healthcare. This practice was at issue in a case in which it was argued that a failed asylum seeker had come to the United Kingdom not because of a fear of persecution but to receive free medical care, and was not entitled to it notwithstanding that he was likely to remain indefinitely in the country since he was unable to return home (to the Palestinian occupied territories). The Court of Appeal held that the relevant statutory wording (the Secretary of State is obliged to promote a ‘comprehensive health service designed to secure improvement in physical and mental 95  European Union Agency for Fundamental Rights https://fra.europa.eu/en/publication/2016/ healthcare-­entitlements-­migrants-­irregular-­situation-­eu-­28 96  S Thomson, T Foubister and E Mossialos, Financing Healthcare in the European Union, Challenges and Policy Responses (WHO, 2009). 97  S Spencer and V Hughes, Outside and In: Legal Entitlements to Health care and Education for Migrants with Irregular Status in Europe, Report (Compas, University of Oxford 2015) 12. 98  Department of Health (n. 40), Principle 1.

Access to Healthcare in Europe   263 health in the people of England’, not people in England)99 suggested ‘that the beneficiaries of this free health service are to be those with some link to England so as to be part and parcel of the fabric of the place. It connotes a legitimate connection with the country . . . as a rule, the benefits were not intended by Parliament to be bestowed on those who ought not to be here’, such as the failed asylum seeker in the case in question.100 Those without the requisite connection are not excluded from access to treatments and services in an NHS system, but these may not be provided free of charge. Thus, current rules in England stipulate that those who are not eligible will be charged upfront and in full for any care deemed by a clinician not to be ‘immediately necessary’ or ‘urgent’, although no charge is currently made for general practitioner and nurse consultations, nor for attendance at hospital accident and emergency departments, and there is also a range of exempted categories covering both services and patients.101 The Italian system is, in principle, more generous, in that undocumented immigrants are granted free access to emergency care and treatments related to communicable disease, pregnancy, and childbirth,102 but may also be granted a ‘temporary residing foreigner code’, by means of which continuing primary and secondary care may be accessed for conditions which are not dangerous in the short term but may become so over time.103 However, this must be read against the backdrop of the greater use of co-­ payments in the Italian system by comparison with the English NHS. The controversy which may be generated by measures which appear to contravene the principle of universal access in NHSs is amply demonstrated by the case of Spain. As late as 2012, it was argued that ‘in theory, Spain provides some of the broadest health coverage to undocumented migrants in Europe’.104 However, the position altered significantly in that year with the passage of RDL 16/2012, which adopted an approach more characteristic of an SHI system, restricting eligibility to the ‘insured’ (or their beneficiaries), defined as someone who contributes—or has previously contributed—to the system of social security.105 This meant that undocumented migrants were excluded from access to free healthcare services, with exceptions for emergency care for serious illness or accident (with ongoing specialized care until discharge) and obstetric and child care. The reform provoked widespread opposition, with 15 of the 17 Autonomous Regions

99  National Health Service Act 2006, s 1(1)(a). Emphasis added. 100  R (YA) v Secretary of State for Health [2009] EWCA Civ 225, [55] (Ward LJ). See also R(A) v Secretary of State for Health [2017]EWHC 2815 (Admin), especially at [44] (Irwin LJ and Haddon-­Cave J). 101  National Health Service (Charges to Overseas Visitors) Regulations 2015, SI 2015/238, amended by National Health Service (Charges to Overseas Visitors) (Amendment) Regulations 2017, SI 2017/256. 102  LD 286/1998, Article 35 (3). 103  See C Cuadra, Policies on Health Care for Undocumented Migrants in EU27 Country Report: Italy (Malmo University 2010). 104  B Gray and E Van Ginneken, ‘Health Care for Undocumented Migrants: European Approaches’ (2012) 33 Issues Int’s Health Policy 1–12, 6. 105  See L Peralte-­Gallegi, J Gené-­Badia, and P Gallo, ‘Effects of Undocumented Immigrants Exclusion from Healthcare Coverage in Spain’ (2018) 122 Health Policy 1155–1160, 1156.

264   André den Exter and Keith Syrett taking measures to limit its impact,106 while some healthcare professionals, supported by national and international medical organisations, refused to comply.107 Nevertheless, it has been estimated that the restrictions to access caused an increase in the mortality rate of undocumented migrants of 15% in the first 3 years of the strategy.108 Perhaps unsurprisingly, the replacement of the conservative government which had initiated the measure by a socialist-­led administration in 2018 has brought about the restoration of full eligibility.109

4.3  Europe-­wide Barriers to healthcare access for irregular migrants in Europe are not only matters of political debate. A general conclusion derived from studies published since 2007 is that national law has eroded governments’ international human rights obligations under the right to health by making access conditional upon an individual’s nationality or legal status.110 Viewed from relevant treaty provisions such as Article 12 of the UN’s International Covenant on Economic, Social, and Cultural Rights (ICESCR) and General Comment no. 14 (GC14), it is concluded that provision only of emergency or urgent care ‘cannot be regarded as an acceptable minimum level of healthcare’.111 Instead, GC14’s emphasis on inclusive and equitable care does not justify any differentiation between emergency, primary, and secondary care. Requesting upfront, full-­cost payment when visiting primary and secondary care facilities denies irregular migrants’ rights to healthcare. This has most clearly been confirmed by the European Committee of Social Rights, the Charter’s monitoring body, declaring that ‘legislation or practice which denies entitlement to medical assistance to foreign nationals, within the territory of a State Party, even when they are there illegally, is contrary to the Charter’.112 An excessive financial burden should not be placed upon individuals, meaning that states have the obligation to assist 106 Id. 107  ibid, 1157–1158. 108  A Mestres, G Casanovas, and J Castelló, ‘The Deadly Effects of Losing Health Insurance’, CRES-­UPF Working Paper #201804-­104, Centre de Recerca en Economia i Salut, Barcelona (2018) 16. 109  H Legido-­Quigley et al., ‘Spain Shows That a Human Response to Migrant Health Is Possible in Europe’ (2018) 3 Lancet Public Health e358. 110  E.g., Platform for International Cooperation on Undocumented Migrants, Access to Healthcare for Undocumented Migrants in Europe (2007); P Chauvin, I Parizot, and N Simmonot, Access to Healthcare for Undocumented Migrants in 11 European Countries (Médicins du monde 2009); HUMA Network, Access to Healthcare for Undocumented Migrants and Asylum Seekers in 10 EU Countries: Law and Practice (2009); European Union Agency for Fundamental Rights, Migrants in an Irregular Situation: Access to Healthcare in 10 European Union Member States (2011) and Cost of Exclusion from Healthcare—The Case of Migrants in an Irregular Situation (2015). 111  D Ingleby and R Petrova-­Benedict, Recommendations on Access to Health Services for Migrants in an Irregular Situation: An Expert Consensus (International Organization for Migration Regional Office Brussels 2016) 23. 112  European Committee on Social Rights (ECSR), International Federation of Human Rights Leagues (FIDH) v France, Complaint no 14/2003.

Access to Healthcare in Europe   265 citizens regardless of their residence status.113 And finally (with particular pertinence to the Spanish situation outlined previously), ‘the economic crisis cannot serve as a pretext for a restriction or denial of access to healthcare that affects the very substance of the right of access to healthcare.’114 Even though the decisions and conclusions of the European Committee are not directly enforceable in domestic legal systems, these authoritative interpretations must be respected and can help the development of social rights. Although the ECHR contains no specific right to access healthcare, Article 3 was engaged in two UK cases concerning the deportation of individuals who had been receiving free treatment for HIV/AIDS on the NHS but whose status had subsequently changed, rendering them no longer eligible. In the first of these, the House of Lords ruled that returning the patient (who was in the terminal stages of illness) to a country which had no facilities to provide adequate medical and palliative care amounted to a violation of Article 3 ECHR, which prohibits subjection to degrading treatment.115 However, this conclusion was not followed in the latter case, although it should be noted that the facts were distinguishable in so far as the patient’s symptoms were well controlled and he would receive care upon his return. In this instance, the House of Lords expressed concern that permitting the patient to remain on human rights grounds ‘would risk drawing into the UK large numbers of people already suffering from HIV in the hope that they too could remain here indefinitely so that they could take the benefit of the medical resources that are available in this country. This would result in a very great and no doubt unquantifiable commitment of resources’. This decision was subsequently upheld by the ECtHR, which stated that placing an obligation to provide ‘free and unlimited healthcare without a right to stay within its jurisdiction . . . would place too great a burden on Contracting States’.116 This statement suggests that the ECtHR’s general position on allocative questions, outlined in Section 3.2.3, will be determinative. However, when irregular migrants are held in custody or under strict control over their place of residence, an analogy with prisoners is not far-­fetched.117 Generally considered as a vulnerable group, the ECtHR has accepted the authorities’ duty to protect those in custody under the Convention. Since the ECtHR’s case law on the treatment of prisoners is very strict, covering all types of necessary medical care, this could substantially strengthen the position of the individual irregular migrant excluded from necessary healthcare, including in national courts.118 113  B Inverardi, Accessing HIV Prevention, Testing, Treatment Care and Support in Europe as a Migrant with Irregular Status in Europe: A Comparative 10-­Country Legal Survey (European HIV Legal Forum 2015) 61. 114  ECSR Conclusions XX-­2 Spain (2014) 14. 115  D v United Kingdom (n. 77). 116  N v Secretary of State for the Home Department [2005] UKHL 31, [53] (Lord Hope); N v United Kingdom [2008] ECHR 453, [44]. 117  As suggested by C Slingenberg, Between Sovereignty and Equality. The Reception of Asylum Seekers under International law (PhD thesis, Vrije Universiteit Amsterdam, 2012), ch. 12. 118 E.g., Kudla v Poland, App no 30210/96 (ECtHR, 26 October 2000); Vasyukov v Russia, App no 2974/05 (ECtHR, 5 April 2011); Marian Chirita v Romania, App no 9443/10 (ECtHR, 21 October 2014); Dorneanu v Romania, App no 55089/13 (ECtHR, 28 November 2017); and Botnari v the Republic of Moldova, App no 74441/14 (ECtHR, 5 June 2018).

266   André den Exter and Keith Syrett

5  When Can Patients Access Healthcare? 5.1  Social Health Insurance Systems In case of medical need, patients may directly visit primary care providers (family physicians, midwives, pharmacists, etc.), but access to specialist secondary care is sometimes contingent on referral by a primary care provider. In such instances, primary care then functions as the ‘gate-­keeper’, managing patient entry into the system, particularly when there are long waiting times for secondary care. Nowadays, measuring unmet needs has been used by policy-­makers to address barriers in healthcare access, such as waiting times. Unmet needs for specific healthcare-­related services exist for a variety of reasons, for example because of costs (too expensive), distance (too far to travel), or timeliness (waiting times or lists for examination or treatment). Europe-­wide, statistics show that, in particular, long waiting times and substantial out-­of-­pocket expenses (i.e., healthcare costs not paid for or reimbursed) are major causes of unmet needs and thus delaying access to necessary healthcare. Eurostat results in 2014 show that more than one-­quarter of the EU-­28 population had an unmet healthcare need due to costs, distance, or waiting lists.119 Among SHI systems, long waiting times were identified as the main reason for delay in healthcare access, varying from 26.2% (Poland) to 31% (Luxembourg).120 Notably in central and eastern European countries, high costs were reported as the second most important barrier in delaying healthcare access (Slovakia, Hungary, Estonia, and Latvia). By contrast, 2016 data reveal an average decrease in unmet needs of the population aged 15 years and older, with some exceptions (e.g., dental care, mental health services).121 At the national level, policy-­makers have dealt with waiting lists for healthcare services in different ways. For instance, since the Dutch legislature introduced more market competition in healthcare and deregulated the strict hospital planning system in 2006, newly established private clinics providing specialised care (e.g., orthopaedic surgery, ophthalmology, etc.) have been contracted by competing health insurance companies, decreasing waiting times substantially. Consumer choice, or the ‘exit’ option, has forced health insurers to contract and increase healthcare supply. In addition, health insurers have been quite active in ‘waiting time mediation’, which is based upon the insurer’s statutory ‘duty of care’ to guarantee adequate access to the insured entitlements (Art. 11(1) HIA). Alternatively, the insured may also successfully claim reimbursement of noncontracted healthcare (abroad) in case of exceeding maximum waiting times.122 Still, long waiting lists in mental healthcare services (for minors) confirm the Eurostat 119 Eurostat, Statistics Explained Data 2018 www.ec.europa.eu/eurostat/statistics-­explained/index. php?title=Main_Page 120 Id. 121 Id. 122 ECLI:NL:CRvB:2012:BY1363.

Access to Healthcare in Europe   267 data on unmet needs and delay in both examination and treatment. As a result, by law, hospital providers are obliged to inform health insurers and the public in case of waiting lists.123 The rationale is that transparency allows health insurers to act as prudent buyers, purchasing adequate and timely healthcare services elsewhere. Other SHI systems are also familiar with extending healthcare supply options (higher spending, incentive mechanisms) to tackle long waiting times.124

5.2  National Health Systems Delays in access to treatment are familiar features of NHSs. In England, the NHS took on a waiting list of 476,000 patients at its establishment in 1948. By 1956, inpatient waiting times for trauma and orthopaedic, ear, nose, and throat, and gynaecological specialty services stood in excess of 80 days.125 It was not until the turn of the century that waiting times began to fall consistently, and this trend has been reversed since 2010, as the rate of growth of funding for the NHS has slowed. The law may play a role in regulating waiting times in such systems. In England, the NHS Constitution sets out a ‘right to access certain services commissioned by NHS bodies within maximum waiting times, or for the NHS to take all reasonable steps to offer you a range of suitable alternative providers if this is not possible’.126 The standards are set out in secondary legislation—there is a duty placed upon the local CCGs and NHS England to ‘make arrangements to ensure’ that not less than 92% of eligible persons have been waiting to commence consultant-­led treatment for less than 18 weeks from referral in the case of non-­urgent conditions and a further duty to ensure that 93% of urgent referrals for suspected cancer are seen by a specialist within 2 weeks; as provided by the Constitution, ‘all reasonable steps’ must be taken to offer an alternative provider of healthcare if the original body is unable to comply with these standards.127 In Spain, Law 16/2003 imposes a legal obligation to guarantee maximum waiting times, with exceptions for organ and tissue transplants: this is fleshed out by RD 1039/2011, which provides for maximum 180-­day waiting times for various surgical 123  www.volksgezondheidenzorg.info, search for ‘wachttijden’ (waiting times). 124  E.g., B Zurl et al., ‘ÖGRO Survey on Radiotherapy Capacity in Austria: Status Quo and Estimation of Full Demands’ (2018) 194 Strahlentherapie und Onkologie 284–292; K Osowiecka et al., ‘How Long Are Cancer Patients Waiting for Oncological Therapy in Poland?’ (2018) 15 Int’l J Environ Res Public Health 577–591. 125  King’s Fund, Briefing: In the Beginning (2008) 7–9. 126  Also a non-­binding pledge to provide convenient, easy access to services within the waiting times set out in Department of Health, Handbook to the NHS Constitution (2015) 33. 127  Department of Health (n. 40), section 3a; The National Health Service Commissioning Board and Clinical Commissioning Groups (Responsibilities and Standing Rules) Regulations 2012 (n. 67), Part 9. Certain exceptions to the maximum waiting times apply, such as cases where delaying treatment is in the clinical interests of the patient (e.g., where smoking cessation or weight management prior to the treatment is likely to improve outcomes), or where monitoring in secondary care is clinically appropriate; certain services are also excluded, including maternity services and public health services provided by local government rather than the NHS.

268   André den Exter and Keith Syrett interventions (heart surgery, cataracts, knee and hip prosthesis). This guarantee is allied to a system of information on waiting times established by RD 605/2003, with publication of data on a biannual basis.128 As for Italy, a 2002 agreement between the national and regional governments set maximum waiting time targets for various forms of ambulatory care, oncology services, cataract surgery, hip replacement, and percutaneous transluminal coronary angioplasty;129 90% of patients must be treated within the maximum period.130 If these targets are exceeded, patients have the right to free provision of the treatment or service from the private sector (or merely to pay the requisite co-­payment).131 Subsequent National Plans for the governance of waiting lists, issued pursuant to Law No. 26 of 23 December 2005, have imposed additional requirements upon regional authorities in this regard,132 and there is an obligation to publish waiting information online.133 The effectiveness of these legal provisions is, however, highly questionable. In the United Kingdom, the targets have been missed on a frequent basis since 2016, and NHS England acknowledged in a strategic document published in March 2017 that there was unlikely to be compliance ‘over the next couple of years’.134 Nonetheless, while this has generated political controversy and a legal opinion that the position was unlawful,135 the matter has not (yet) resulted in litigation. On the SNS, waiting times have increased since 2010 for those services covered by RD 1039/2011,136 most likely as a consequence of the introduction of austerity measures by the Spanish government in light of the financial crisis.137 And in Italy, lengthy waiting lists have resulted in significant shifts to private healthcare, as well as deployment of clandestine strategies to gain swifter access, such as bribery and use of personal connections.138 As noted earlier, free movement principles in EU law, now embodied in Directive 2011/24/EU, provide an additional avenue for access to treatment for patients suffering ‘undue delay’ in access to treatment in NHS systems. In this context, it is important to 128  Article 4(1). 129  Ministero della Salute, Rilevazione nazionale degli interventi attuali in tema di liste e tempi d’attesa dalle Aziende Usl ed ospedaliere (30 June 2002). 130  See Ministero della Salute, Piano Nazionale di contenimento dei tempi di attesa per il triennio 2006–2008 (3 April 2006). 131  LD 124/98, Article 3(13). 132  E.g., Ministero della Salute, Piano Nazionale di Governo delle Liste Diattesa per il Triennio 2010–2012 (October 28, 2010); Piano Nazionale di Governo delle Liste di Attesa 201-­2021 (February 21, 2019). 133  Law N. 266, 23 December 2005, Article 1, para 280 (e), referring to State-­Regions Understanding, 23 March 2005, Article 3, para 6. 134  NHS England, Next Steps on the NHS Five Year Forward View (2017) 47. 135  See www.drphilhammond.com/blog/wp-­content/uploads/2017/05/Legality-­of-­Dropping-­18-­ week-­RTT-­target.pdf. 136  See OECD/EU, Health at a Glance: Europe 2018: State of Health in the EU Cycle (OECD Publishing 2018) 189. 137  Bernal-­Delgado et al. (n. 37), 137–138. 138  See ‘Wait List Mismanagement Reveals Corruption in Italy’s Public Healthcare System’, La Stampa in English (19 January 2018).

Access to Healthcare in Europe   269 understand that waiting lists may not merely be symptomatic of bureaucratic inefficiencies, but also function as a means of managing the mismatch between the demand for and supply of healthcare resources which are free at the point of use—that is, ‘rationing by delay’. In the Watts case, the United Kingdom argued that ‘patient migration would be liable to put at risk the competent member state’s planning and rationalisation efforts in the vital healthcare sector so as to avoid the problems of hospital overcapacity, imbalance in the supply of hospital medical care and logistical and financial wastage’.139 However, the Court considered that this was, on its own, insufficient rationale for refusal to authorise treatment elsewhere in the EU in the absence of consideration of an ‘objective medical assessment’ of the particular patient’s need for treatment.140

6  Provider’s Duty to Provide Care In most SHI systems, the patient–purchaser relationship is based on the legal duty of the insurance fund to organise care, as set by SHI law. This is fundamentally different in the doctor–patient treatment relationship, which is essentially a contractual relationship based on civil law (services contract)141 or more specific medical treatment agreement rules.142 Freedom of contract, as a key principle in civil law, therefore regulates the doctor–patient relationship (in terms of rights and obligations). Therefore, there is no legal duty of care.143 Patients are free to choose the preferred physician, and physicians are also free to contract with patients of their choice. Clearly, some general restrictions to freedom of contracting are applicable (e.g., prohibition of discrimination based on race, gender, etc.), as well as restrictions set by SHI law and the insurer–provider agreement. Both civil law and health insurance law may refer to professional rules, directly and indirectly regulating the doctor–patient relationship.144 Generally characterised as a civil law contract, contractual obligations may concern general topics including provision of medical services and necessary information, patient’s consent, the reimbursement of services provided, etc.,145 but also more specific topics in the doctor–patient relationship, such as medical confidentiality and data protection.146 Finally, it is generally accepted that the services or treatment contract includes an obligation to provide services according to the standards of medical science and not an obligation of result.147 139  [2006] QB 667, [71]. 140  ibid, [123]. 141  As in the German Civil Code, §630b BGB ‘Dienstvertrag’ (BGH, NJW 1975, 305); or ‘auftragsrecht’ according to §1151(1) ABGB (Austrian CC); §394 OR (Swiss Code of Obligations). 142  As in the Netherlands, in which the Medical Treatment Agreement Act (WGBO) is part of the Civil Code, specifying rules for the doctor–patient relationship. 143  Except for emergency situations, e.g., § 48 Ärtzegesetz (Austria); also § 7–2 (Muster) Berufsordnung M-­BO (Germany). 144  E.g., the professional standards, Art. 453 WGBO (Dutch CC) and HIA (Bzvw) art. 2.1. 145  E.g., § 630c-­g BGB Germany. 146  E.g., Arts 457, 459 WGBO (NL). 147  C Uleer, J Miebach, and J Patt, Abrechnung van Arzt- und Krankenhausleistungen. Kommentar (Beck Verlag, 2006) Einleidung, A, p 1.

270   André den Exter and Keith Syrett Medical science standards do not necessarily mean the most expensive treatment available, as the treating physician should also take cost considerations into account, opting for the most clinical and cost-­effective treatment option. Whatever intervention is decided upon, the physician should inform the patient adequately, including the excluded alternative(s). The situation in NHS systems is somewhat complex. In the United Kingdom, there is no contract between patient and physician other than in the private healthcare context. A physician’s duty to provide care to a patient is therefore founded upon tort law principles backed by a system of professional regulation, although the relationship between certain physicians (notably, general practitioners) and the NHS itself is a contractual one. However, in Italy, the physician–patient relationship has been deemed to be contractual, premised (in the case of hospital care) on the basis that ‘the acceptance of the patient in the hospital for admission or for a clinical control, involves the conclusion of a contract,’ with the relationship being seen as a species of social contract.148 As for Spain, the dichotomy between contract and tort is irrelevant in the context of public healthcare, whereas in respect of private healthcare, patients in practice have a choice between the two forms of action.149

6.1  Conflicts and Termination of the Provider–Patient Relationship Conflicts in respect of contractual obligations are primarily regulated by civil law, whether or not combined with use of medical disciplinary boards reviewing the doctor’s conduct against professional standards. This, however, does not exclude legal action taken against the SHI entity, for example for failure to comply with its legal duty to organise necessary care adequately (e.g., in case of long waiting times or contracting substandard care). As regards termination of the physician–patient relationship, since there are no lifetime contracts, the doctor–patient relationship can in principle be terminated by both parties respecting mutual interests. Due to the specific nature of the doctor–patient relationship, physician-initiated termination of the treatment agreement has generally been approached with due caution (‘compelling reason’ and ‘ensuring continuity of care’).150 Courts, as well as medical disciplinary boards, have accepted that the physician may end the agreement but only with due caution and after careful attempts to resolve the problem. Noncompliance with the treatment plan, lack of trust, aggressive behaviour, repeat148  See Italian Court of Cassation, United Sections, Case 577/2008, cited in F Traina, ‘Medical Malpractice: The Experience in Italy’ (2009) 467 Clin Orthopaed Related Res 434–442, 436. 149  See M Paz García Rubio and H Trigo García, ‘The Development of Medical Liability in Spain’ in E Hondius, The Development of Medical Liability, Volume 3 (Cambridge University Press 2012) 160–187, 173–174. 150  E.g., Art. 627(2) BGB (Germany).

Access to Healthcare in Europe   271 edly not showing up for appointments, non-­payment of services, absence of medical need, and distance (when a patient has moved outside the practice’s designated catchment area) have all rendered justified reasons for physicians terminating their patient relationships.151 A lack of trust is an abstract notion that has been accepted, for instance, in cases of sexual advances made by the patient or when a patient persistently acts inconsiderately or unreasonably, but not solely because of a complaint they have made.152 Proper termination requires a reasonable notice, after which the physician will no longer be available for clinical issues, by notifying the patient in writing, with provision of reasons for the decision.153 Also, the physician should support the patient in the transfer by ensuring arrangements are made for continuing care. In case of hospitals, however, ending an existing treatment contract (if relevant) is not allowed as long the patient is in need of inpatient care, except when the necessary transfer to an alternative hospital has been safeguarded (Austria).154 A survey of more than 300 UK GPs showed that more than half of the practices had removed patients from their list (i.e., ending their professional relationship) in the preceding 3 years for reasons of violence or threatening behaviour, criminal behaviour, and/ or deception or other issues resulting in irretrievable breakdown of the doctor–patient relationship.155 The breakdown of trust means the physician cannot provide good clinical care and therefore justifies ending the professional relationship with a patient.

7 Conclusion Searching for high-­quality care, cost effectiveness, and more efficiency in healthcare services accessible to all will remain the main drivers for healthcare systems in Europe. In this process, steps toward more transparency in the allocative decision-­making process of scarce healthcare resources have been made or are being considered. But, in the end, transparency and participatory decision-­making, and thus procedural democratic requirements, will surely appear insufficient to satisfy virtually infinite medical needs. Hence, more substantive criteria are needed to justify limits to healthcare access. Economic criteria like cost effectiveness and QALY thresholds are politically highly 151  E.g., Germany: OLG München 1 U 3395/07; AG Karlsruhe 9 C 251/97; Austria: OGH 5 Ob 578/80; NL: ECLI:NL:RBLIM:2016:2725; ECLI:NL:RBOVE:2017:4481; ECLI:NL:RBLEE:2003:AI1813; ECLI:NL:TGZEIN:2018:47; ECLI:TGZCTG:2017:228. In the UK, the only plausible basis for a legal challenge to termination would be an action in defamation, although the patient might argue that there had been a failure to comply with professional codes of conduct. 152  See also General Medical Council, ‘Ending Your Professional Relationship with a Patient’ (22 April 2013). Also Parliamentary and Health Service Ombudsman decisions on patient removal: (no 139, 2013), (no 598, 2014), (no 1023, 2015). 153  ECLI:NL:RBOBR:2015:2183; General Medical Council, id. 154  Art. 24 (1) Federal Law on Hospital Facilities and Sanatoria (KAKuG) Austria. 155  J Molodynski, ‘Removing Patients’, www.medicalprotection.org/uk/articles/removing-­patients (6 April 2013).

272   André den Exter and Keith Syrett controversial and also problematic from a human rights perspective, as these thresholds exclude categories of vulnerable groups (e.g., persons suffering from rare diseases). What remains is the search for additional substantive principles to ensure a fair distribution of scarce healthcare resources. This will constitute the real and ongoing challenge of healthcare systems in Europe, as well as elsewhere.

R EGU L AT ION OF H E A LT H C A R E SERV ICE S , FAC I L I T I E S , A N D T R A NSAC T IONS

Chapter 13

I n troduction to R egu l ation of Professiona l s a n d Facilitie s Deirdre Madden and Isaac D. Buck

The following two chapters by Isaac D. Buck and Deirdre Madden examine the regulation of healthcare providers and facilities within the United States and Europe. There are many aspects to the administration of healthcare systems and clinical practice that are subject to legal or regulatory oversight to ensure that appropriate standards are maintained and the interests of the public are protected. The following chapters seek to present a summary of the respective regulatory structures. Even though the regimes differ in key areas, systems in the European Union (EU) and United States share certain overlapping themes. At a basic level, the regulation of professionals and facilities differs from a medical malpractice regime that compensates injured patients following a violation of the standard of care or from a fraud and abuse regime that focuses on financial harms. Instead, quality regulation in both systems seeks to prevent harm to patient health from occurring in the first place. Both European and American systems also face challenges posed by definitional limitations around the practice of medicine, and both employ different types of regulation (and softer incentives) in different scenarios. Making sure the rules protect patient access to care can often aggravate or complicate the size and scope of regulatory structure, as does the threat of regulatory capture. Finally, both systems grapple with balancing an appropriate level of regulation to protect patients without imposing excessive costs, all the while trying to maintain public trust and confidence. These challenges are universal in nature. Nonetheless, the systems also differ in key ways. Most prominently, the American system has a more comprehensive and robust regulatory structure for hospitals and entities, particularly its private ones, than some European systems although there is

276   Deirdre Madden and Isaac D. Buck wide divergence between national systems in Europe. Additionally, the European system may rely less on “soft” tools, such as consumer decision-­making, to bring about regulatory goals as, by and large, citizens either use publicly provided healthcare in which there is less opportunity for consumer decision-­making or, in the context of private healthcare, expect the state to impose regulatory standards and oversight on private providers. In this way, there is less scope for “consumerism” in European systems whereas America frequently relies on consumer-­based tools to bring about or encourage key changes to healthcare delivery. The first chapter focuses specifically on the American project, seeking to summarize the types of regulation that govern healthcare providers and institutions in the United States. In the midst of an era of great policy uncertainty, healthcare in the United States is regulated and shaped by a varied number of actors and rules. Because quality control is essential in all healthcare settings, its broad, prophylactic rules apply universally throughout the industry. The rules govern the entirety of medical practice: market entry, licensing and accreditation, surveys and review, and sanction and discipline, as well as definitional limitations around what it means to practice medicine in America. And as mentioned earlier, whether the class of quality-­enhancing and protecting rules are appropriately calibrated or harmfully overbroad is the topic of considerable discussion. Before focusing on the specific regulatory structures that govern medical care in the United States, important themes that underlie America’s quality regulation scheme, such as patient protection, data gathering, standardizing medical care, and promoting patient trust, will be summarized. Once summarized, the discussion of America’s quality-­enhancing regulatory structure will begin by focusing on the regulation of practicing professional healthcare providers, followed by the regulation of American healthcare entities. The summaries will focus on different types of medical regulation seen in the American system, from self-­regulation, to state regulation, to so-­called soft regulation and consumer-­based tools. Needless to say, American regulation is multifaceted and multilayered. The second chapter focuses on a broad analysis of the European regulatory frameworks and some of the regulatory issues that have arisen in recent years. In part the growth (and revision) of regulation in healthcare, particularly regulation of professionals, has developed as a reaction to high-­profile, individual instances of misconduct and consequent public alarm leading to a perception of risk on which governments must act. Other regulatory frameworks have been introduced because of the influence of EU law in Member States, such as the regulation of clinical research trials and the quality and safety of human tissue and blood products. This chapter also looks at difficulties that have arisen because of professional mobility within the EU, the maintenance of independence by regulators, and the challenging issues that can arise in the imposition of sanctions such as withdrawal of licenses from healthcare facilities.

Chapter 14

R egu l ation of Professiona l s a n d Facilities i n Eu rope Deirdre Madden

1  History of Regulation of Healthcare Many aspects of healthcare systems and clinical practice in Europe are subject to legal or regulatory oversight to ensure that appropriate standards are maintained and that public safety is protected. However, the mechanisms by which this oversight is achieved vary between jurisdictions and have changed over time. The history of healthcare regulation in many European countries shows a shift in recent years from self-­regulation, which depends on public trust in healthcare providers and professionals to hold themselves and their peers to account for failings, to a demand from citizens and politicians for increased involvement of state actors and patient voices in regulatory bodies and calls for greater transparency and accountability in public life more generally. This is consistent with the shared understanding in most European jurisdictions of healthcare as a matter of public life and welfare rather than simply a private transactional relationship between patients and doctors. Therefore, as expectations about accountability in other areas of public life change, so do expectations in the context of healthcare. Traditionally, healthcare professionals enjoy a high status in communities around the world and are often unchallenged by a deferential patient population. In many jurisdictions in Europe (and indeed in the United States), the autonomous healthcare professional was subject only to his or her own conscience as guided by cultural and professional norms. Professional bodies were largely interested in education and training activities at their early stages but soon became active in seeking registration mechanisms to ensure that only those who had successfully navigated the rigours of medical schools were granted the privilege of the protected title of ‘doctor’. Historically other health professionals were commonly vocationally trained rather than acquiring university

278   Deirdre Madden qualifications and were often seen as having a lesser status, although this has now changed with the recognition of university or professional qualifications for a range of allied health professionals. Registration systems controlled by the professions themselves act as both an assurance of competence to the public and also as a means by which to control access to the profession. The success of this self-­regulatory framework for many decades relied on trust in the professions to ensure adequate educational criteria were met, appropriate training was provided, and that members behaved in a manner that was considered safe and professional. As will be seen later, where self-­regulation is perceived by the public to have failed to protect patients and users of social care services from harm, governments have changed the regulatory model to one involving greater participation of state-­ appointed experts from outside the relevant profession, representatives of patient populations, and accountability to government ministers to ensure appropriate corporate governance, financial integrity, and alignment with international best practice. This is often opposed by the professionals themselves who object to the loss of the privilege of self-­regulation and resist state interference in their profession. Regulation of healthcare institutions also has seen a shift from self-­regulation to government regulation, driven in large part by the organisation of healthcare systems. Historically, governments in Europe did not perceive that they had an obligation to provide healthcare to all citizens, and many governments only operated a system of access to doctors and infirmaries for the very poorest in society to try to contain the spread of infectious disease.1 Hospitals and other healthcare institutions were often owned and funded by charitable organisations, with the state paying healthcare providers to provide care to those who could not afford to pay for it themselves. As mentioned, the healthcare providers operated under a self-­regulatory model, and the government also relied on the mission of charitable organisations to protect patient welfare. Ironically, World War I had a heavy influence on the provision of healthcare in European countries, with governments becoming more involved in funding healthcare due to a drop-­off in charitable donations and setting standards for control of infectious diseases (such as tuberculosis), sanitation, and other public health issues. The end of World War II also saw a political push toward the establishment of a national health serv­ice in some countries in order to further the aims of postwar reconstruction, which were seen to require the elimination of disease, squalor, and poverty. As governments became more involved in operating or funding hospitals and other healthcare facilities, they assumed a greater role in regulating practices in those facilities. There were notable exceptions to this general trend. In countries such as Ireland, which had a strong Roman Catholic tradition and presence, there was fierce resistance to the establishment of a national health service due to fears that state involvement in health would curtail the power of the religious hierarchies to control matters relating to 1  Doctors who were contracted to provide such care to the poor remained free to charge fees to their wealthier patients. This led to the growth of two-­tier health systems which remain the basis of many European health systems today.

Regulation of Professionals and Facilities in Europe   279 family planning and reproductive health. This resulted in the State agreeing to fund voluntary organisations which were largely faith-­based in the provision of health and social care, but the state did not interfere with their ethos and values. This hands-­off approach made it more challenging in later years to attempt to regulate access to particular serv­ices which are often opposed on grounds of ethos. Overall, then, the development of state regulation of healthcare providers in more recent decades reflected the confluence of several factors. There was an increasing public expectation that if the state was paying for the provision of healthcare, it should be in a position to direct and monitor how that healthcare was delivered. In addition, there was a growing separation of church and state in many countries, a recognition of the need for standardisation of healthcare in the interests of patient safety, and a decrease in trust in self-­regulating professions and consumerism more generally. A further influence on the growth of state regulation was membership of the European Economic Community (EEC) and later the European Union (EU), which dictated Member State regulation of aspects of healthcare such as clinical research trials2 and the quality and safety of human tissue and blood products.3 In recent years there has been an increased focus on what kind of regulation is appropriate and necessary in terms of protecting the public from harm but also in ensuring that regulatory frameworks do not impose too high a burden through compliance costs, negative perception of the burden of bureaucracy, or disproportionate restriction of practice. Whereas light regulation might be ineffective and seen as an abdication of responsibility by governments, heavy regulation might be seen as excessive and as an interference with liberty. In attempting to strike the right balance (so-­called ‘right-­touch regulation’), governments have tried in various areas of healthcare to make regulation more straightforward and to simplify the compliance process. This has involved using information technology solutions for self-­certification, providing effective guidance and support to comply, analysing the potential impact of regulation in advance of its introduction, engaging in public consultation about regulation, and assuring compliance through random audit rather than mandatory inspections. Right-­touch regulation is best seen as the minimum regulatory force required to achieve the desired result. It means fully understanding the problem before arriving at a solution and ensuring that the level of regulation is commensurate with the level of risk to the public. Right-­touch regulation should therefore be proportionate, consistent, targeted, transparent, accountable, and agile.4 Healthcare regulators (both professional and systems-­based) aim to provide protection against harm, not elimination of risk; 2  Regulation (EU) No 536/2014 of the European Parliament and of the Council of 16 April 2014 on clinical trials on medicinal products. 3  See https://ec.europa.eu/health/blood_tissues_organs/blood_en for further details of EU Directives and Regulations in this area. 4  Professional Standards Authority, Right-­touch regulation (2015), http://www. professionalstandards.org.uk/what-­we-­do/improving-­regulation/right-­touch-­regulation

280   Deirdre Madden oversight of regulated activities; and monitoring compliance with rules. They help the public make informed choices about service providers and provide expert and in­de­pend­ent feedback to government on the safety or otherwise of the system. Although the concept of right-­touch regulation applies to all sectors and all regulators, it can be more challenging to put into operation in a health system which is underfunded and has a mix of health facilities—state-­owned, state-­funded, private, voluntary, and charitable. In particular, demands for financial oversight and governance of non-­state facilities tend to be more onerous than those under direct control of the state. The use of public revenue to fund non-­state health facilities is sometimes viewed  with suspicion, resulting in heavier reporting requirements to ensure state oversight of compliance with standards and micro-­management by state employees of operational matters within these facilities, contrary to the concept of ‘right-­touch’ regulation.

2  Regulation of Healthcare Professionals 2.1  Purpose and Role of Professional Regulation Regulation of healthcare professionals aims to serve a number of important functions. First, it seeks to protect and promote patient safety through a model of registration or licensure that ensures that those who practise in any area of healthcare have (and maintain throughout their careers) the necessary qualifications and competencies to do so. By providing public registers of qualified practitioners which are freely and easily accessible, states try to ensure that potential patients seek diagnosis, advice, and treatment only from those who have satisfied a regulatory authority that they have met minimum educational qualifications and are fit to practise that profession. Most regulatory systems also incorporate systems of recognition of foreign qualifications and criminal ­penalties for those who practise while unregistered. These penalties are important in ensuring that maverick practitioners do not take advantage of potentially vulnerable and desperate patients who may be persuaded to seek unconventional treatments without fully comprehending the professional qualifications (or lack thereof) of the person who is purporting to provide them with treatment. This satisfies one of the goals of the global patient safety movement, which is to ensure that healthcare is provided only by appropriately qualified practitioners. Professional regulatory models seek to provide assurance to the public that there is oversight of the educational standards to be achieved before registration and before medical practice is permitted, that there is appropriate investigation of complaints about practitioners allegedly falling below professional standards, and imposition of appropriate penalties in order to maintain confidence in the professions.

Regulation of Professionals and Facilities in Europe   281 Oversight of educational standards is commonly exercised in accordance with world standards, such as those of the World Federation of Medical Education,5 in order to ensure that national standards strive for international excellence and comparability across jurisdictional boundaries and to ensure that mobility of professionals across ­borders does not jeopardise the standard of patient care. Once registered or licensed, practitioners are subject to professional standards of practice set by the regulator, against which a practitioner’s professional competence or behaviour will be measured. Although regulators in many jurisdictions still tend to be comprised of professionals, this is not the case in all jurisdictions. For example, in Ireland, the membership of some professional regulators is a majority of persons who are not members of the relevant profession. The standards are commonly developed across a range of professional competencies and professional conduct or ethical behaviour. Failures by practitioners that may lead to complaints being investigated can include allegations of a serious falling below standards of competence, ethics, or both. Strict legal mechanisms are put in place for the investigation of such complaints in order to ensure fairness to the professional against whom the allegations are made and to ensure that patient safety is paramount. Following due process, if complaints are upheld against a practitioner, sanctions such as censure, suspension, and, ultimately, removal of a practitioner’s name from the register may be imposed. These systems operate to ensure the integrity of the register of professionals, providing public assurance that only those practitioners who are fit to practise may maintain their names on such registers. Professional healthcare regulatory authorities in Europe are not consistent in their structure and approach, having evolved at different times and in different ways. There is considerable diversity in the range of topics that regulatory bodies oversee, with almost all covering healthcare quality and safety and others encompassing reputation, respect, and trust.6 In most European countries such as Ireland and the United Kingdom, there is a separate regulator for doctors. In others, such as Sweden and Estonia, a single organisation regulates all health professionals. In some countries, there are separate regulators for nurses and midwives, while in others, such as Ireland, they are combined.7 These inconsistencies have significant implications for professional mobility, patient safety, and quality of care. The question of independence of function on the part of professional regulatory authorities is a vexed one. Professionals generally guard their independence very jealously and resist any attempt by governments to impose their political will through accountability and reporting frameworks, provisions in legislation enabling the relevant Minister/Secretary of State to impose or demand certain courses of action on the part of 5 http://wfme.org/ 6  Risso-­Gill I, Legido-­Quigley H, Panteli D, and McKee M, ‘Assessing the Role of Regulatory Bodies in Managing Health Professional Issues and Errors in Europe’ (2014) 26(4) Int’l J. Qual. Health Care 348–357. 7  McKee M, ‘Healthcare Professionals Crossing Borders,’ Summer Edition Issue 39, https://www. hpcb.eu/Issue_39___HPCB__Update_Briefing.pdf_71213377.pdf

282   Deirdre Madden the regulator, or potential interference for political purposes. Where professionals associate together for purposes of setting educational and professional standards on a voluntary basis, this can be very productive and useful in achieving consensus. However, if these professional-­led organisations do not have statutory powers and functions they will be missing the teeth of enforcement which may be necessary to deal with outliers or those who opt out of non-­statutory schemes. In some countries, statutory regulation of health professionals has a long and not always happy history, with professionals complaining about interference with clinical autonomy for political purposes, the appointment of nonprofessional/lay members to the boards of regulatory authorities to assuage the public demand for transparency and accountability, and the cost of regulation, which in most cases is borne by the profession itself with individuals paying on an annual basis for the retention of their name on the public register. There have been long-­standing debates in many countries about whether self or state regulation of healthcare professionals is more effective in securing the aims just described. Proponents of self-­regulation generally argue that it is justified where the practitioners possess an unusual level of skill and knowledge, the trustworthiness of the practitioners means they can work without supervision, and, where there are issues of incompetence or unethical conduct, the profession can undertake the necessary disciplinary action itself.8 Self-­regulation also offers the advantages of independence and clinical freedom, bringing insider knowledge to bear on a problem, and is more acceptable to those being regulated, thereby encouraging compliance.9 It is also said to be more responsive because self-­regulating bodies can move faster without the need to change legislation, and it is cheaper because it requires less monitoring.10 However, for self-­ regulation to be effective the public must be able to trust the profession. Where this trust has been shaken by controversies and scandals, this can lead to public perception that the profession is failing to self-­regulate effectively in order to prevent harm to patients. Such events in the past have led to accusations that self-­regulatory bodies lack legitimacy, engage in protectionism of the profession, and fail to provide public accountability.

2.2  Move from Self-­Regulation By comparison to the growth of regulation of healthcare facilities, state regulation of professionals has sometimes developed as a reaction to high-­profile instances of misconduct or adverse events in healthcare and consequent public alarm leading to a perception of risk on which governments have to act. Examples include the investigation into paediatric cardiac surgery in Bristol, UK; the retention of organs in postmortem

8  Friedson E, Profession of Medicine (University of Chicago Press 1988). 9  Madden D, Medicine, Ethics and the Law, 3rd ed. (Bloomsbury 2016) para 2.40. 10  Irvine D, ‘The Performance of Doctors: Professionalism and Self-­Regulation in a Changing World’ (1997) 314 BMJ 1540.

Regulation of Professionals and Facilities in Europe   283 practice in several countries11; and the practice of Dr Harold Shipman in the UK.12 When there is a crisis or controversy of this kind, there is often a shared demand in the public domain that ‘the government must do something’ to ensure patient safety. This leads to a reactive political response which can often be restrictive or punitive or, more rarely, reforming, to ‘ensure that this never happens again’. An example of this can be seen in the public and political concern which emanated from a State Inquiry in Ireland into the practice of a consultant obstetrician who had carried out a very high number of peripartum hysterectomies, seemingly unchecked by his peers, hospital management, or his professional regulatory body, the Medical Council, for many years. Disquiet grew about the perceived failure of the Medical Council, which at the time was a self-­regulatory body, to act and be accountable to the public for whose safety it was established and designed to protect. This was in part assumed, perhaps unfairly, to be due to its professional majority membership and consequent lack of accountability to the public. The Report of the Inquiry, which was published in 2006,13 acted as a catalyst to the introduction of the Medical Practitioners Act 2007, which fundamentally altered the composition of the Medical Council from a body with a majority of medical practitioners to a body with a majority of nonpractitioners or lay people. The government took the view that it was important for public confidence that members of a profession themselves do not have majority control in the governance of their regulatory body. Despite opposition from within the profession, this principle was later extended to other regulatory bodies in the country with the aim of ensuring effective regulation of all health and social care professions. It is difficult to say whether this move away from self-­regulation has been successful in enabling the Medical Council to better fulfil its statutory mandate of protecting the public. Would success be measured by an increase or a decrease in the level of complaints upheld against doctors? The number of complaints against doctors in Ireland rose from 343 in 2007 to 411 in 2016,14 but this is due to many factors and cannot simply be regarded as having been caused by a change in the membership of the Medical Council. An increase in the number of complaints might be seen as an indication of the wider reach of a lay-­majority council, an increased effort to make the public aware of the functions and remit of the Council, or a combination of these and other factors. However, simply looking at the number of complaints which have been upheld in that period would not provide an answer to this question as a preliminary screening committee of the Council, which decides whether a prima facie case exists against a doctor 11  The organs were retained for diagnostic purposes or for education or research purposes rather than being left in the cadaver for burial. See Royal Liverpool Children’s Inquiry Report (HC 12, 2001) in the UK and the Madden Report into Post-­mortem Practice in Ireland (DoHC 2005). 12  Harold Shipman was an English doctor who was found guilty of murdering fifteen of his patients. This led to the Shipman Inquiry and, in particular, the 5th Report of Dame Janet Smith entitled ‘Safeguarding Patients: Lessons from the Past—Proposals for the Future’ (2004) (Cm 6394). 13  The Lourdes Hospital Inquiry—An Inquiry into Peripartum Hysterectomy at Our Lady of Lourdes Hospital, Drogheda (DoHC 2006). 14 www.medicalcouncil.ie

284   Deirdre Madden complained against, is still comprised of a medical majority. An increase in the number of doctors sanctioned by the Medical Council might be a reflection of a lower tolerance amongst lay people for failures of competence, poor behaviour by doctors, or an illustration of a push for higher standards amongst professionals generally. Perhaps the symbolism of a lay majority and the promise of greater public involvement and accountability is itself an adequate reason for change, but it is difficult to provide an evidence base to say whether or not it has had an impact in terms of patient safety or protection of the public interest.

2.3  State Regulation As outlined earlier, in recent years states have increasingly become involved in the ­regulation of healthcare professionals by introducing legislation to ensure standardisation of educational qualifications, oversight of registration requirements, establishment of regulatory bodies to develop codes of conduct, and provision of legal mechanisms to inquire into the behaviour and competence of those against whom complaints are made. These measures, although sometimes resisted as state interference by the regulated professions, are designed to ensure patient safety and public accountability through easily accessible registers of professionals, public disciplinary hearings of complaints, and appointment of board members through public appointment mechanisms. State regulation of professionals is not without its flaws as there is an associated high cost of establishing and maintaining such complex mechanisms which either the state must fund or are paid for by registration fees collected from practitioners. The mandatory payment of fees by professionals to a body which subjects them to strict regulatory controls legislated for by the state is often open to criticism within the profession, although the control exerted by the regulatory body over registration requirements often also brings a welcome limitation on access to and competition within the profession. A further complication in the European context arises out of membership of the EU. Although health professionals are subject to state regulation in many European jurisdictions, membership of the EU has brought about certain challenges to the ability of states to control access to employment and qualifications within their own borders. A few decades ago, a qualified professional in one EU country would not necessarily have met the requirements to practise in another EU country unless he or she had completed an entire training course in the latter (host Member State). In order to comply with one of the basic tenets of EU law (i.e., free movement of people) and to thus ensure free movement of doctors within the EU, Member States agreed from an early stage on common standards for basic and specialist medical training. Subsequently, similar arrangements were created for other health professionals. European rules on mutual recognition of qualifications were later consolidated into the Professional Qualifications Directive,15 15  Directive 2005/36/EC of the European Parliament and of the Council of 7 September 2005 on the recognition of professional qualifications. The Directive was later reformed and modernised to reaffirm

Regulation of Professionals and Facilities in Europe   285 which provides for automatic recognition for a limited number of professions based on harmonised minimum training requirements (sectoral professions), a general system for the recognition of evidence of training, and automatic recognition of professional experience. On the basis of this Directive, if a person qualifies in one Member State and wishes to access employment on a permanent basis in another Member State, he or she will need full recognition of her or his qualifications by the regulator in the host state in order to be able to access the profession and practise it on the same terms as nationals of, and those who qualified in, that country. Professional mobility is common amongst healthcare professionals, especially of doctors and nurses, the two most mobile professions in the EU based on the number of decisions taken on recognition of professional qualifications for the purpose of permanent establishment within the EU Member States, European Economic Area countries, and Switzerland. In the EU, there is a system of automatic recognition of professional qualifications for five health professions: nurses, midwives, doctors, dentists, and pharmacists who meet specified criteria in relation to their qualifications. Some states, such as Ireland, have interpreted the legal provisions to mean that the regulator cannot impose an English-­language requirement on doctors who qualified within the EU as this would constitute discrimination against other EU nationals. Employers are however at liberty to impose such language proficiency tests. For those who are not eligible for automatic recognition, their qualifications may still be recognised in another EU country under the general system for recognition of qualifications. In such cases Member States proceed on a case-­by-­case basis and have a fair bit of discretion in terms of granting access to a given profession. If there are major differences between the qualifications, applicants may be asked to sit an aptitude test or complete an adaptation period16 after which they should be granted full access to the profession in that state. However, although in keeping with fundamental principles of the Single Market and facilitating professionals to travel, work, and live in any EU Member State, the downside of this positive and attractive feature is that mobility of health professionals changes the composition of the health workforce in source and destination countries and may aggravate or mitigate existing problems such as shortages, maldistribution, and skill-­ mismatches of health professionals.17 Some Member States rely to a large extent on the underlying philosophy of mutual recognition and mutual trust between Member States whilst exploring innovative ways to better reflect it in practice. These new features include the introduction of a European professional card to offer interested professionals the possibility to benefit from easier and quicker recognition of their qualifications, modernisation of harmonised minimum training requirements, and an alert mechanism set up for all professions with patient safety implicationsm which effectively introduces an obligation for competent authorities of a Member State to inform the competent authorities of all other Member States about a professional who has been prohibited, even temporarily, from exercising his professional activity or who made use of falsified documents. 16  Directive 2005/36/EC of the European Parliament and of the Council of 7 September 2005 on the recognition of professional qualifications (as amended), http://eur-­lex.europa.eu/legal-­content/EN/ TXT/?uri=CELEX:02005L0036-­20160524 17  Glinos IA et al., ‘How Can Countries Address the Efficiency and Equity Implications of Health Professional Mobility in Europe? Adapting Policies in the Context of the WHO Code of Practice and

286   Deirdre Madden foreign health professionals, while others experience important outflows. An example of dependency on overseas graduates in the UK can be seen in the debate that ensued following the case in which a British patient died after he was accidentally administered a lethal dose of diamorphine by Daniel Ubani, a Nigerian-­born doctor based in Germany.18 These trends can change over time depending on economic climates in individual states and perceptions of the health systems in different countries so ‘countries are faced with the constantly changing conundrum of efficiency and equity, that is, between the free mobility of health professionals in the European labour market on one hand, and the planning requirements of health systems.’19 Most, if not all countries experience both inflows and outflows; Italy, for example, experiences outflows of medical doctors but inflows of nurses, while in Ireland inflows of foreign-­trained doctors replace outflows of domestically trained doctors.20 However, EU legislation is, at best, indifferent with regards to health systems in source and destination countries as it focuses on the individual (health) workers and their rights in the context of their freedom to move and work in other EU Member States. Challenging issues also arise in the context of imposition of disciplinary sanctions. Within the EU there have been some high-­profile cases of individual practitioners who had been subject to disciplinary sanctions in one Member State going to work in another Member State without much difficulty. Once a practitioner has been registered to practise in any state, that practitioner can usually only be removed from the register following investigation of a complaint and a legal process that may be vigorously defended. Pending the making of a complaint and its investigation, the practitioner is free to practise in any state in which he or she is registered. This clearly leads to potential risks to patients and the public. This risk led to the introduction of a proactive alert mechanism within the EU, whereby the regulator in each Member State notifies those in all other Member States about individuals whose fitness to practice is in question. This provides a clear example of countries reaching consensus on a way forward to protect patient safety. However, a recent survey found that some professional regulators have not registered with this system and that there is considerable variation among those registered in what would trigger an alert to be sent or how they would respond to an alert from another regulator.21 So, in theory, regulators see and act on a common purpose but differences in legal regimes, accountability standards, complaint mechanisms, standards of proof, availability of sanctions, and so on make the sharing of information on a common understanding challenging. EU Freedom of Movement’, http://www.euro.who.int/__data/assets/pdf_file/0008/287666/OBS_PB18_ How-­can-­countries-­address-­the-­efficiency-­and-­equity-­implications-­of-­health-­professional-­mobility-­ in-­Europe.pdf?ua=1 18  See discussion in Simpson J and Esmail A ‘The UK's Dysfunctional Relationship with Medical Migrants: The Daniel Ubani Case and Reform of Out-­of-­Hours Services’ (2011) 61(584) Br J Gen Pract 208–211. 19  Glinos, ‘How Can Countries Address’, section 3. 20  Glinos, ‘How Can Countries Address’, section 2.1. 21  McKee, ‘Healthcare Professionals Crossing Borders’.

Regulation of Professionals and Facilities in Europe   287 From the perspective of patients and members of the public, regulation of professionals can commonly be an alien and somewhat intimidating mechanism. Regulators need to work on communicating their role and statutory functions to the general public irrespective of whether the regulator is self-­regulating or not.22 Regulators exist in a different space to medical malpractice litigation in that they do not offer financial compensation to those who have been harmed by a health professional but rather focus on the broader issue of trying to ensure that all those on the register of practitioners are fit to practise and imposing sanctions on those who fall short of standards in order not to punish but to ensure protection of the public. There is often a high degree of misunderstanding amongst the general public about the role and function of professional regulators; many complaints to regulators show that members of the public perceive the regulator to offer a system of individual conflict resolution as a means by which their complaint will be investigated and an apology provided. Although some complainants state that their motivation in making a complaint is to ensure that the event will not happen again, in general there is little understanding of the statutory function of the regulator, which is to investigate complaints and promote the highest professional standards in order to protect the public interest. A further challenge exists in the way that complaints are dealt with in regulatory systems. Some complaints could be better resolved at the local level rather than through investigation by a regulator as this would avoid high financial and emotional costs for both parties and may serve to help repair the damaged relationship between professional and patient in a way that a statutory investigation, with all its legality and formality, cannot. Complainants are generally more likely to proceed to making a complaint if they feel disrespected or ignored by the professional, and therefore professionals should be advised that it is in their interests as well as the interests of their patients to respond quickly and effectively to any concerns or complaints made. There is of course a balance to be struck here between these complaints that may be resolved effectively and quickly at a local level and those that may raise significant issues relating to the fitness to practise or professional competence and ought to be dealt with by the regulator to ensure the safety of other patients and the public. This is an ongoing challenge for regulators who must have clear and effective legal processes in place to both protect the public and also uphold the right of professionals to their good name and integrity. Holding statutory investigations into every complaint is neither feasible nor desirable as there may often be complaints that are frivolous or vexatious and yet the regulator must reassure the public that complainants are taken seriously—this can be a challenging balance to strike, and professional regulatory bodies often bear the brunt of criticism from the public and media who fail to appreciate the difficulties involved.

22  Madden D and O’Donovan O, ‘Qualitative Review of Complaints to the Medical Council’ (2016), https://www.medicalcouncil.ie/News-­and-­Publications/Publications/Listening-­to-­Complaints-­ Qualitative-­Report.pdf

288   Deirdre Madden

3  Regulation of Healthcare Facilities 3.1  Regulating Mixed Healthcare Systems Regulation of institutions where healthcare is provided can be challenging on a number of levels. First, in many countries there is a mix of public and private providers and often, for historical reasons, a third sector of voluntary or not-­for-­profit providers which are funded in whole or in part by the state to provide health and social care services. The latter make a significant contribution to health and social care in many European countries, having arisen primarily from the historical failure or inability of states to provide adequate care for their sick and poor before statutory healthcare systems existed. They are generally strongly motivated organisations focussed on their founding values, closely engaged with their local communities, powerful advocates for patients and users of social care services, and can often be more responsive and flexible in relation to service reform and innovation than the public sector. The differences among these three sectors can result in difficulties in applying a ‘one size fits all’ regulatory model as their distinct models can sometimes complicate the oversight of corporate governance, funding, and compliance with national standards that necessarily fall within a regulatory framework. Although regulatory frameworks should apply equally to all hospitals or healthcare institutions in terms of safety and quality of care, there is not always a ‘closing of the loop’ between the standards imposed by a state regulator and the funding of the steps that need to be taken in order to ensure compliance with those standards. Where compliance with national standards is required for accreditation or licensure, public and voluntary providers may struggle more with compliance due to lack of resources as compared to private providers that may be able to raise funds more easily through fee-­for-­service payments, as well as through investors and shareholders. If public providers are unable to maintain compliance, the state then often has no option but to purchase care from private providers, often ultimately at a higher cost to the taxpayer. This issue can be a source of significant complaint and dissatisfaction amongst service providers and is ultimately best resolved by a meeting of minds between what are essentially two arms of the state: the regulator and the funder. The regulation of voluntary or not-­for-­profit providers, particularly religious and faith-­based organisations, also poses ethical challenges as there are legitimate questions about the extent to which the state, which funds the majority of these providers, can and should regulate such providers to ensure that they provide all services lawfully available even where some services might conflict with the ethos of the organisation. The conflict between religious freedom and state regulation of care can be particularly problematic in the context of reproductive healthcare, as demonstrated by an example from Ireland which has had a long and difficult history of attempting to legislate in the area of reproduction. The Irish government’s decision to build a National Maternity Hospital was welcomed by many clinicians as a much needed development in maternal care. However

Regulation of Professionals and Facilities in Europe   289 the location of this hospital on land owned by a Catholic religious order also caused public controversy due to the backdrop of historic claims of clerical child abuse and attempts to liberalise the very strict constitutional regime in Ireland for the termination of pregnancy.23 Questions arose as to whether the full range of modern reproductive healthcare services would be available in this hospital if the religious order were to seek to influence board decisions about service planning and provision of assisted reproduction, genetic testing, termination of pregnancy, and other services that conflict with the teaching of the Catholic Church.24 Although public disquiet on this particular issue was dampened by an announcement from the religious order that it would not have any involvement in the ownership or management of the new maternity hospital,25 the Minister for Health subsequently established an independent review group to examine the role of voluntary organisations, particularly religious and faith-­based organisations, in health and social care.26 This example illustrates the difficulties that can arise in a broader context about the extent to which, if any, public/state funding of healthcare institutions empowers state agencies to require those institutions to provide a full range of services and to secularise their mission statements and values in order to remove overt affiliation to any particular religious ethos. This issue seems to be less problematic in other (larger) jurisdictions where freedom of choice for patients who wish to be treated either by religious or nonreligious providers is possible due to a greater number of service providers and where commissioning models of healthcare based on activity-­based funding appear to have taken the heat out of this debate to some extent because the state is not seen as funding the institution per se but rather the activities or services it provides.

3.2  Independence of Function and Imposition of Sanctions for Noncompliance The importance of independence in regulatory functions cannot be underestimated. Although there often is and must be a triangular working and well-­functioning relationship among the healthcare funder, provider, and regulator, the regulator must maintain distance and objectivity from both funder and provider in order to assure the public of the independence with which it carries out its statutory mandate of protecting the public. This can sometimes prove to be a difficult tightrope to walk as the regulator must 23  http://www.thejournal.ie/religion-­health-­care-­catholic-­church-­3360849-­Apr2017/ 24  In light of the change to the constitutional position on abortion in Ireland in May 2018, it will be interesting to see in coming years if this repudiation of conservative and religious values evolves into other areas of healthcare traditionally influenced or sometimes dominated by faith-­based perspectives. 25  http://www.thejournal.ie/sisters-­of-­charity-­st-­vincents-­maternity-­hospital-­3414044-­May2017/ 26  Report of the Independent Review Group established to examine the role of voluntary organisations in publicly funded health and personal social services (Dept of Health, Dublin, February 2019), https://health.gov.ie/blog/publications/independent-­review-­group/

290   Deirdre Madden work with the political system, advocacy bodies, and members of healthcare professions to secure engagement on development of appropriate standards and buy-­in from those expected to operate the standards, and yet the regulator is also the enforcer when those standards are not met. One of the major difficulties in regulation of hospitals and other healthcare facilities lies in the imposition of sanctions for noncompliance. In other sectors, sanctions including fines, warnings, improvement notices, and, ultimately, withdrawal of licences may be contemplated, but in healthcare it is extremely problematic for legal, social, political, and ethical reasons to close a hospital or residential care setting through withdrawal of its license. The state remains obligated to those patients or residents who continue to need care, but the public sector may not have bed capacity, and relocation of long-­stay residents in nursing homes may be extremely disruptive and traumatic to those individuals, their families, and carers. In such extreme cases, again the state may have no option but to purchase care from private providers, sometimes at higher cost to the state and, ultimately, taxpayers. An example of the complexities involved in navigating this relationship is provided by the Health Information and Quality Authority (HIQA) in Ireland. HIQA is an in­de­ pend­ent authority established to drive high-­quality and safe care in health and social care services. HIQA has a broad remit across standard setting, regulation, monitoring of children’s social services, monitoring healthcare safety and quality, health technology assessment, and health information, but, in the context of regulation of facilities, its main role is to develop standards, inspect and review services, and support informed decisions on how services are delivered. HIQA’s licensing (known as registration) mandate is currently confined to nursing homes for older and dependent people and to residential services for children and adults with disabilities. The relationships at play here are among the political arm, the funding agency, and the watchdog: the Department of Health sets policy for the healthcare system, the Health Service Executive funds the public health system, and HIQA acts as the standard-­setting agency and enforcer. HIQA was introduced into the health system at a difficult economic time in Ireland, and many smaller centres were vocal in their dissatisfaction at being criticised for not meeting standards at a time when funding to achieve compliance with those standards was not available from the state. The reality is that HIQA cannot excuse noncompliance or allow centres to offer excuses on financial grounds when there are issues of resident safety and dignity at stake. However, it undoubtedly causes tensions between stakeholders that can become public and political controversies. A further example of the difficulty in regulating facilities relates to systems in which there are a mix of providers—public, private, voluntary, charitable, community, and so on. Again, to take the example of HIQA, it currently has a limited remit to inspect and publish reports into public acute hospitals’ compliance with specific national standards, such as those for the prevention of healthcare-­associated infections. However, unlike in the area of residential social care services, outside of those specific national standards, HIQA does not have the authority to regulate acute general hospital services and has no enforcement powers. Although there is some traction to be gained from the public and

Regulation of Professionals and Facilities in Europe   291 media attention that a negative report from HIQA receives, it is nonetheless open to question whether a regulator can be effective without teeth. In the absence of licensing in Ireland (which is currently the subject of proposed legislation), private hospitals and clinics are not subject to any state regulation, but they do usually have to comply with very exacting standards in order to continue to receive funding from private health insurers, and therefore they apply for accreditation from international organisations such as Joint Commission International (JCI).27 It is expected that licensing of healthcare facilities will be introduced in 2021/22 but careful thought will have to be given to many difficult issues such as whether a license will be given to a facility, service, hospital group, or nominated responsible person and the consequences of noncompliance with national standards by an acute hospital. If, after repeated warnings and due process, the hospital remains noncompliant, the question remains as to whether a facility or hospital will actually lose its license to offer services and what the implications of this might be for local communities, patients, and the public. These are serious political and social issues to be grappled with by any regulatory system and are issues that do not come without financial cost.

4 Conclusion Regulation of healthcare is necessary to provide a safe, high-­quality health system, and regulators should always aim to comply with modern expectations of independence, accountability, conformity with international best practice, and representation of the patient/public interest. However, as this chapter has demonstrated, regulatory mechanisms are not straightforward given the complexity of healthcare systems, the balancing of interests involved, the need to avoid political agendas taking priority over the public interest, and the financial and bureaucratic cost of compliance. Right-­touch regulation in all sectors tries to strike an appropriate and proportionate balance between liberty and safety, between freedom to innovate and public protection, but the stakes are perhaps higher in healthcare than in other areas due to the potential consequences involved. The increasing level of participation of public interest or lay representatives on regulatory boards and statutory authorities may go some way to providing the public with assurance of the independence of these bodies and their prioritisation of patient safety above all else, but much also depends on the statutory framework underpinning the work of these bodies and the extent to which they are mandated to be rigorous, in­de­ pend­ent, effective champions of the public interest. Much work has been done in this context, but there remains much more to do. The change in the relationship between professions and governments is also mirrored by changes in the relationship between patients and healthcare professionals. In recent years this relationship has both benefitted from and been challenged by increasing levels 27 https://www.jointcommissioninternational.org/

292   Deirdre Madden of awareness, education, access to information, and greater empowerment on the part of patients, leading them to be more inquisitive about their care and wanting to be more involved in decision-­making. These changes have not only had an impact on the practice of professionals but have also led to increased public expectation that those who practise any of the health professions must be subject to a system of regulation by which their competence can be measured and assured. There is an open question about the extent to which the shifting of power from the profession to the state (at least in some respects) has served the patient safety and quality agenda well or whether it is simply part of an overarching strategy of transparency and accountability which many states have been moving to in order to satiate public demand.

chapter 15

R egu l ation of Professiona l s a n d Facilities i n the U n ited States Isaac D. Buck

1 Introduction This chapter examines the regulation of healthcare professionals and facilities within the United States. In the midst of an era of great policy uncertainty, healthcare in the United States is regulated and shaped by a number of actors and rules. This chapter seeks to summarize the types of regulation that govern American healthcare professionals and institutions while providing an overview for comparison with European counterparts. Differing from a medical malpractice regime that compensates injured patients following a violation of the standard of care or a fraud and abuse regime that focuses on financial harms, quality regulation seeks to prevent harm to patient health from occurring in the first place. Due to the nature of quality control regulation which focuses on preventing harm before it occurs, its broad, prophylactic rules apply universally throughout the industry. The rules govern the entirety of medical practice: market entry, licensing and accreditation, surveys and review, and sanction and discipline, as well as definitional limitations around what it means to practice medicine in America. And whether the class of quality-­ enhancing and protecting rules is appropriately calibrated or harmfully overbroad is a topic of considerable discussion. Before focusing on the specific regulatory structures that govern medical care in the United States, important themes that underlie America’s quality regulation scheme will be presented. Following that, a summary of America’s quality-­enhancing regulatory

294   Isaac D. Buck structure will begin by focusing on the regulation of practicing healthcare professionals, followed by the regulation of American healthcare entities.

2 Themes Needless to say, quality regulation of healthcare professionals and entities is a broad subject. The rules range from licensing and discipline of professionals by the professions themselves to the corporate practice of medicine (CPOM) doctrine preventing the direct employment of doctors in some states. Even though the maze of quality regulations is complex, important themes and values underlie its landscape. In addition to the obvious goal of a quality-­protecting system focused on protecting patient welfare, the system seeks to both standardize and improve medical practice, with the ultimate goal of fostering and cultivating patient trust. Initially, quality regulation in the United States serves an important data-­gathering function. By gathering information on how professionals and facilities administer healthcare, the regulatory system can collect, surface, and encourage the adoption of best practices, thereby addressing a serious gap in medical knowledge. For decades, American healthcare had little understanding of its most pressing quality challenges. A groundbreaking 1999 Institute of Medicine report, To Err Is Human, finally began to characterize the scope and type of quality challenges facing American healthcare.1 The main conclusion from the report highlighted the fact that between 44,000 and 98,000 American patients died annually due to medical errors.2 Besides culminating in a comprehensive report that identified a major challenge in American healthcare, the report also served as a foundational moment in American medicine: professionals and patients were finally able to access data that described a bleak picture of American healthcare quality. Beyond its information-­gathering function, America’s quality regime seeks to use the information to standardize medical practice in an effort to address medical heterogeneity. Quality regulation seeks to instantiate a “best practice” or accepted medical stand­ ard; without such a standard, medicine becomes a more fragmented, idiosyncratic, and—ultimately—conflicted enterprise. Without standardization among professionals and facilities, the quality that an American patient experiences could be up to the whim of which individual professional administers care to that patient. In a clearly unsettling way, patient experiences can vary in such a system—with potentially disastrous results. Ultimately, the regulatory regime’s goals of information-­gathering and standardization can build and protect patient trust. Without a coordinated rule-­based enforcement system, patients may lack trust in the administration and delivery of American 1  Institute of Medicine, To Err Is Human, Nov. 1999, http://www.nationalacademies.org/hmd/~/ media/Files/Report%20Files/1999/To-­Err-­is-­Human/To%20Err%20is%20Human%201999%20%20 report%20brief.pdf. 2 Id.

Regulation of Professionals and Facilities in the United States   295 ­ ealthcare. And, indeed, patient trust is vital to the doctor–patient relationship. In sum, h quality regulation gives patients the comfort to trust their doctor’s recommendations, with positive impacts on the therapeutic relationship and treatment outcomes.

3  The Regulation of Healthcare Professionals The regulation of a professional’s practice of medicine within the United States relies on a multilayered mix of laws, standards, and incentives that all seek to ensure the quality of American healthcare delivery and administration. The norms and standards making up this regulatory regime are both voluntary and involuntary, reliant on both penalties and incentives, and enforced by government agencies, quasi-­government actors, and the profession itself. Regarding government enforcement, it, too, is multilayered: for instance, professionals are both regulated and influenced by the state in which each professional practices, as well as by the federal government—most notably through the US Department of Health and Human Services (HHS) and Centers for Medicare and Medicaid Services (CMS). Differing from America’s patient-­compensating medical malpractice and corporate negligence regimes or its taxpayer-­protective fraud and abuse enforcement system, laws and rules regulating the practice of medicine largely seek to prevent patient harm before it happens. When working well, America’s medical regulatory regime achieves its primary goals of (1) protecting public health and welfare and (2) building and maintaining a culture of trust between professionals and their patients. Given the heterogeneity within medical practice, the increasing number of specialties and subspecialties in the medical community, and the geographic size of the United States, it is unremarkable to observe that striking the appropriate balance of enforcement and discretion is difficult. Indeed, a number of persistent challenges complicate the regime. These include whether, and to what extent, regulation has struck the appropriate balance in protecting physicians’ ability to self-­regulate or whether a self-­regulatory regime is simply an anti-­ competitive tool, negatively impacting the costs of care; how other professionals, such as nurse practitioners, fit into the regulatory regime given America’s gaps in access to healthcare; and, on a foundational level, how enforceable medical standards are reached in a country long characterized by heterogeneous medical practice. There also are challenges from a number of contemporary developments that impact American medicine. These include new challenges brought on by a more mobile population, a national primary care shortage and rural healthcare access crisis, the move away from fee-­for-­service reimbursement and toward value-­based reimbursement, and an increased reliance on consumer-­based tools. Before the challenges are presented, the regulation of American professionals is summarized—from self-­regulation, to state regulation, to reimbursement-­focused and incentive-­based regulation. These topics, which can holistically be described as quality regulation, will be the focus of this chapter. Notably, while the regulatory impact of

296   Isaac D. Buck medical malpractice litigation—and, specifically, its quality-­protecting effects—cannot be dismissed, it is not the subject of focus of this chapter.

3.1  Self-­Regulation Within the Medical Professions Probably the most visible type of professional regulation is the enforcement of quality-­ based standards that are promulgated by the profession itself. Through the police power vested in each American state, all 50 states (as well as US territories) have empowered professionally run licensing boards to regulate the practice of various professionals within their borders. For example, state medical boards regulate the practice of medicine, boards of nursing regulate the practice of nursing, and boards of dentistry regulate the practice of dentistry. While the boards include lay members, professionals predominate.3 These professional boards are responsible for two main regulatory actions: licensing and discipline. Through licensure, which includes re-­registration and re-­licensure, each state empowers the professions to be responsible for admission into their ranks. Standard admission requirements typically include attainment of various educational degrees, completion of fellowship or residency training, and successful passage of comprehensive examinations. Control over admission gives the various boards substantial power over the future and makeup of the profession. And the professions use their licensing power to employ market-­restricting techniques, thus limiting the number of individuals who can practice within the state’s borders. Boards also are responsible for imposing disciplinary sanctions against those who violate various state statutes that govern their practice. Once a violation is alleged— which can be reported by patients—the alleged violation is investigated by the boards, and the board makes a decision. For serious allegations, the board will hold hearings before meting out a penalty against the professional.4 Sanctions can include a host of various penalties, from letters of concern to competency evaluations, to fines or restrictions.5 The most serious penalty can include

3  See, e.g., Dental Board Members, Indiana State Board of Dentistry, https://www.in.gov/pla/3208. htm; Board of Dentistry, Board Members, Tennessee Dep’t. of Health, https://www.tn.gov/health/ health-­program-­areas/health-­professional-­boards/dentistry-­board/dentistry-­board/members.html; Members of the Medical Board and Executive Staff, Medical Board of California, http://www.mbc.ca. gov/About_Us/Members/; Board of Medical Examiners, Montana Department of Labor and Industry, http://boards.bsd.dli.mt.gov/med#0?4; Statutory Composition and Current Membership, Nursing, New York Office of the Professions, http://www.op.nysed.gov/boards/bdcomp.htm#nur; Current Nursing Board Members, Minnesota Board of Nursing, https://mn.gov/boards/nursing/about-­us/ about-­the-­board/current-­board-­members.jsp. 4  Federation of State Medical Licensing Boards, US Medical Regulatory Trends and Actions, 2016, at 10–11, https://www.fsmb.org/globalassets/advocacy/publications/us-­medical-­regulatory-­trends-­ actions.pdf. 5  Id. at 10.

Regulation of Professionals and Facilities in the United States   297 s­ uspension or revocation of the professional’s license.6 This, for a healthcare p ­ rofessional, can be seen as the equivalent of a professional “death penalty” because revocation of one’s license ends the professional’s ability to see patients or otherwise practice ­medicine. And even though this regulatory regime is state-­centric, professionals cannot simply relocate: at least regarding medical doctors, states communicate with one another through the Federation of State Medical Board Physician Data Center by sharing physician disciplinary actions with other states.7 The American licensure and disciplinary system—in which the profession itself takes primary responsibility—has a number of positive attributes. From a clinical expertise perspective, state boards are often in the best position to sanction professionals. Populated by professional experts, the boards can employ clinical expertise both (1) in initially licensing professionals and (2) in determining whether a sanction is appropriate when a violation has been alleged. The interests of the profession—which should mirror the interests of the patients who access care in that system—should enjoy full discretion in such a self-­regulatory system. Professionals—who presumably practice medicine on a day-­to-­day basis—can argue that they are best-­positioned to determine when other professionals’ practices merit sanction. Professionals who follow the newest developments in the field can bring that expertise to bear. In this way, the board may be able to access additional information about the clinical episode due to the comfort level of physicians with colleagues and due to their increased expertise; relatedly, decision-­making professionals may make better decisions when it comes to determining whether professionals’ actions warrant sanction; and, ultimately, the disciplinary process may enjoy additional buy-­in and legitimacy from professionals because they believe their peers will be fairer in sanctioning poor treatment than either a non-­expert, nonclinician judge or a faceless bureaucracy. Furthermore, the structure of the licensing and disciplinary regime allows it to be dexterous and responsive to radical and rapid clinical changes. Regarding the specific penalties imposed, the board’s discretion allows for flexibility in fashioning the sanction, appropriately allowing for the nuance and individual details of a given professional’s case to influence the ultimate result. When it comes to medical issues, patients often trust healthcare professionals’ abilities to properly distinguish proper care from substandard care, and boards can access additional expertise to adequately mitigate or aggravate penalties for a given violation. But self-­regulation is not without a downside. Perhaps most prominently, the positive attributes of allowing a professional to discipline another professional can seemingly dissolve into the problem of professional self-­preservation and unevenness within the regulatory regime. First, self-­regulatory regimes may be susceptible to protecting friends or professional colleagues by not bringing disciplinary actions where they should and not imposing stringent enough penalties when the facts require them. Systems of self-­regulation inevitably suffer from collegial sympathies, as well as from the inherent incentive for professionals to avoid punishing their colleagues out of concern 6 Id.

7 Id.

298   Isaac D. Buck for retribution in the future. These and other concerns have led some in the US Congress to question whether medical boards have had appropriate oversight, especially in light of cases that have resulted in clear patient harm but have lacked sufficient physician penalties.8 It is also the case that boards are often underfunded, which can severely constrain their ability to investigate and sanction poorly performing professionals.9 Lack of resources may exacerbate the very real problems of unevenness and underenforcement. These concerns may put the perception of the board’s legitimacy at risk, resulting in damage to its public image. Beyond underenforcement, self-­regulatory regimes can also create the potential of an opposing concern: an appearance of, or built-­in incentive for, board members engaging in self-­interested behavior by overusing the power of sanction. Different from the aforementioned concern where the professional may be pressured to protect one’s friends or colleagues, professionals may also be incentivized to more harshly target enemies or competitors. This may be illustrated best by sanctions of professionals who employ unique clinical or alternative medicine techniques. In cases where the self-­regulatory enforcement mechanism is misused—most saliently in cases lacking any discernable patient harm—the regime may be perceived as approaching something of an anti-­ competitive tool, violative of antitrust principles and harmful to patient welfare. This is often highlighted by the fact that state medical licensing boards do not require proof of patient harm, most basically because the boards act not only to punish patient harm but also to prevent harm. Given the goals of such a regime, the lack of an allegation of harm should be expected. Nonetheless, a regulatory regime that can discipline professionals absent documented patient harm opens that regime up to allegations of unevenness. Indeed, where a profession is seeking to penalize professionals for administering potentially harmful care—in a profession that often pushes the standard of care forward by allowing professionals the discretion to try new clinical treatments—the regime creates an immiscible conflict of perception: one professional’s clinically innovative treatment that may constitute a breakthrough is another professional’s example of dangerously unproven and substandard care. Furthermore, as opposed to federal standards of care or guidelines published by various national associations, a state’s professional board may—comparatively—lack a robust standardization of rules and standards. In this way, professional discretion can cut both ways. Whereas the regime can positively fashion an individualized penalty, the self-­regulatory structure lacking clearly defined standards of care may be susceptible to unfair penalties or standards as well. States can prevent these claims by building a comprehensive set of rules and processes to investigate and determine alleged violations in a 8  Peter Eisler and Barbara Hansen, Thousands of Doctors Practicing Despite Errors, Misconduct, U.S.A. Today, Aug. 20, 2013, https://www.usatoday.com/story/news/nation/2013/08/20/ doctors-­licenses-­medical-­boards/2655513/. 9  See, e.g., Briana Vannozzi, Senate Testimony Alleges “Crisis” at State Nursing Board, njtv, Oct. 5, 2017, https://www.njtvonline.org/news/video/senate-­testimony-­alleges-­crisis-­state-­nursing-­board/.

Regulation of Professionals and Facilities in the United States   299 transparent manner. In this way, board hearings resemble more formalized legal proceedings. Finally, this state-­centered regulatory structure that places primacy on geography may also cause a number of tensions in an increasingly interconnected and mobile world. Rapid developments in the delivery of healthcare services—that seek to ameliorate societal harms—complicate the structure of state-­based regulation because they most notably challenge the geographic-­centric notion of regulation. For example, when one considers the solution of telemedicine to the rural healthcare crisis—where a patient would visit a mobile technology center or clinic and then have a clinical encounter with a professional hundreds of miles away—the clarity of state-­ based self-­regulation gets cloudy. Would it matter where, geographically, that professional was located? Would it matter where the patient lived? In the case of a conflict, would one state’s rules regarding licensure and discipline outweigh the other state’s rules? And, most foundationally, does one state have the authority for its patient-­ protective regime to apply it to protect the citizens of another state? Illustrated by the example of telemedicine, as the practice of medicine becomes more interconnected and more mobile, what to do with incomplete, ineffective, or outdated regulatory solutions borne out of the traditional strictures of the past will be vexing for policymakers.

3.2  State Regulation Between the Medical Professions In addition to regulating the practice within the medical profession itself, states have the duty to regulate the boundaries between various professions within medicine and between licensed professionals and unlicensed individuals. These rules seek to prevent incompetent individuals from providing medical care and to ensure that professionals offering healthcare services to patients in the state are adequately trained. In the United States, this type of regulation has most commonly been handled at the state level through rules that are known as “scope of practice” laws. Typically set by state statute, scope of practice laws set and define the type of medical services that can be ­provided by different professionals within the state’s borders. These state laws govern the practice of a number of different types of healthcare professionals, setting out which procedures and services are permitted and which are prohibited. Thus, for example, while chiropractors and physical therapists cannot prescribe medication, nurse practitioners and physician assistants generally can. Or, while some states grant independent prescribing authority to nurse practitioners, other states require supervision by a physician, especially for narcotic drugs or other controlled substances. Enforcement of these boundaries are serious: violations of scope of practice laws typically carry ­criminal penalties. Mirroring other quality-­ensuring rules like the self-­regulatory set of rules governing the practice of medicine, the purpose of scope of practice laws is harm prevention: specifically, to protect the public from harm that can be caused by the unlicensed provision of healthcare services. This can occur when an unlicensed professional administers

300   Isaac D. Buck healthcare services, but also can occur when professionals who are licensed under one set of professional rules practice beyond the scope of their professional licensure. Implementation and enforcement of scope of practice laws provide a clear market-­ restricting function, granting the state considerable control over the provision of healthcare services within its borders and over the boundaries between the health professions. Additionally, these quality-­protective laws signal to the public that the provision of healthcare services are tightly regulated, legitimizing a service that so often depends on patient trust. Nonetheless, as the provision and delivery of healthcare in America becomes more team-­oriented and collaborative, scope of practice rules have been subject to criticism and increasing tension. Rigid boundaries between the legitimate functions of different professionals are more difficult to enforce when all members of a healthcare team are providing collaborative care. As American medicine seeks to address its harmful and persistent fragmentation and siloization between a number of professionals through new payment and delivery models, state scope of practice laws may constitute a real barrier to these efforts. Importantly, scope of practice laws may impact the rural health crisis. Considering that a number of American counties have few—if any—licensed physicians,10 scope of practice rules can impose substantial hurdles for access to healthcare for Americans living in rural areas. When viewed in the context of the rural health crisis, critics claim these laws ham-­handedly protect market share and a ritualistic and outdated notion of quality to the exclusion of substantial concerns about access to care. In this way, scope of practice laws may constrict feeble markets that desperately need to be stimulated. Finally, scope of practice laws—like many of the self-­regulatory rules that govern various healthcare professionals—are criticized for being harmfully anti-­competitive and self-­interested. As new professionals—such as physician assistants11 and nurse practitioners12—proliferate in number and enhance their expertise—the goals focused on bringing down healthcare costs and expanding access to vulnerable or impoverished populations may increasingly conflict with traditional scope of practice statutes. The challenges facing American healthcare apply additional pressure to the uninterrupted application of scope of practice rules.

10  See Olga Khazan, Why Are There So Few Doctors in Rural America?, The Atlantic, Aug. 28, 2014, https://www.theatlantic.com/health/archive/2014/08/ why-­wont-­doctors-­move-­to-­rural-­america/379291/. 11  See Maria Clark, Why Physician Assistant Is the Fastest-­Growing Job in Medicine, New Orleans Times-­Picayune, May 15, 2018, https://www.nola.com/health/index.ssf/2018/05/physician_assistants_ grow_in_n.html?__vfz=rtw_top_pages%3D1446500015704. 12  See Number of Nurse Practitioners Hits New Record High, American Ass’n. Nurse Practitioners, Mar. 19, 2018, https://www.aanp.org/press-­room/press-­releases/173-­press-­room/2018-­ press-­releases/2190-­number-­of-­nurse-­practitioners-­hits-­new-­record-­high.

Regulation of Professionals and Facilities in the United States   301

3.3  Soft Power: Regulation Through Reimbursement, Board Certification, and Consumer-­Based Transparency In addition to self-­regulatory professional rules and scope of practice statutes, American professionals are also subject to “soft” regulation through reimbursement rules and board certification. Additionally, American healthcare has sought to inject additional consumer-­ based tools into the administration and delivery of healthcare. Reimbursement-­based regulation includes traditional health insurance incentives and directives as well as new payment models such as accountable care organizations (ACOs) and shared savings ­programs that also impact the practice habits of healthcare professionals. These policy pressures do not have the force of law or the market-­restricting power of self-­regulation, but nonetheless influence professionals to offer and supply healthcare services and ­treatments in particular ways. The most basic example of soft regulation is health insurance reimbursement. Pervasive within American medicine, health insurance companies create payment policies that seek to influence the provision of care. For instance, if a private insurance company wants a professional to administer care in a certain way or use a certain product or treatment, it can place limitations on its use through prior authorization.13 Societally, there is a tight balance between insurance companies’ goals in containing healthcare costs and ensuring quality, and professionals’ desires to maintain discretion over a course of treatment. When the public feels that insurance companies are influencing care in an illegitimate manner, they revolt—as was seen in the 1990s during the backlash to health maintenance organizations (HMOs).14 Timelier examples of reimbursement-­based regulation during the Affordable Care Act (ACA) era include a number of newly implemented payment models, including bundled payment systems, shared savings programs, and ACOs, as well as Medicare’s new merit-­based incentive payment system (MIPS) for professionals that was established under the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA). Under the MIPS program, if physicians (1) treat Medicare patients, (2) are not enrolled in an alternative payment model, and (3) do not qualify for some other exception from the program, their Medicare reimbursement rates are affected by their performance on four quality-­related metrics.15 In fact, the majority of the composite score, which determines whether or not the physician will experience an increase or a decrease in 13  See, e.g., What Is Prior Authorization and How Does the Process Work?, Cigna, Jul. 2018, https:// www.cigna.com/individuals-­families/understanding-­insurance/what-­is-­prior-­authorization. 14  Patients especially objected to obstacles to receiving care from specialists rather than primary care physicians and to select doctors. See, e.g., Robert J. Samuelson, HMO Backlash, Wash. Post, Mar. 4, 1998. See also Milt Freudenheim, H.M.O.’s Cope With a Backlash on Cost Cutting, N.Y. Times, May 19, 1996, https://www.nytimes.com/1996/05/19/us/hmo-­s-­cope-­with-­a-­backlash-­on-­cost-­cutting.html. 15  The metrics are quality measures, promoting interoperability, improvement activities, and cost measures. See Explore Measures and Activities, Quality Payment Program, Centers for Medicare and Medicaid Services, https://qpp.cms.gov/mips/explore-­measures/promoting-­interoperability?py =2019#measures.

302   Isaac D. Buck r­ eimbursement, is made up of a quality performance assessment of the care the physician provides—as opposed to the other three metrics.16 For example, physicians might be assessed based on the percentage of eligible patients who receive flu shots or the rate of complications from surgery. Nonetheless, due to the relative infancy of the MIPS ­program and recent changes brought about by the HHS that expanded exceptions, it  remains unclear whether or not MIPS will be fully implemented as initially contemplated. Similar to the involuntary MIPS program, physicians who participate in voluntary ACOs are also financially rewarded when they meet quality metrics as well as financial efficiency targets. Medicare’s new Shared Savings Program (MSSP) also rewards professionals who meet quality benchmarks and financial savings goals. By building these soft incentives into the reimbursement mechanism, payers—including and featuring public payers—can have a major impact on the quality of the care that physicians and other professionals administer without using the force of hard law, including its criminal or legal penalties. Furthermore, in the case of a mandatory program like MIPS, professionals who want or need profitable margins must heed the rules. In this way, smarter reimbursement techniques that promote medical quality and financial efficiency can play a long-­needed role in American medicine. Additionally, in an apt example of a “market channeling” function, American professionals can become “board certified,” which is a voluntary certification for professionals who want to be categorized as a professional with unique expertise. Many subspecialties, through medical associations in the United States, can use board certification as a signal to American healthcare consumers and patients that a particular professional has exceeded the minimum requirements for licensure and has exceptional expertise in a specialty. This serves as a channeling function in that board certification may encourage patients to seek care from those professionals who have these more impressive credentials. In addition, group practices may require board certification as a condition of joining the group, and hospitals may require certification before granting privileges to provide in-­hospital care. Finally, in an era of increased transparency—with a focus on stoking consumer-­based tools to encourage quality—physicians who participate in the Medicare program are facing increased disclosure of clinical information. Through CMS’s Physician Compare program, patients can access information about their current or potential professional, learning information that may be helpful in deciding which professional to visit for a clinical encounter.17 On a related note, CMS has also instituted the Open Payments disclosure program that makes public important information regarding financial relationships between physicians and drug or device manufacturers or between physicians and

16 Id. 17  About Physician Compare: An Overview, Centers for Medicare and Medicaid Services, Apr. 21, 2016, https://www.cms.gov/Medicare/Quality-­Initiatives-­Patient-­Assessment-­Instruments/physician-­ compare-­initiative/About-­Physician-­Compare-­An-­Overview.html.

Regulation of Professionals and Facilities in the United States   303 drug or device group purchasing organizations.18 These disclosures include honoraria, consulting fees, research funding, and ownership stakes held by physicians in the manufacturers or group purchasing organizations.19 Theoretically, these disclosures may impact patient decision-­making, particularly when the professional is recommending a certain course of treatment.

3.4  Conclusion The regulation of professionals in the United States is characterized by a multilayered regulatory scheme. This scheme includes self-­regulation, one in which the profession itself sets priorities for quality and enforcement. It is also organized by the state’s definitions of what it means to practice medicine in a given jurisdiction, which often defines the boundaries between and among the healthcare professions. Finally, soft and voluntary quality regimes—including reimbursement incentives, voluntary board certification, and other consumer-­based tools—are used to protect and promote the quality of American medicine. All regimes seek to both standardize heterogeneous medical practice throughout the industry and enforce quality standards in an effort to promote patient trust in the healthcare system.

4  Regulation of Healthcare Facilities Many of the same regulatory challenges facing professionals also face healthcare facilities, even though the regulatory regimes do differ in foundational ways. All healthcare entities—from hospitals, to nursing homes and other long-­term care facilities, to clinics—face a complex web of laws, regulations, and standards. And, similar to the multilayered and multifunctional regulation governing the practices of American healthcare professionals, healthcare facilities are subject to different kinds of regulation—primarily traditional hard law regulation by the government and soft regulation by payers and accrediting bodies. Differently from the regulation of professionals, however, facilities are subject to less self-­regulation. At the same time—and similar to the regulation of American professionals—the regulation of American facilities has tried to incorporate more consumer-­based tools that seek to empower patients to make better decisions about where, when, and how they seek care. This includes imposing data-­ gathering measurements and publicizing the  information for patients. These transparency-­ ensuring initiatives have been 18  Open Payments, Centers for Medicare and Medicaid Services, Jun. 29, 2018, https://www.cms.gov/ openpayments/. 19  The Facts About Open Payments Data, 2018, Open Payments Data Centers for Medicare and Medicaid Servs., Jun. 28, 2019, https://openpaymentsdata.cms.gov/summary.

304   Isaac D. Buck c­ omplemented with other “softer” pressures in recent years, including reimbursement-­ related incentives.20 In America, the force of regulatory quality for facilities seeks to strike a complicated balance: regulators want to enforce quality-­ensuring rules to protect patients, but must also be cognizant of the negative consequences impacting patients that come with penalizing an institution that has to reduce medical services in a community. Not completely unlike the regulatory regime facing professionals, often the regime for entities is confronted with the challenge of trying to appropriately regulate and punish poorly performing facilities without substantially compromising patient access to care. This makes for a complicated enforcement mechanism. As mentioned earlier, in an era of rural hospital closures in the United States, those who regulate the quality of entities in the United States may have to grapple with the question of whether the existence of a poorly performing facility is better than no healthcare facility at all. In its starkest terms, such an enforcement mechanism forces an intractable conflict between quality and access. Additional organizational differences, like, for example, religiously affiliated facilities, also complicate the regulatory structure. Although seemingly not as prevalent as in Ireland’s, Malta’s, or Croatia’s healthcare systems,21 America faces the same challenges as in European countries, particularly when it relates to abortion and other reproductive healthcare. Regulators must carefully balance religious freedoms with quality-­ enhancing and protecting regulations, often encountering difficult conflicts.22 Finally, market shifts have added tension to the regulatory regime. Most prominently, with healthcare entities increasingly consolidating and merging in recent years, an integrated marketplace has placed old rules governing healthcare delivery at risk of becoming obsolete. Corporate parents are buying hospitals, once-­competitor hospitals and clinics are merging, and physicians and other professionals are being hired as employees by healthcare facilities. As the primary actor in American healthcare has shifted from the fiercely autonomous action of the physician or professional to the standardized rules of a hospital system or facility, the importance of facility-­based regulation has only increased. Old rules that prevent corporations from “practicing medicine” or require a “certificate of need” (CON) before a hospital can change or add services can seem increasingly incoherent. This section also grapples with these new developments. In analyzing these developments, this section covering professional regulation will provide an overview of the regulation of healthcare facilities in the United States, mindful of the access, consolidation, and cost challenges facing those facilities in a time of rapid change for American hospitals and clinics. It also seeks to analyze how this transitional period will impact future changes in the regulatory landscape for healthcare 20  See Hospital Compare, Medicare.gov, https://www.medicare.gov/hospitalcompare/search.html?. 21  See, e.g., Legal, But . . . Abortion Access Under Threat in Catholic Croatia, Times of Malta, Mar. 28, 2019, https://timesofmalta.com/articles/view/ legal-­but-­abortion-­access-­under-­threat-­in-­catholic-­croatia.705790. 22  See Religious Beliefs and Healthcare Necessities: Can They Coexist?, American Bar Ass’n, Jun. 30, 2017, https://www.americanbar.org/groups/crsj/publications/human_rights_magazine_home/ human_rights_vol30_2003/spring2003/hr_spring03_religiousbeliefs/(last accessed Jan. 31, 2018).

Regulation of Professionals and Facilities in the United States   305 f­ acilities in an effort to provide a synopsis of American healthcare apt for comparison with America’s European counterparts.

4.1  State-­Based Licensure and Federal Accreditation American hospitals and other healthcare facilities are regulated on a dual-­track system. First, facilities must be licensed by the state in which they are located. In order to be licensed under state law, hospitals and other healthcare facilities must certify that they offer a number of services and treatments.23 These requirements ensure that healthcare facilities are adequately staffed and resourced to promote patient health. A state agency board, made up of a number of healthcare experts, is empowered to certify and license many different types of healthcare facilities—including hospitals, nursing homes, birthing centers, home care organizations, assisted living facilities, recuperation centers, ­residential hospices, ambulatory surgical treatment centers, and laboratories.24 Boards are empowered to set and enforce regulations to ensure that the healthcare facilities are safe for their patients. As further part of the certification and recertification process, state Survey Agencies (SAs) serve a primary role. SAs conduct surveys that are integral to the federal government’s review and certification process for participation in the Medicare program.25 In addition, these SAs also often conduct the survey process for certification for state Medicaid programs. Generally speaking, state requirements are relatively straightforward, and, once licensed, the healthcare facility then has the ability to market itself as such as long as it continues to comply with state law. Once the state has fully licensed the facility, the institution can decide to seek additional accreditation. Facilities in the United States can pursue many different types of accreditation, but, most commonly, American entities seek to achieve Joint Commission (JC) accreditation.26 Although facilities can be accredited through other private accrediting bodies, many do so through the JC because it is responsible for accrediting more than 21,000 healthcare facilities nationwide.27 Founded in 1951, the JC is the largest and oldest accrediting body for healthcare facilities in the United States.28

23  See, e.g., AHCA: Hospital and Outpatient Services Unit—Hospitals, Florida Agency for Healthcare Administration, http://ahca.myflorida.com/MCHQ/Health_Facility_Regulation/ Hospital_Outpatient/hospital.shtml. 24  See, e.g., Board for Licensing Healthcare Facilities, Tennessee Dep’t. of Health, https://www.tn. gov/health/health-­program-­areas/health-­professional-­boards/hcf-­board/hcf-­board/board.html. 25  See Quality, Safety & Oversight—Certification and Compliance, Jan. 10, 2018, https://www.cms. gov/Medicare/Professional-­Enrollment-­and-­Certification/CertificationandComplianc/index.html. 26  See Facts About Joint Commission Accreditation and Certification, Joint Commission, Oct. 9, 2017, https://www.jointcommission.org/facts_about_joint_commission_accreditation_and_certification/. 27  See Facts About Joint Commission, Joint Commission, Jul. 8, 2016, https://www.jointcommission. org/facts_about_the_joint_commission/. 28  See History of the Joint Commission, Joint Commission, https://www.jointcommission.org/ about_us/history.aspx.

306   Isaac D. Buck Once a facility receives JC accreditation, Medicare and Medicaid certification occur simultaneously. Facilities that fail to maintain JC quality standards and accreditation face the real possibility of a loss of Medicare certification, resulting in substantial financial penalties and/or potential suspension or closure. Accordingly, facilities seek and protect Medicare certification aggressively. To retain the JC’s “gold seal,” a healthcare facility is subject to survey and review every 3 years.29 Thus, certification of organizations can be secured based on a SA survey that, regarding Medicare, is performed at the behest of the federal government, or by a separate accrediting body like the JC “that has been recognized by CMS as having standards and a survey process that meet or exceed Medicare’s requirements.”30 Either can occur. In addition to Medicare certification, smaller, rural hospitals can also seek to become certified as a Critical Access Hospital (CAH), in which the hospital receives different and additional funding seeking to protect it from the financial threats that often accompany operating a hospital in rural America. The JC also specifically accredits facilities with subspecialties, including behavioral healthcare, home care, laboratories, and nursing homes.31 JC’s hospital accreditation includes general hospital, children’s hospital, long-­term acute care (LTACH), oncology, psychiatric, or rehabilitation hospitals, or even a different specialty hospital, including cardiac, orthopedic, or surgical hospitals.32

4.2  The Special Cases of Nursing Homes As America’s population ages, another important regulatory accreditation process is focused on nursing home and long-­term care facilities. Nursing homes and other long-­ term care facility services are more commonly covered by state Medicaid programs, and their quality-­enhancing regulations are often squarely within the domain of the state. These quality-­protective measures are often characterized as command and control licensure, where the state sets a standard, monitors for compliance, and penalizes those who violate the standards. Similar to other healthcare facilities, nursing homes require state licenses to open their doors, then also have to qualify for Medicaid coverage—and for those who accept it—Medicare coverage. Accreditation and quality review processes for nursing homes are impacted by a number of additional challenges that are not present in the hospital context. Most recently, the ACA shifted the regulation of nursing homes. Through the Nursing Home Transparency and Improvement Act and new consumer-­based tools like Nursing

29 Id. 30  See Facts About Federal Deemed Status and State Recognition, Joint Commission, Dec. 13, 2018, https://www.jointcommission.org/facts_about_federal_deemed_status_and_state_recognition/. 31  What Is Accreditation?, Joint Commission, 2018, https://www.jointcommission.org/ accreditation/accreditation_main.aspx. 32  What Is Accreditation?, Joint Commission, https://www.jointcommission.org/accreditation/ hospital_audience.aspx

Regulation of Professionals and Facilities in the United States   307 Home  Compare,33 the ACA updated the regulatory regime to an outcomes-­based ­regulatory regime. These new tools seek to increase the amount of transparency and public accountability for the quality of care that American nursing homes administer. The reason why—in the nursing home context—public regulation has occupied such a position of primacy is because other quality-­enhancing tools—such as the tools of medical malpractice, internal quality review, and training and accreditation—are all muted in nursing home regulation. For the many nursing homes that are for-­profit, licensure and accreditation processes for opening and running a nursing home are less strenuous. On top of these challenges, medical malpractice litigation—an important quality-­enhancing and patient-­protective regime in other medical contexts—is less available in the nursing home context due to a number of complicating factors: the presence of mandatory arbitration clauses, the fact that residents experience difficulty with causation in that is often hard to prove that an injury was caused by inappropriate care and not advanced age or disease, witness testimony is hard to uncover and secure, and, foundationally, the standards by which nursing home care are judged are often difficult to discern. Further complicating regulation, the power of the consumer is blunted because comparative quality metrics in nursing homes have historically lacked transparency. The hope is that the new tools installed by the ACA make real headway in achieving increased nursing home transparency of quality information for residents and their families.

4.3  Corporate Practice of Medicine Depending on one’s jurisdiction, CPOM statutes can provide a substantial practical limitation on a facility’s ability to administer healthcare services. Long the province of states regulating healthcare delivery and administration within their borders, the CPOM is a species of state statute that prevents corporations from practicing medicine, either directly or indirectly through employing physicians to deliver care. With increasing integration and consolidation in the healthcare space—and the resultant ascendance of the healthcare “system”—these rules seem to be increasingly in tension with broader trends in American medicine. Unremarkably, CPOM statutes have been the target of legal and policy-­based critique,34 tracking state efforts to systematically peel back their comprehensive prohibitions.35 These rules were originally promulgated with noble goals most generally interested in insulating professionals and medical care from any moneyed or profit interests: protecting the medical judgment of professionals, protecting the practice of medicine from commercialization, and ensuring that the interests of physicians and shareholders did 33  Five-­Star Quality Rating System, Jun. 27, 2018, https://www.cms.gov/medicare/professional-­ enrollment-­and-­certification/certificationandcomplianc/fsqrs.html. 34  See Nicole Huberfeld, Be Not Afraid of Change: Time to Eliminate the Corporate Practice of Medicine Doctrine, 14 Health Matrix 243 (2004). 35  See, e.g., Berlin v. Sarah Bush Lincoln Health Center, 688 N.E. 2d 106 (Ill. 1997).

308   Isaac D. Buck not conflict.36 Many states now have exceptions to these broad prohibitions, allowing certain hospitals (like non-­profit hospitals or Critical Access Hospitals) to employ physicians.37 Other states allow employment, but prohibit entities from limiting the professional’s independent medical judgment.38 These laws highlight the importance of management agreements in overseeing the administrative functions of hospitals.39 Changes in application of CPOM laws throughout numerous states have tracked shifts in healthcare delivery and administration more broadly; allowing hospitals to directly employ physicians and administer healthcare has overlapped with unprecedented recent consolidation in healthcare.40 In some states, this has increased tensions between market trends and state laws, and, in others, it has led to increased repeal and reshaping of CPOM statutes altogether. Practitioners in each state must make themselves aware of the appropriate structure of physician employment and healthcare delivery.

4.4  Certificate of Need Another state-­based limitation on the type and amount of healthcare delivered in a state includes CON laws. States with CON laws require healthcare facilities to receive state-­ sanctioned preclearance before the facilities can legally expand or build new hospitals or clinics; if hospitals wish to add on a wing, for example, they have to receive state approval beforehand. The states that impose CON laws say that they are necessary to track the need of health services in the state, to limit rising costs, and to secure access for their citizenry.41 Other states note that the goals of CON laws “promote delivery of high quality healthcare and ensure that services are aligned with community need.”42 Originally created to limit increasing costs by better controlling the supply of health  services, CON laws have also been subject to critique.43 The American Medical Association has argued that CON laws have not contained costs, constitute an 36  Issue Brief: Corporate Practice of Medicine, American Medical Association, 2015, https://www. google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=2ahUKEwjAotCGi9znAhWPY98K HUpvDqEQFjAAegQIBBAB&url=https%3A%2F%2Fwww.ama-­assn.org%2Fmedia%2F7661%2Fdownlo ad&usg=AOvVaw3I0sn8SPujJDC8-­wyv9tsD. 37 Id. 38 Id. 39  See Jennifer Brunkow, 3 Steps to Navigate Through the Corporate Practice of Medicine, Becker’s Hospital Review, Mar. 26, 2012, https://www.beckershospitalreview.com/legal-­regulatory-­issues/3-­ steps-­to-­navigate-­through-­the-­corporate-­practice-­of-­medicine.html 40  See Alex Kacik, For the First Time Ever, Less Than Half of Physicians Are Independent, Modern Healthcare, May 31, 2017, http://www.modernhealthcare.com/article/20170531/NEWS/170539971; Bonnie Darves, New England Journal of Medicine Career Center, Jul. 23, 2014, Understanding the Physician Employment “Movement,” http://www.nejmcareercenter.org/article/ understanding-­the-­physician-­employment-­movement-­/. 41  Certificate of Need (CON), Georgia Dep’t. of Community Health, https://dch.georgia.gov/ certificate-­need-­con. 42  See, e.g., Certificate of Need, New York State Dep’t. of Health, May 2018, https://www.health. ny.gov/facilities/cons/ 43  Patrick John McGinley, Beyond Healthcare Reform: Reconsidering Certificate of Need Laws in a “Managed Competition” System, 23 Fla. St. U. L. Rev. 141 (1995).

Regulation of Professionals and Facilities in the United States   309 anticompetitive regime, may constrain choice, and do not improve access or quality in American healthcare.44 Like other proposed policy solutions in the healthcare space, CON laws have been unable to slow the rising cost of American healthcare.45

4.5  Consumer-­Based Tools As in the professional context, consumer-­based tools impact facilities as well. The era of increased transparency has both increased the amount of information available to patients and the number of requirements that are applied to healthcare facilities. Because this era is still in its infancy, there are limited examples of these new data-­ gathering avenues resulting in policy changes. Nonetheless, these efforts have given American patients important information in determining where to seek care. Hospital Compare, the facility-­based analog of Physician Compare, provides patients and citizens quality information about hospitals participating in Medicare on 57 different metrics.46 Data are available to patients for a number of different types of hospitals, as well as from a number of different programs—including the Hospital Readmissions Reduction Program (HRRP) and Hospital Value-­Based Purchasing Program (HVBP).47 Also, teaching hospitals are subject to the Open Payments database, which—tracking a similar application seen in the physician context—makes public information about financial relationships between teaching hospitals and drug or device manufacturers or between teaching hospitals and drug or device group purchasing organizations.48 By publicizing the extent of these financial arrangements, CMS is seeking the disinfectant of public scrutiny to apply increased pressure to facilities that may have financial arrangements that—prior to the Open Payments database’s advent—were shielded from public view and accountability. Since these efforts are still quite new, these programs primarily serve as data reporting mechanisms, but their ability to influence law and ­policy in the future remains an important possibility.

4.6  Conclusion Like the regulation of its professionals, the regulation of facilities in the United States is characterized by an overlapping and dizzying array of care standards, modern federal incentives, and traditional state statutes. As American healthcare has increasingly 44  Certificate of Need: Evidence for Repeal, American Medical Association, 2015, https://www. google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=2ahUKEwjinKqUjNznAhUMh-­ AKHSGFAqUQFjAAegQIAhAB&url=https%3A%2F%2Fwww.ama-­assn.org%2Fmedia%2F14736%2Fdo wnload&usg=AOvVaw1UrMv8qy2df0CM-­2p60WHw. 45  See James Bailey, Can Health Spending Be Reined in Through Supply Restraints? An Evaluation of Certificate-­of-­Need Laws, 27 J. Pub. Health 755 (2019). 46  See Hospital Compare, Medicare.gov, https://www.medicare.gov/hospitalcompare/search.html?. 47 Id. 48  See Open Payments, Centers for Medicare and Medicaid Services, Jun. 29, 2018, https://www.cms. gov/openpayments/.

310   Isaac D. Buck become system-­based—moving away from the raw autonomy of an individual family physician, for example—the regulation of healthcare facilities has become more important to protecting American patients. As in the professional context, through the ACA, the regulatory regime has increasingly sought to employ consumer-­based tools to inject transparency into the system. Accordingly, older state regimes—like CON laws and the CPOM doctrine—have been subject to rising critique. And as the delivery and administration of American healthcare faces rapid change, the regulation of facilities—how and what that regulatory structure looks like—will only likely increase in importance.

Chapter 16

I n troduction to A n titrust a n d the Prov ision of Hea lthca r e i n the U n ited States a n d Eu ropea n U n ion Common Challenges Thomas Greaney and Okeoghene Odudu

Common to both the United States and the European Union is a sharp increase in the cost of healthcare provision. Both jurisdictions continue to face questions about the affordability of healthcare services and increased scrutiny of whether the resources committed to the healthcare system are used to best effect.1 For scholars and policymakers, two important issues arise: Do those with power to extract payment and profit from the healthcare system unfairly exploit such power? And how might such exploitation be prevented? In particular, to what extent can antitrust law be used to control and curtail the excesses of market power? A prevailing but misleading caricature paints the US healthcare system as exclusively market-­driven, populated by for-­profit providers afforded a high degree of autonomy and funded, whether by medical insurance or otherwise, from private sources. By this account, constraining the power of healthcare 1  Michael E Chernew and Joseph P Newhouse, ‘Healthcare Spending Growth,’ in Mark V Pauly, Thomas G Mcguire, and Pedro P Barros (eds) Handbook of Health Economics: Volume 2 (North Holland 2012), 17–26; Michael E Chernew and Dustin May, ‘Healthcare Cost Growth,’ in Sherry Glied and Peter Smith (eds) The Oxford Handbook of Health Economics (Oxford University Press 2011), 308–328, 7–9, 17–26; KE Thorpe, ‘The Rise in Healthcare Spending and What to Do About It’ (2005) 24(6) Health Affairs, 1436–1445.

312   Thomas Greaney and Okeoghene Odudu ­ roviders requires antitrust to be applied with full and unyielding force. At the other p extreme is what may be described as the European Social Model, which considers healthcare amongst the things that cannot be left to the market.2 Rather than the market, it is an obligation of government or civil society to ensure that healthcare services are available irrespective of a person’s ability to pay.3 On this account, markets are anathema, and competition law has no legitimate business in the domain of healthcare. These caricatures gloss over the subtle interplay of law and markets. In both the United States and the European Union, healthcare services are provided in a number of different ways, not only in different jurisdictions, but even within a single jurisdiction.4 In the United States, despite the appearance (and political rhetoric) proclaiming a market-­based system, government plays a large role, in common with systems operating within the European Union. Much healthcare is government-­sponsored, every sector is subject to extensive state and federal regulation so that provider autonomy is constrained, and a majority of hospitals operate on a not-­for-­profit basis (though physicians themselves are deemed to operate on a for-­profit basis). There is evidence of satisfaction with both government insurance (Medicare and Medicaid) and employer-­sponsored insurance and support for government supervision of all health services and insurance.5 In the European Union, despite the appearance (and political rhetoric) eschewing markets, attempts continue to be made ‘to capture some of the virtues we generally associate with the private sector and combine them with the social responsibility and concern for equity we associate with democratic government.’6 Market incentives are introduced, though, of course, there is variety in the way healthcare services are funded, in the extent to which healthcare services are provided by public or private bodies, in the extent to which providers operate on a for-­profit or not-­for-­profit basis, and in the extent and intensity of any regulation to which healthcare service providers are subject. The important point is that whilst citizens are guaranteed treatment or care, healthcare providers are no longer guaranteed funding but instead obtain resources according to the number of patients they attract and the services they provide, thus making it necessary to offer services that are attractive to patients and users.7 2  AG Colomer, Opinion in Case C-­265/08, rec. 3, and Case 155/73 Giuseppe Sacchi [1974] 409, AG Reischl Opinion at 437. 3  Jeremiah Hurley ‘An Overview of the Normative Economics of the Health Sector,’ in AJ Culyer and JP Newhouse (eds) Handbook of Health Economics: Volume 1 A (North Holland 2000) 55–118, 87–96. 4  The European Observatory on Health Systems and Policies systematically describes the functioning of health systems in the World Health Organisation (WHO) European Region as well as in some additional countries of the Organisation for Economic Cooperation and Development (OECD). These analyses are available at: http://www.euro.who.int/en/about-­us/partners/observatory/ publications/health-­system-­reviews-­hits/full-­list-­of-­country-­hits 5  Justin McCarthy, ‘Most Americans Still Rate Their Healthcare Quite Positively’ (2018, 7 December) https://news.gallup.com/poll/245195/americans-­rate-­healthcare-­quite-­positively.aspx 6  AC Enthoven, ‘Internal Market Reform of the British National Health Service’ (1991) 10 Health Affairs, 60–70, 69. 7  RB Saltman and C Von Otter, Planned Markets and Public Competition: Strategic Reform in Northern European Health Systems (Buckingham, Open University Press 1992), chapter 9.

Introduction to Antitrust and the Provision of Healthcare   313 Taking the United States and the European Union as examples, this brief introduction sets out some of the challenges faced and solutions adopted when seeking to use antitrust law to address market power problems arising in systems of healthcare provision. Common to both systems is a recognition that market power in private hands can be problematic. Common to both systems also is the use of antitrust law to challenge and restrain this source of power. In the United States, where market solutions have greatest acceptance, antitrust has played and continues to play an important—but not exclusive—role in setting boundaries regarding conduct that providers and payers may undertake. Myriad antitrust cases have challenged price agreements, market allocations, tying arrangements, and boycotts in various healthcare sectors. In European jurisdictions, despite the apparent absence of markets, antitrust has played a similar role on the provider side whilst leaving the payer side largely untouched. Thus antitrust has been used to prevent the use or abuse of market power held by healthcare professionals, institutional healthcare providers, or a combination of the two groups, to ensure that such agreements are in the interests of healthcare service users rather than the professionals or providers themselves. Where antitrust has struggled in European jurisdictions is with the behaviour of payers and the decisions that payers make, thus revealing—in Europe at least—some limits of antitrust.8 As both jurisdictions recognize, antitrust law is not a panacea. Gaps in the laws, market imperfections, and natural monopolies necessitate a range of government interventions. Hence regulation in support of improving consumer welfare is often needed, particularly where issues of quality and access to healthcare services arise. In the United States, antitrust law additionally oversees market structures, limiting the extent to which providers of healthcare services or health insurance may merge or engage in joint ventures. The course of applying antitrust law to healthcare mergers has not been entirely smooth. American merger enforcement faltered as a result of a series of court decisions (subsequently shown by economic studies to be erroneous) that rejected challenges to consolidation followed by an extended period of lax enforcement—which together contributed to a sharp increase in provider market concentration.9 However, recent court decisions applying sound analysis have resulted in a succession of government litigation victories challenging mergers.10 European jurisdictions have also experimented with systems of merger control, preventing the creation of structures in which healthcare providers may accumulate or exploit market power, but with less enthusiasm than in the United States. And even this limited enthusiasm is on the wane.11 This may reflect the view that the purchaser has the whip hand in systems in 8  See T Prosser, The Limits of Competition Law: Markets and Public Services (Oxford University Press 2005) echoing the famous FH Easterbrook, ‘The Limits of Antitrust’ (1984) 63 Texas L R, 1–40. 9  See Thomas L Greaney, ‘Coping with Concentration’ (2017) 36 Health Affs, 1564. 10  See, e.g., FTC & State of Illinois v. Advocate Healthcare Network, 841 F. 3d 460 (2016). 11  ME/6875-­19 Anticipated Merger Between the Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust and Poole Hospital NHS Foundation Trust (27 April 2020); available at https:// assets.publishing.service.gov.uk/media/5eb2dc08e90e070835525d24/Poole_Bournemouth_full_text_ decision_CMA.pdf

314   Thomas Greaney and Okeoghene Odudu which the payer is the sole purchaser of healthcare services for a given population. It may also reflect the view that a merger of providers does not significantly alter the way those providers are able to interact with the payer, though this is separate from the issue of how the merger affects interactions with patients.12 It is important to remember that antitrust law functions in the broader context of diverse public policies, cultures, norms, and economic conditions. Although the European Union and the United States are applying essentially the same antitrust doctrines and are similarly concerned with market power, given the heterogeneity of health systems, what can explain the different recourse to and reach of antitrust in the two model jurisdictions? In the United States, while by no means insignificant, direct government regulation of providers and payers is strictly cabined. Federal regulation stops short of regulating prices or allocating markets. This space for the market to operate can invite anti-­competitive behaviour. Accordingly, governmental supervision of markets under American antitrust law is robust. First, governmental antitrust enforcement in the United States is broadly dispersed. Two federal authorities, the US Department of Justice and the Federal Trade Commission share enforcement responsibilities, and every state attorney general has authority to bring antitrust cases based on both federal and state antitrust laws. Second, competitors and customers can seek both injunctive relief and damages for harm caused by violations of antitrust law. Vesting such power in such a variety of actors (and incentivizing private suits with treble damage recoveries adjudicated by lay juries) adds a democratic element to antitrust policy. Nurse midwives, labor unions, small independent practitioners, consumer class actions, and countless other ‘little guys’ have availed themselves of antitrust’s protective shield. Although states are permitted to immunize conduct from antitrust scrutiny, they can do so only if their statutes expressly authorize displacing competition and provide for active regulatory supervision of behaviors. Lest antitrust enforcement become too aggressive, other doctrines including standing, antitrust injury, and the overarching maxim that ‘antitrust law protects competition not competitors’ temper (though some say, stifle) impulses for litigation that advances private interests at the expense of socially beneficial competition. By contrast with the United States, the role of states in healthcare systems in the EU Member States reduces the autonomy healthcare providers have and thus might systemically reduce the extent to which market power exists or can be exercised. In some European systems, healthcare providers are assumed (or trusted) not to exploit any power they have over purchasers or patients. This may be, for example, because the provider operates on a not-­for-­profit basis, is controlled by a publicly accountable body, or is otherwise deemed to lack the incentive to exploit power. It may also be the case that countervailing power exists within the system so that the ability to exercise power is neutralised. Even if a market power problem exists, the question arises whether antitrust presents the best means of addressing use or abuse of market power held by dem12  Z Cooper, S Gibbons, S Jones, and A Mcguire ‘Does Hospital Competition Save Lives? Evidence from the English NHS Patient Choice Reforms’ (2011) 121(554) Econ J, F228–F260.

Introduction to Antitrust and the Provision of Healthcare   315 ocratically or politically accountable (national) bodies or whether alternative bodies of law, such as the rules on free movement, procurement, state aid, sector-­specific regulation, and judicial review, provide sufficient oversight and are to be preferred.13 The space for alternative means of addressing the problem to be applied exists because, unlike in the United States, there are a number of barriers to antitrust enforcement. Rather than diffuse, a relatively centralised enforcement system operates within the European Union, and the enforcement bodies have a great deal of discretion as to whether antitrust law should actually be applied.14 A final issue, in both the European Union and the United States, is how to describe the problems that remain untouched by antitrust and consider how such problems are best addressed. Beyond antitrust, there is experimentation with organisational forms, new forms of contracting, and government direction. For example, in the United States, there is innovation in payment methods by encouraging reimbursements based on outcomes and quality metrics. In addition, there are strong incentives fostered by both competitive pressures and government regulation for provider integration. In the European Union, there is near constant experimentation with modes of financing and provision—described as serial reorganisation or, less charitably, re-disorganisation.15 All of which raises important issues for future research. Do incentives for provider integration or other policies actually promote affordability and efficiency, and do they do so without creating market power problems? How is market power in these new modes of service delivery to be addressed? If a system or situation emerges in which market power is present, will calls to apply antitrust to curtail such power prevail, or will governments turn to command and control regulation?

13  See Case 1006/2/1/01 Bettercare II at para. 60 and ME/6815/19 Anticipated Acquisition by Aintree University Hospital Foundation Trust of Royal Liverpool and Broadgreen University Hospitals NHS Trust: paras. 5–6, 32–45. 14  For an example of the use of warnings rather than formal enforcement action, see OFT Welcomes Action by NHS Trusts to Ensure Compliance with Competition Law (OFT Press release 71/12 of 16 August 2012) available at https://webarchive.nationalarchives.gov.uk/20140402155501/http://oft.gov.uk/ news-­and-­updates/press/2012/71-­12 15  Christopher Pollitt, ‘New Labour’s Re-­disorganization’ (2007) 9(4) Pub Mgmt Rev, 529–543; doi:10.1080/14719030701726663; and Report of the Expert Panel on Effective Ways of Investing in Health (EXPH), on Typology of Health Policy Reforms and Framework for Evaluating Reform Effects, available at https://ec.europa.eu/health/expert_panel/sites/expertpanel/files/013_healthpolicyreforms_reformeffects_en.pdf

chapter 17

Hea lth Ca r e Serv ices a n d EU Competition L aw Okeoghene Odudu

1 Introduction European Union (EU) Member states organise their healthcare systems in various ways in order to ensure that services are available according to need rather than being determined by ability to pay. The systems of provision vary according both to the ­mech­an­ism of funding and the status of the providers operating within each European healthcare system.1 In the dimension of funding, there are three broad options. The first may be described as discretionary funding, by which individuals exercise autonomy on how to make provision to pay for any treatment or care that may be required. Next lies compulsory coverage, by which individuals are mandated to participate in a particular scheme or range of schemes that will pay for treatment or care required. This mech­an­ ism is a key characteristic of the Bismarck systems of healthcare provision prominent in Europe. Finally, payment may be made by the state from general taxation.2 This mech­ an­ism is characteristic of Beveridge systems of healthcare provision. In the dimension of provision, healthcare services may be provided by entities that are owned or controlled by the state or by entities that are independent of the state, operating either on a ­for-­profit or a not-­for-­profit basis.

1  The European Observatory on Health Systems and Policies systematically describes the functioning of health systems for countries in the World Health Organisation (WHO) European Region as well as some additional countries of the Organisation for Economic Cooperation and Development. These analyses are available at: http://www.euro.who.int/en/about-­us/partners/ observatory/publications/health-­system-­reviews-­hits/full-­list-­of-­country-­hits. 2  Chalkley and Malcomson (2000).

318   Okeoghene Odudu Funding Source discretionary

compulsory

taxation

for profit

Healthcare Providers

not-for-profit

State-owned

Figure 17.1  Health system matrix.

The two dimensions of a healthcare system can be used to produce a health system matrix, with the method of funding along the horizontal axis and the provider type on the vertical axis (see Figure 17.1). A common misconception is that the way EU Member States organise their healthcare systems, which ensures services are available according to need rather than being determined by ability to pay, results in the market—and there­ fore competition law—having little or no role to play.3 This fails to recognise that the problems that competition law is designed to address, problems of market power, may arise irrespective of the way a healthcare system is organised and are particularly preva­ lent in healthcare systems.4 Market power is to be addressed because it enables those that possess it to charge a higher price or offer a lower quality than those lacking market power, with the result that some individuals are denied access to goods or services that could otherwise be afforded. Possessors of market power are not only insulated from competition that would force them to lower price. They are also less able to minimise costs and so consume more resources to generate a given level of output than is con­ sumed by those without market power. Finally, those with market power are less willing or able to innovate—causing delay in the emergence of new and better goods and serv­ices.5 Competition law, alongside the law on procurement, sector-­specific regula­ tion, and judicial review, inevitably features in the conversation about how these market power problems are identified and addressed.6 The extent to which market power and its concomitant problems arise and the form and vigour with which market power problems manifest themselves may vary 3  Temple Lang (2005, 65), Case 1006/2/1/01 BetterCare Group Limited [2002] CAT 7, para. 71–­103; and Case T-­319/99 FENIN at para. 36. 4  Havighurst and Richman (2010–2011). 5  On the first concern see Scherer and Ross (1990, 15–55). On the second concern, see Frantz (1988, 54, 64–65) and Posner (1975, 807–828, 809). On the third concern, see Scherer (1992, 1425–1430). 6  See Hancher and Sauter (2012), Guy (2019), Pieters and van den Bogaert (1997), van de Gronden and others (2011, ch. 11–15), Department of Health (1994), and Dawson (1995).

Health care Services and EU Competition Law   319 depending on the structure of the healthcare system adopted. For example, in systems in which healthcare services are provided by not-­for profit providers, the extent to which market power is a problem or a problem that manifests itself in harmful ways is con­ tested. At the same time not-­for-­profit entities receive special treatment in many areas of law, including competition, and it is important to ask whether and when that special treatment is justified.7 Perhaps the most controversial elements in the healthcare system matrix are those where market power is held by a public body responsible for the provi­ sion of healthcare services.8 Controversy is most intense in elements when not only are the providers owned and operated by the state, but they are also financed by the state through taxation.9 Does market power manifest itself in these elements of the matrix, and does it do so in harmful ways? Does the ability of the state to control both the health­ care provider and bodies financing healthcare provision mean that market power prob­ lems do not arise or that hierarchical or political power is best used to address any problems that do arise? Although not without controversy, within a number of EU Member States, competi­ tion law has been used to address problems of market power in the healthcare services sector. In Section 2, the relevant EU and national competition laws are summarised. Section  3 considers the experience of applying those laws to providers of healthcare serv­ices. This chapter is chiefly concerned with healthcare services in England, though examples are drawn from other EU Member States. England allows healthcare services to be funded by all means on the system matrix.10 Funding from each source may be used for all healthcare provider types (i.e., those that are state-­owned and those that are independent of the state, the latter being able to operate on both a for-­profit and a not for profit basis so long as the provider is licenced).11 Examination of the English experience thus affords a view of the use of competition law to address market power problems in most elements of our health system matrix.12 In Section 4, three challenges that emerge from that experience of using competition law to address problems of market power in healthcare service markets are considered. The first challenge is the applicability of the competition law to healthcare service providers operating in each or every element of the healthcare system matrix. The second, accepting applicability, questions the appro­ priateness of the substantive rules for healthcare services. Finally, a battle of authority and autonomy considers whether decisions made by healthcare service providers should be subject to external review and the type of review that competition law offers.

7  Hansmann (1980, 836–837). 8  Department of Health (1988/89, para. 3.9). 9  Lear, Mossialos, and Beatrix (2010, 346). 10  For details of how the health service is financed in England, see Boyle (2011). 11  NHS Act 2006, s 12(1); Health and Social Care Act 2012, section 81; Regulations 7(2) (a) and 7(3) of the National Health Service (Procurement, Patient Choice, and Competition) (no 2) Regulations and accompanying guidance; Health and Social Care Act 2008 s 10 and Health and Social Care Act 2008 (Regulated Activities) Regulations 2014, SI 2014/2936, Sch 1. 12  On the extent and nature of the competition, see Office of Health Economics (2012).

320   Okeoghene Odudu

2  Applicable Rules There is no special EU regime governing market power possessed by those providing healthcare services. Instead, the general regime set out in Articles 101 and 102 of the Treaty on the Functioning of the European Union (TFEU) is applicable and applied.13 Article 101 TFEU prohibits any agreement or concerted practice made between two or more ‘undertakings’ (independent businesses) that would enable those undertakings to obtain or exercise market power.14 In contrast with Article 101 TFEU, Article 102 TFEU applies to those that possess market power unilaterally to ensure such power is not abused.15 A generally expressed concern is that, in applying competition law, there is an unwillingness or inability to balance national policies on the provision of health services against the EU’s market integration objectives.16 There is therefore a tendency to seek to exclude the matter from the scope of competition law rather than to balance conflecting interests and objectives within the norm.17 Within competition law, however, conduct caught by either of the prohibitions is exempted when the benefits (efficiencies) flowing from that prohibited conduct outweigh the identified harm. Additionally, Article 106(2) TFEU allows competition to be restricted when a polity demands that services be pro­ vided in a manner different from what would occur if market operators were unre­ strained.18 As examples, it may be that the provision of an accident and emergency service or the provision of training require competition to be restricted.19 Articles 101 and 102 of the TFEU are supplemented by systems to assess situations in which a first undertaking is able to exercise a degree of influence over a second under­ taking such that the two undertakings cease to offer an independent force on the market— a situation commonly described as a ‘merger.’20 EU Member States also operate their own competition law regimes in parallel with the EU regime and may capture conduct that falls outwith the scope of EU competition law. In the United Kingdom, the Competition Act 1980, the Competition Act 1998, and the Enterprise Act 2002, along with a special regime developed to apply competition principles in healthcare service

13  See guidance about how these powers are exercised in relation to healthcare services in England at note 22 and in Procurement Lawyers’ Association (2016). 14  Cowen and Sutter (1999), Caplan and Stringham (2003), and Cowen and Sutter (2005). 15  Communication from the Commission—Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings [2009] OJ C45/7, paras 8, 28–31. 16  Davies (1995) and infra note 46. 17 Id. 18  This argument is made in Cicoria (2006, 46). 19  On the former see Monopolies and Mergers Commission (1994, para. 11.29), and on the latter see Department for Health (1988/89, para. 4.29–4.30). 20  Article 3 (1) and (2) of Regulation 139/2004 on the control of concentrations between undertakings [2004] OJ L 2004/1 and Commission Consolidated Jurisdictional Notice under Council Regulation (EC) no 139/2004 on the control of concentrations between undertakings [2008] OJ C95/1, para. 16–23.

Health care Services and EU Competition Law   321 markets are available.21 There is extensive guidance setting out how those provisions are intended to be applied to healthcare services in England.22 The Competition and Markets Authority and the sector regulator, National Health Service (NHS) Improvement, share responsibility for ensuring compliance.23 Private parties are also able to enforce some of the obligations competition law imposes and may recover damages.24 In the Netherlands, a specialist healthcare regulatory, the Nederlandse Zorgautoriteit (NZa), was initially established and endowed with power to set prices and apply (some) aspects to competition law. Competition powers have now transferred to the general competi­ tion authority, the Autoriteit Consument and Markt (ACM).

3  The Rules Applied Competition law has been applied to address market power in the healthcare services sector in three discernible ways. The first concern is with those offering healthcare serv­ ices in a particular way or in a particular setting seeking to prevent others from offering equivalent services in a different way or from a different setting. An example of such exclusion or foreclosure would be, when healthcare providers with certain qualifications seek to prevent the provision of services by those with alternative qualifications or pro­ viders from using teams with a different skill mix than incumbents. Such reconfigura­ tion may enable provision at a lower cost or greater convenience and, if so, would manifestly be in the interest of patients and those funding healthcare services. The sec­ ond concern has been with agreements between those offering similar or equivalent serv­ices—so-­called horizontal collaboration. Whilst collaboration between those ­providing similar services may prove beneficial and improve the situation that would otherwise prevail, it may also result in increased prices or reduced levels of innovation, quality, or quantity of service. If two healthcare service providers agree to divide out-­of-­ hours availability so that they each are available on alternate weekends only, is this harmful or beneficial? Competition authorities have been called on to assess whether horizontal collaboration is of the harmful or the beneficial type. The final concern has been with preserving an environment that ultimately enables patients or purchasers to switch to another provider in response to a detrimental change in the price or quality of services on offer. This is important as it is observed generally that providers operating in such environments offer higher quality, charge lower prices, and are more innovative

21  Sanchez-­Graells (2015). 22  Competition & Markets Authority (2015a), Competition & Markets Authority (2015b), Monitor (2013a, 2013b, 2013c), Monitor and Competition & Markets Authority (2014), Office of Fair Trading, Competition Commission and Monitor (2013), Office of Fair Trading (2013), Cooperation & Competition Panel for NHS-­funded services (2010a), Monitor (2014a), Monitor (2013d), Monitor (2014b), and Cooperation & Competition Panel for NHS funded services (October 2010b). 23  See Sanchez-­Graells (2014) and Prosser (2010, 136–152). 24  Brealey and George (2019).

322   Okeoghene Odudu than those operating in an environment where the possibility of switching is absent. These three concerns are examined in turn.

3.1 Exclusion/Foreclosure Competition law has been used to scrutinize the extent to which those providing health­ care services create barriers that prevent others from offering those same services. In England, curiously, there is no legal impediment to those who have not undergone a period of medical training from offering healthcare services.25 Registered medical practitioners do, however, enjoy certain advantages over those not registered.26 In turn, registration is conditional on evidencing a particular level of competence, and it is inevitable that those who have not evidenced such a level of competence will be at a ­disadvantage.27 It is however recognized that the need to evidence competence in a particular manner may stifle innovation, since [e]limination of the ‘quack’ practitioner may involve also elimination of any practi­ tioner whose methods are new or unorthodox. Existing bodies of the qualified tend to be reluctant to recognise any potential advantages that such methods might have for future development.28

The concern with barriers to the provision of a healthcare service being imposed by other healthcare service providers has prompted action in a number of EU Member States, examples being action taken against those offering healthcare services below cost29 or refusing to supply patient records and other clinical information necessary for patients to receive treatment of care from competing providers of healthcare services.30 The Dutch Competition Authority has imposed a fine of €7.72 million (£6.4 million; US$9.8 million) on an association of medical practitioners for advising its members not to allow new general practitioners (GPs) to set up in competition, thus denying patients the opportunity to see new GPs with fresh ideas.31 Similarly, the competition authority in Ireland examined the ability of incumbent GPs (through their self-­regulatory body) to prevent the entry of new GPs if such entry would threaten their ‘viability.’32 As the authority observes

25  Monopolies and Mergers Commission (1994, para. 8.1). 26  See Section 2(2) of the Medical Act 1983 and Monopolies Commission (1970, paras. 43, 309 and appendix 11, at p 154–157, 161). 27  Id., para. 39 and 61. 28  Id., para. 51. 29  Finnish Competition Authority (2003) and Cooperation & Competition Panel for NHS-­funded services (2012a). 30  Cooperation & Competition Panel for NHS-­funded services (2012b), NMa (2003a, 2003b, 2004). 31  NMa (2012). 32  The Competition Authority (Ireland) (2010, paras. 2.59, 3.3, 6.17–6.31, 7.1–7.20, 8.8–8.13).

Health care Services and EU Competition Law   323 The main situation in which a new GP practice would threaten the viability of an existing practice is if it was providing a better service to patients or offering a better price to private patients.33

Fees to see a medical practitioner in Ireland had risen by 87% over a period when prices in general had risen by only 30%, and such an increase could in part be attributed to the exclusion of new GPs.34 In England, limits on the ability of healthcare service providers to promote their serv­ ices have been scrutinized as part of broader scrutiny of restrictions placed on members of the professions generally.35 Measures imposed by the General Medical Council (GMC), British Medical Association (BMA), and Royal College of General Practitioners (ReGP) preventing their members from engaging in promotional advertising have been found to prevent providers of healthcare services from attracting patients away from providers less able to meet the patients’ needs.36 The UK competition autherity accepted that some restrictions on advertising may be justified, and, to understand this, a separate examination of primary and secondary care was conducted.37 To the extent that a patient has a choice of primary care provider, the patient makes the choice unaided, and, in such a situation, advertising provides patients with useful information.38 Certainly, information relevant to that choice should not be suppressed, as would be the case if GPs are not free to advertise to patients.39 For sec­ ondary providers the competition authority accepted a limit on promotion direct to patients because the merits of a particular course of treatment are difficult to judge— even after the event—and such promotion may undermine the role of a primary pro­ vider in informing that choice.40 A restriction on secondary care providers promoting themselves to primary care providers was not, however, justified.41 The ability of those supplying healthcare services to prevent others wishing to supply healthcare services from doing so is most acute in the system by which patients are referred to a specialist (provider of secondary care) for diagnosis, treatment, or care by a GP (provider of primary care). Both the General Medical Council’s Professional Conduct and Discipline: Fitness to Practice and the British Medical Association’s Philosophy and Practice 33  Id., para. 7.14. Also Office of Fair Trading (2003a, para. 5.28). 34  The Competition Authority (Ireland) (2010, para. 2.18, 6.11). 35  Monopolies and Mergers Commission (1989, para. 7.6, 7.21, and 8.5, in which action is taken under section 7(1) (c) and (2) of the Fair Trading Act (1973). Consider also the Competition Authority (Ireland) (2010, para. 5.1–5.16), the European Commission’s approach at https://ec.europa.eu/ competition/sectors/professional_services/overview_en.html (accessed 24 November 2019), and Miller (1992). 36  Monopolies Commission (1970, para. 214, 251–252 and 272). 37  A general claim that ‘patient safety’ justified a broad restriction was not accepted because such concerns are to be addressed by the legislative regime: Id., para. 291. 38  Monopolies and Mergers Commission (1989, para. 3.9). On the choice of primary provider see https://www.nhs.uk/using-­the-­nhs/nhs-­services/gps/patient-­choice-­of-­gp-­practices/ 39  Id., para. 8.55–58.59. 40  Id., para. 7.3, 7.16, 7.21, 7.27, and 8.43. 41  Id., para. 8.46.

324   Okeoghene Odudu of Medical Ethics provide that specialists should not normally accept a patient without reference from the patient’s GP.42 The benefits of a referral system are acknowledged to be that a GP (i) is best placed to give authoritative advice on the type of specialist required; (ii) may reassure the patient that specialist treatment is unnecessary, thus reducing unnecessary cost; and, (iii) can ensure continuity of care by being the reposi­ tory of the patients’ medical history.43 These benefits are not universally accepted: self-­ referral is the norm in some branches of medicine, either for practical or social reasons (e.g., accident and emergency services, clinics for the treatment of sexually transmitted diseases, postnatal physiotherapy, and family planning and abortion clinics), or it takes place even when it is not generally accepted as appropriate (cosmetic plastic surgery). Still, the general need for a referral presents a barrier to those providing secondary care, and, as such, the referral decision and factors influencing that decision have attracted scrutiny to ensure that referral is not influenced by anything other than the patient’s interest.44 The concern with foreclosure prompted an enquiry into regulations restricting entry into the community pharmacies market in England.45 Regulations limited pharmacists entitlement to dispense medicines on behalf of the state, and since such dispensing accounts for approximately 80% of the turnover of an average pharmacy, in practice, the regulations controlled market entry. Authorisation to dispense medicines on behalf of the state was granted only when ‘necessary’ or ‘desirable’ in a particular area. Consequently an authorisation would not normally be granted in an area already served by one or more pharmacies. The removal of the entry control restrictions was recom­ mended because (i) one of the reasons for choosing one pharmacy over another is the quality of service on offer, such as medical advice given, and a correlation between increased quality of service and lower levels of concentration had been identified and (ii) it was estimated that the restriction resulted in prices being up to 30% higher than they would be absent the restriction.46

42  See Competition & Markets Authority (2014b, para. 2.40–42.44). 43  Monopolies and Mergers Commission (1989, paras. 6.7, 7.5, 7.27 and 8.41). 44  See as examples Monopolies and Mergers Commission (1989, paras 5.10–15.14, 8.25–28.26, and 8.41); Competition & Markets Authority (2014b, para. 2.39), Competition & Markets Authority (2014a, para. 6.65); Cooperation & Competition Panel for NHS-­funded services (2010a, 7.44–47.49), Cooperation & Competition Panel for NHS funded services (2010b, para. 6.94–96.101), Monitor (2014b, 22–23), and Dixon et al. (2010, 96 and 108). 45  Office of Fair Trading (2003a, para. 2.20–22.22, 3.8–3.12, 3.18–3.22). The investigation is carried out under section 2 of the Fair Trading Act 1973. See also Ofcom (2006a), in which the competition authority examined (and condemned) an agreement to exclude all but one provider of telecommunications facilities in state-­owned hospitals. This resulted in the use of mobile telephones on hospital premises being prohibited and high charges to use the facilities provided. 46  Office of Fair Trading (2003a, para. 4.9, 4.13, 4.33, 4.45–4.49, 4.66–4.67). But see House of Commons Health Committee (2014) expressing a ‘key criticisms of the OFT analysis and proposals is that they are based on issues of competition but that they neglect the healthcare perspective.’

Health care Services and EU Competition Law   325

3.2 Collaboration Collaboration between those offering equivalent or substitute services may prove beneficial.47 Competition authorities have intervened, however, when such collabora­ tion results in a more expensive or a lesser quality or quantity of service. A first situation when this might be the case is when healthcare professionals, as individuals or small groups, agree within each specialty how much each will charge third parties or on other terms of service. Unless each healthcare professional is an employee (in which case negotiations are through a trade union), each professional is viewed as a separate eco­ nomic entity operating in competition with all other healthcare professionals in the same medical speciality.48 Each healthcare service provider, as a separate economic entity or undertaking, is expected to ‘determine in­de­pend­ently the policy which it intends to adopt on the common market including the choice of persons and undertak­ ings to which he makes offers or sells.’49 Acting in­de­pend­ently, each healthcare service provider succeeds in attracting patients only by making their best offer in terms of qual­ ity and price.50 In Finland, Ireland, and England, prices have been found to be higher than they otherwise would have been when healthcare service providers breach that obligation and instead, for example, agree with other members of their speciality (often via the mechanism of their professional association) on the terms and conditions under which they offer services.51 Collective action to resist pressure on fees has been argued as necessary because Professional people, in common with others who have undergone higher education and specialised training . . . may expect to receive incomes regarded as being com­ mensurate with their qualifications.52

Alternatively, it has been claimed that collective action to resist pressure on fees is necessary in order ‘to maintain professional standards of integrity and independent judgment.’53 Both claims have been rejected, it being decided that the risks and disadvantages generally thought to flow from price regulation are pres­ ent in the professions as elsewhere, and that the absence of price competition in any 47  Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-­operation agreements OJ [2011] C 11/01. 48  On the employment status of healthcare professionals in OECD countries see the OECD Health System Characteristics Survey: http://www.oecd.org/els/health-­systems/physicians-­employmentstatus-­ training-­regulation.htm 49  Joined cases 40 to 48 etc. Suiker Unie ECLI:EU:C:1975:174, para. 173. 50  Cooperation & Competition Panel for NHS-­funded services (2010a, para. 7.62–7.63). 51  Monopolies and Mergers Commission (1994, paras. 5.4–5.5, 11.5–11.13, 11.26–11.28,11.109–111.120, 11.158–11.164), finding a restriction of competition within the meaning of section 7(1)(c) and (2) of the Fair Trading Act and the Competition Authority (Ireland) (2010, para. 7.25–27.32); on Finland, see http://www.oecd.org/finland/15360244.pdf (accessed 23 November 2019). 52  Monopolies Commission (1970, para. 127). 53  Id., para. 102.

326   Okeoghene Odudu profession is likely to have some disadvantageous effects on average costs and efficiency.54

Medical specialists may decide to cease independent practice and instead be employed or form a partnership so that the employer or group takes responsibility for determining the terms on which services will be offered to third parties.55 Whether medical specialists are able to benefit from this intra-­enterprise exception is determined by applying a two-­stage analysis to identify the degree to which the individual members are integrated.56 The first stage identifies services the group offers to third parties (as distinct from services to group members). In relation to such services identified, the second stage asks whether group members remain free to offer those same services other than through the group.57 This has involved consideration of whether members (i) generate profits for the common benefit of the group; (ii) operate under a common name; (iii) share administrative functions, such as joint billing; and (iv) there is a bank account (or accounts) in the name of the group and/or a single set of accounts is pro­ duced in respect of the group’s commercial activities.58 It is only when group members are unable to offer services to third parties other than through the group that each mem­ ber has ceased to be a separate undertaking and so benefits from the intra-­enterprise exception.59 In the UK competition agency’s ophthalmology investigation, the intra-­ enterprise exception could not be applied once it was found that the relevant healthcare professionals were able to provide the same services both through the group and in competition with the group of which they are members.60 It is not only collaboration between individual specialists that has been investigated but also that between institutions. For example, harm has been identified when state-­ owned healthcare service providers agree to observe minimum waiting times before making treatment available to patients, thereby eliminating competition based on speed of service.61 Harm has also been identified when six state-­owned healthcare providers exchanged information about the price each would charge for privately funded health­ care services.62 54  Id., para. 138. 55  Office of Fair Trading (2003b), finding that since ‘each . . . group . . . operates as a single undertaking . . . an agreement between the members of each group (within their respective groups) as to the levels of fees to be charged for their private professional services, will not amount to an agreement between undertakings.’ 56  Competition & Markets Authority (2015c, para. 3.59). 57  Office of Fair Trading (2003b). 58  Office of Fair Trading (2003b). Compare with the criteria set out by the European Commission in case no IV/M.1016 Price Waterhouse/Coopers & Lybrand, a Commission decision of 20 May 1998 (para. 7–8). 59  If answered in the affirmative, the formation of the group is properly analysed under the merger control provisions of the Enterprise Act 2002 rather than under the Competition Act 1998. 60  Competition & Markets Authority (2015c, paras. 1.16; 3.37; 3.60–63.114; 3.134–3.136). Not all the bodies subject to the decision seem to meet this criteria; see id., para. 3.15, note 32. 61  See Cooperation & Competition Panel for NHS-­funded services (2011). 62  Office of Fair Trading (2012a, 2012b).

Health care Services and EU Competition Law   327

3.3 Concentration It is observed generally that if users or customers are in an environment in which they may switch to another provider in response to a detrimental change in the price, quantity, or quality of services on offer, the users receive more favourable treatment than those in an environment in which such exit is not possible.63 The extent to which switching is possible is described as concentration, and the general observation is equally apposite in healthcare service markets.64 Concentration may be examined either ex ante using merger control powers or ex post using (primarily) the rules on market dominance. Ex ante control is preferred as problems are difficult to address ex post, and ex ante control is the focus of what follows.65 For the EU merger regime to be applicable, it is necessary that the matter affect ‘a substantial part of the common market.’ The patients’ desire to minimise travel has, however, resulted in markets for healthcare services being seen as subnational or local and therefore outwith the scope of the EU merger regulation.66 Consequently, it has been national merger control regimes that have been applied, as with the Enterprise Act 2002 in England.67 Here, transactions are examined when (i) two or more enterprises (ii) cease to be distinct.68 An enterprise is an activity carried on for gain or reward or supplied ‘otherwise than free of charge.’ Health service providers that are owned and funded by the state have been found to fall within this concept.69 In England, though each state-­owned healthcare provider is a separate legal entity with its own management and governing body exercising control over its commercial policy, certain providers remain subject to a power of direction enabling the Secretary of State to determine uni­ laterally their commercial behaviour.70 Transactions involving exclusively bodies sub­ ject to such powers of direction are not felt to involve the requisite change in control necessary to engaged the Enterprise Act 2002.71 Otherwise, transactions involving acute 63  Demsetz (1973) and Hirschman (1970). 64  Department of Health (1994, 10–11 and 16), Ferguson and Goddard (1997, 67–82), Goddard and Ferguson (1997, 15–20, 36–38), and the Netherlands Authority for Consumers & Markets (2017). 65  Schmid and Volker (2013, 303). 66  See Case no COMP/M.4367 APW/APSA/Nordic Capital/Capio, paras. 33–36. 67  For wider consideration, see Schmid and Varkevisser (2016). 68  Enterprise Act 2002, section 22 for completed mergers and 33 for proposed mergers. In addition to enterprises ceasing to be distinct, either the UK turnover of the target enterprise must exceed £70 million or the acquisition must enable the acquirer to increase to at least 25% its share of the supply of any goods or services in the UK or a substantial part of it. 69  Competition & Markets Authority (2014a, para. 1.8, 5.3–5.4, and 5.19–5.23) and Office of Fair Trading (2009b, para 3.8–3.9). For historical antecedents, see Bramwell v Lacy (1879) 10 Ch. D 691, 694–695 (per Jessel MR), (a hospital operating on a non-­for-­profit basis but offering services for which patients made payment according to their means was found to be a business) and Rolls v Miller (1884) 27 Ch D 71, at 88 (per Lindley LJ) (an organisation operating on a not-­for-­profit basis and not charging users of the services it provided was nonetheless engaged in business). 70  NHS Act 2006, sections 8, Sch 4 paras 3(1)(a) and 3(1)(b), and Sch 7 paras 7(1) and 15(1). 71  Competition & Markets Authority (2014a, paras. 2.1, 5.7), Office of Fair Trading (2013, para. 25), and Cooperation & Competition Panel for NHS-­funded services (2010b, para. 4.11–14.13, 4.20–24.23).

328   Okeoghene Odudu healthcare service providers, whether between private providers, such as the acquisition by Spire Healthcare Limited of Classic Hospitals Group Limited; between state-­owned provid­ ers, such as the acquisition by University College London Hospitals NHS Foundation Trust and Royal Free London NHS Foundation Trust’s neurosurgery services; or between state-­owned and private providers, such as the Pathology Services Joint Venture between Basildon and Thurrock University Hospitals NHS Foundation Trust, Southend University Hospital NHS Foundation Trust and Integrated Pathology Partnerships Limited have been reviewed, having satisfied the two criteria.72 In reviewing such transactions, the first stage— market definition—identifies the services affected by the change in control, the second stage identifies harm, and the final stage seeks to mitigate the harm identified.

3.3.1  Market Definition Market definition is an exercise performed to enable us to understand the constraints under which those providing goods or services operate.73 The standard approach for identifying antitrust markets is to ask whether it would be possible for a provider to impose on its customer a small but significant, nontransitory increase in price (SSNIP).74 When services are free at point of use or the price is fixed, price is not a factor in decision-­making, and the standard test is modified to ask whether it would be possi­ ble for a healthcare provider to impose a small but significant, nontransitory reduction in quality (SSNRQ).75 The aim is to understand the reasons why an increase in price or diminution in quality is or is not possible, and specific features of healthcare service pro­ vision contribute to this understanding. Application of the test may result in separate markets for privately funded healthcare and state-­funded healthcare.76 Privately funded healthcare may be further divided into that funded by an insurance arrangement and that differently funded, for example, self-­ pay or consumer credit.77 State-­funded healthcare services may be further divided into those funded through capitation and those funded on a fee-­for-­service basis.78 Each medical speciality may be considered separately, taking account of the extent to which specifically trained staff and dedicated facilities or equipment are required, such that a provider of one speciality cannot quickly shift to provide another speciality if conditions of demand change.79 The focus on supply-­side factors yields broadly defined clusters of services necessary to provide particular treatments.80 Although there is a risk of over­ stating the size of the market, Baker explains that the cluster market approach 72  See Cases ME/3610/08; ME/5574-­12, and ME/6252-­13. 73  Coscelli and Overd (2007, particularly in relation to regulated prices) and Baker (2007). 74  Office of Fair Trading (2010a, paras 5.2.9–16) and OXERA (2011). 75  Competition Commission (2013a, para. 5.5). 76  Office of Fair Trading (2010b, para. 13). 77  ME/4560/10 General Healthcare Group/Covenant Healthcare Group, para. 40. 78  Competition & Markets Authority (2014a, para. 6.80). 79  Competition Commission (2013c, para. 5.13–15.26 and Appendix E, paras. 1–22). 80  The competition authority typically focuses on demand-­side substitutability to determine the product market but also aims to understand whether others not currently offering a service are able to do so in a timely fashion: Office of Fair Trading (2010a, paras 5.2.6, 5.2.17) and Competition Commission (2013a, para. 5.8 and 5.10).

Health care Services and EU Competition Law   329 can be defended as a matter of analytical convenience: there is no need to define separate markets for a large number of individual hospital services, for example, when market shares and entry conditions are similar for each, or when data limita­ tions will effectively require that the same proxy (such as the number of hospital beds) be employed to estimate the market share for each individual service.81

Finally, there is a geographic dimension to the constraint felt by a provider of a par­ ticular cluster of healthcare services owing to limits on the willingness and ability of patients to travel. Noting a greater propensity to travel for more sophisticated services, so that specialized services may have a larger geographic market, a starting point is to consider alternative clusters of services as imposing constraints on behaviour if they are available within a 30-­minute travel time, though for accident and emergency services, this is reduced to between 14 and 19 minutes travel time.82

3.3.2  Assessment of Harm Within the defined market, the competition authority determines whether the merger may be expected to result in harm—that harm being articulated in terms of a substantial lessening of competition.83 Although not often made explicit, the concern is not directly with a substantial lessening of competition, but with the harm that such a lessening is expected to cause in ways such as increased length of hospital stay, increased travel times, the environment in which treatment or care is provided, the time required to wait for treatment or service, the method used to provide the service, and the availability of services or cost of provision.84 An indication of whether the transaction is likely to be harmful may be given by com­ paring the pre- and post-­transaction market share in each specialty.85 For example, in British United Provident Association Ltd and HCA United Kingdom, an increase in mar­ ket share from 9% to 14% did not give rise to increased market power, particularly as there were other sizable competitors.86 Greater weight, however, is now placed not on the size of the market served by a competitor but on its importance.87 This approach ulti­ mately led to the prohibition of a proposed merger between two state-­owned and largely state-­funded acute healthcare service providers in Dorset.88 It was found that each ­hospital provided patients with an important safeguard against poor performance in relation to certain elective in-­patient and out-­patient services. The management of each 81  Baker (2007). On the use of clusters, see Competition Commission (2013a, para. 5.13–15.22). Also, see the interdependence of particular services and ‘domino theory’ set out in West (1998, 172–173). For a critique of cluster markets, see Gaynor and Vogt (2000). 82  Competition & Markets Authority (2014a, para. 6.40–46.42), Department of Health (1994, 15–17; based on obligations imposed on Commissioners under the patients charter), Office of Fair Trading (2011a, para. 4.35–4.37, using 30-­minute drive time), and Beckert, Christensen, and Collyer (2012). 83  Enterprise Act 2002, sections 35 and 36. See guidance referred to at note 22 infra. 84  Competition & Markets Authority (2014a, para. 1.5) and Jung and Perrussel (2014). 85  Competition & Markets Authority (2014a, para. 6.43–46.45). 86  Monopolies and Mergers Commission (1990, 6.47). 87  Competition & Markets Authority (2014a, para. 6.46–6.49) and Monitor (2013b, para. 8.3). 88  Competition Commission (2013a).

330   Okeoghene Odudu hospital had made significant effort to reduce waiting times, implement best practice guidance, deliver innovative models of care, and improve facilities in order to attract patients from the other hospital, and the merger would remove pressure to make such efforts in future. Though a rare occurrence, not least due to the signal such action sends to the sector, unjustified increases in market power have resulted in state-­funded health­ care providers being prevented from merging in a number of EU jurisdictions, notably in Germany, where the Bundeskartellamt has blocked mergers between public hospitals.89

3.3.3 Benefits Nothing in merger control prevents action necessary to create an efficient and effective system of healthcare provision. When a transaction removes the constraint on behav­ iour that an alternative provider would otherwise impose, it is for the competition authority to consider whether the benefits of the transaction outweigh the harm of new or increased market power.90 It is therefore necessary to explain how the consolidation of particular clinical and administrative functions will benefit patients and those that fund the service. In England, the Dorset hospitals failed to put forward a compelling case. Healthcare service providers have, however, advocated mergers to enable efficient but small providers to combine and extract economies of scale and scope or to other­ wise enable the organisation to improve quality.91 Such studies as have been conducted do, however, tend to show that the proclaimed benefits of mergers between state-­owned and -funded healthcare service providers have not materialized.92

4  Three Challenges Competition law has the clear aim of controlling market power.93 Three challenges have been faced when seeking to pursue this objective in relation to healthcare service provi­ sion. The first is that there is debate over the extent to which EU competition law applies to healthcare service providers. There is next a set of questions about what EU competi­ tion law demands: What does compliance entail? Competition authorities have found it difficult to articulate and defend (and may not themselves be convinced of) the need to control market power in relation to healthcare services. The failure to articulate and defend the value of controlling market power fuels a recurring complaint that ensuring

89  Bundeskartellamt (2006). 90  Competition & Markets Authority (2014a, para. 7.4, 7.24–27.28). 91  Balan (2017), NHS Modernisation Agency (2007), and Frontier Economics and Boston Consulting Group (2012). 92  Goddard and Ferguson (1997, 39–52); Broers and Kemp (2017); and Gaynor, Laudicella, and Propper (2012). 93  Havighurst and Richman (2010–2011).

Health care Services and EU Competition Law   331 compliance imposes cost and consumes time for little if any discernible benefit.94 The final challenge is to the authority and autonomy of healthcare providers within a specific structure of provision. Health is a critical component of well-­being, but ill-­health can be caused by factors beyond the control of the individual sufferer.95 Different polities have devised different means of ensuring access to healthcare on the basis of need, rather than on the basis of non–health-­related attributes (such as willingness and ability to pay) and the extent to which competition authorities and competition law have author­ ity to interfere or influence any settlement reached between citizens, service providers, and funders remains a contentious issue.

4.1  The Perception of Uncertain Scope Competition law and healthcare seem dissonant owing to a view that health is priceless and that people therefore do not want to tradeoff quality for financial savings.96 As the British Medical Association has long asked ‘What patient faced with a life-­threatening condition would choose cut-­price surgery?’97 The idea that cost is not part of the clini­ cal decision seems an essential part of clinical autonomy and what has been described as the professional paradigm.98 Particularly in a publicly financed healthcare system it is, however, necessary to challenge the idea that ‘clinical autonomy bestowed an auto­ matic, unfettered right to use public resources without scrutiny or limits.’99 Klein refers to the focus on squeezing the greatest possible amount of healthcare from inevitably limited resources as the ‘ideology of efficiency.’100 The ideology of efficiency follows the realisation that the use of ineffective, inefficient or expensive methods of clinical intervention had implications for the NHS as a whole: funds so spent represented opportunities for­ gone for improving other parts of the NHS.101

To what extent does competition law impose discipline on healthcare expenditure? Health care providers, particularly those that are publicly owned or funded, tradition­ ally have had little understanding of competition law’s demands and of whether their behaviour is consistent with the law’s demands, and, because the law has rarely been enforced, have had little need to increase that understanding.102 The lack of enforcement 94  House of Commons Health Committee (2014, para. 88), Monitor (2014b, 12), Spencelay and Dixon (2014, 5–6, 67–70; though it is noted at fn 90 that the cost relates to preparation for a merger rather than the competition law assessment of the merger). 95  Hurley (2000, 55–118, 87–96). 96  Havighurst (1977, 478). 97  Monopolies and Mergers Commission (1989, para. 7.29). 98  Klein (2010, 27), Ameringer (2008, 21–41), Blumstein (1998, 91), and Blumstein (1994, 463–1486). 99  Klein (2010, 132). See, for example, Audit Commission (1999), Office of National Statistics (2004), Wanless and others (2007, 215–250), Webster (1998, 183–186), and Timmins (1996, 453–458). 100  Klein (2010, 47). 101  Id., 61. 102  Knox (2010, 21; comments of Leibenluft) and Currie (2013).

332   Okeoghene Odudu action taken against particular types of healthcare provider or those providing particular types of healthcare service prompts the question as to whether EU competition law is applicable to such provision. The UK competition authority initially justified its inaction in the sector by expressing doubt as to whether state-­owned healthcare providers that are funded from general taxation are subject to EU competition law, and this doubt has created a significant barrier to the use of competition law to address problems of market power in healthcare service markets.103 The competition authority’s position mirrored the UK government’s more general claim that the operation of the NHS lies beyond the scope of EU law. For example, in Müller-­Fauré, the United Kingdom Government points to the specific characteristics of the NHS and asks the Court to uphold the principle that healthcare provided under such a national sickness insurance scheme does not fall within the scope of Article [57 TFEU] and that the NHS, which is a non–profit-­making body, is not a service provider for the purposes of the Treaty.104

In response, the Court of Justice makes it clear that a medical service does not cease to be a provision of services because it is paid for by a national health service or by a system providing benefits in kind. . . . There is thus no need . . . to draw a distinction by reference to whether the patient pays the costs incurred and subsequently applies for reimbursement thereof or whether the sickness fund or the national budget pays the provider directly.105

There is little doubt that EU competition law applies to healthcare service providers, even when state-­owned and -financed. The rules apply to the conduct of undertakings, and the Court of Justice defines undertakings in Höfner as ‘every entity engaged in an economic activity, regardless of the legal status of the entity and the way in which it is financed.’106 Economic activity exists in relation to activities from which there is a potential to make a profit, and it is the possibility of making profit rather than the making of profit that is considered.107 Profit is possible if, as a matter of fact rather than law, a person can be prevented from enjoying the benefits of a service once that service is produced.108 Applying these criteria in 1997, in a short report titled The Consequences of 103  Office of Fair Trading (2010c, para. 1.8). See Odudu (2011). 104  C-­385/99 Müller-­Fauré [2003] ECR I-­4509, para. 59. 105  C-­385/99 Müller-­Fauré [2003] ECR I-­4509, para. 103, emphasis added. The implications of judgment are recognised by Munby J in In R (On The Application of Yvonne Watts) v Bedford Primary Care Trust and Secretary of State for Health [2003] EWHC 2228 (Admin) para. 107 and May LJ in Secretary of State for Health v Yvonne Watts [2004] EWCA Civ 166, para. 88. 106  Case C-­41/1990 Höfner and Elser v Macrotron [1991] ECR I 1979, para. 21 (emphasis added). The method of finance has mattered under other provisions of EU law, and there are often attempts made to read this jurisprudence into EU competition law. Compare Case 263/86 Humbel [1988] ECR 5365 with Case C-­102/92 Wirth [1993] ECR I-­6447, and see further Odudu (2009, 226–230). 107  Case T-­128/98 Aéroports de Paris v Commission [2000] ECR II-­3929 para 124. 108  Stiglitz (2000, 128–146).

Health care Services and EU Competition Law   333 European Competition Law for National Health Policies, Pieters and van den Bogaert expressed the view that little doubt can exist with regard to the possibility to qualify hospitals, both public and private, as well as individual healthcare providers, such as independent medical doctors, as ‘undertakings’ in the sense of Articles [101 and 102 TFEU].109

Similarly, in 2005, Hatzopoulos was writing, with regard to public hospitals being financed directly from the state b ­ udget . . . accord­ ing to the broad definition of ‘undertaking’ put forward by the Court whereby any activity pursued or susceptible to being pursued for profit is to be considered as an economic activity, even these public hospitals should be treated as undertakings.110

The UK competition authority felt that its doubt about the inapplicability of EU com­ petition law was justified by a ruling of the Court of Justice in FENIN.111 The idea that in FENIN the Court of Justice vindicates the UK position by ruling EU competition law inapplicable to at least state-­owned healthcare providers that are state-­funded is one that has reverberated around the EU, and so it is worth setting out in some detail why this understanding is to be doubted. FENIN involves a complaint to the European Commission about the purchasing practices of 26 organisations, including three ministries of the Spanish Government, which run the Spanish national health system. The Commission did not conduct an investigation, in part because it believed that the bodies complained of were not acting as undertakings. It is the decision not to conduct an investigation that is the subject of review by the courts. Although not the subject of the original complaint, before the General Court it was argued that hospitals within the Spanish healthcare system did act as undertakings.112 The Commission is very clear that its decision relates only to the activities of the organisations managing the health system and not the healthcare pro­ viders themselves, the General Court reporting The fact that certain hospitals may, exceptionally, provide services to private patients against payment is, at most, relevant only to determining whether those hospitals— not the bodies concerned by the complaint—act as undertakings in their dealings with private patients.113

The General Court did not consider the status of healthcare providers within the Spanish system because the conduct of the healthcare providers was not the subject of 109  Pieters and van den Bogaert (1997, 16). 110  Hatzopoulos (2005, 150). Also Hatzopoulos (2002, 705–720). 111  Office of Fair Trading (2004). 112  Case T-­319/99 FENIN [2003] ECR II-­357, para. 19 and 41. 113  Case T-­319/99 FENIN [2003] ECR II-­357, para. 34, emphasis added.

334   Okeoghene Odudu the contested decision.114 Again, the reasons why the status of healthcare providers is not considered are set out clearly by the General Court. An applicant bringing an action against a decision of the Commission rejecting its complaint in a competition matter cannot, therefore, criticise the Commission for failing to take account of facts which it has not brought to the Commission’s atten­ tion and which the Commission could only have discovered by investigation. In this case it must be observed, as indeed the Commission does in its rejoinder, that the applicant made no reference in its original complaint to the services which it alleges are provided for consideration. It mentioned them for the first time before the Court and then only in its reply. Therefore, in its review of the legality of the decision contested in the present action, the Court cannot take the existence of those services into account and it is not necessary in this case for the Court to rule on their potential relevance to the question whether the purchasing operations of those organisations amount to an economic activity.115

In the appeal against the General Court’s decision, the position of healthcare providers was again raised, but the Court of Justice ruled the claim inadmissible as the position of the healthcare provider, as distinct from the bodies managing the healthcare system, was not the subject of the decision under appeal.116 In FENIN neither the Commission, the General Court, nor the Court of Justice expressed a view or made a determination on whether the activities of healthcare providers, state-­funded or otherwise, state-­owned or otherwise, constituted economic activity subject to EU competition law. Advocate General Maduro did consider the matter, but was of the view that ‘The essential infor­ mation for concluding that the activity of providing healthcare of the SNS is of a non-­ economic nature is therefore not available’ and proposed that the Court of Justice refer the case back to the General Court for it to make the findings in fact necessary to deter­ mine the status of healthcare providers with the Spanish healthcare system.117 Given that no such investigation was conducted and no such findings made, FENIN in no way casts doubt on the applicability of EU competition law to healthcare providers. Furthermore, in guidance issued in 2011, the UK competition very much distanced itself from its previous approach.118 Attempts to remove at least certain types of healthcare provider—typically those that are state-­owned or -funded—from the reach of competition law in the United Kingdom continue.119 The NHS Long Term Plan, published by NHS England and NHS

114  Case T-­319/99 FENIN [2003] ECR II-­357, para. 34. 115  Case T-­319/99 FENIN [2003] ECR II-­357, paras. 43–44. 116  Case C-­205/03 P FENIN [2006] ECR I-­6295, para. 19–22. 117  Case C-­205/03 P FENIN [2006] ECR I-­6295, AG Opinion, para. 53 and 57. 118  Office of Fair Trading (2011c). 119  House of Commons Health Committee (2014, para. 91) and Collins (2015, 1). Compare with Entin, Fletcher, and Teske (1994) and Havighurst (2001).

Health care Services and EU Competition Law   335 Improvement in January 2019, proposes such action.120 The proposal does not, however, claim that market power problems do not arise, nor does it explain how such problems are to be resolved or balanced against other interests. If healthcare providers are undertakings within the EU concept, why then is there so little enforcement of EU competition law? The answer is complex, but can be explained in part by the fact that the applicability of EU competition law is triggered by interstate demand, and evidence suggests that patients want treatment or care as close to home as possible—the problems faced are inherently local. The ability of the patient to escape the local market tends to be addressed by national administrative law provisions or by EU free movement provisions.121 The ability of new healthcare providers to enter local mar­ kets in order to provide services proximate to the patient is addressed, at the EU level, by free movement and procurement law, certainly to the extent that the difficulties are caused by the regulatory regime or the way in which healthcare services are funded.122 It is only when the inability to enter a market is caused by other healthcare service provid­ ers that the difficulty is cognized as a competition law problem, but competition law has been consistently applied to such problems.123 The answer would seem to be that com­ petition law is applied to competition law problems—those of market power—and that a measured view as to when such problems arise and are best addressed by competition law has been taken. What is notable is that EU merger control is not applied, but only because the thresh­ olds for its application are not met. Since merger control is such a significant proportion of competition law, and much of a competition authority’s expertise and knowledge of a sector is derived from its examination of mergers, healthcare services have not moved to the forefront of the Commission’s competition law agenda. At the same time this has created space for national competition authorities to exercise their competence. However, whether significant competition concerns are raised sufficiently frequently in health­ care service provider mergers so as to justify ex ante review is certainly being questioned in England.124 A considerable amount of management time is involved in merger review, in addition, of course, to the need to ensure that the transaction does not create harm or that harm can be mitigated.125 There is no obligation to notify authorities of the transaction, but the risk of intervention and the cost of retrospective compliance make it prudent to do so.126 An alternative would, for example, exclude from scrutiny transactions that are the result of regulatory action (i.e., proposed by NHS Improvement), particularly 120  NHS England and NHS Improvement (2019) and House of Commons Health and Social Care Committee (2019). 121  See Directive 2011/24/EU on the application of patients’ rights in cross-­border healthcare [2011] OJ L 88, 45–65. 122  Bovis (2004, 299–300) and Arrowsmith (2005, 265). 123  See Section 3.1 in this chapter. 124  See House of Commons Health and Social Care Committee (2019) HC 2000 and NHS England and NHS Improvement (2019). 125  Competition & Markets Authority (2014a, para. 4.30). 126  Id., para. 4.1–4.3. A notification obligation was contained in Department of Health (1994).

336   Okeoghene Odudu if the system of healthcare provision and regulation allows market power problems to be addressed ex post.127

4.2  The Uncertainty of Competition Law’s Demands A significant question is whether market power problems arise, and arise with the same frequency or intensity, in all elements of the healthcare system matrix. This issue may explain why interest in and application of competition law varies amongst the EU Member States. Reluctance to enforce duties owned by healthcare service providers seems confined to certain elements of the healthcare system matrix.128 Consider, for example, the reticence with which the competition authority acted when finding that state-­owned healthcare providers in England engaged in illegal information exchange to the detriment of privately funded patients.129 Although information exchange relat­ ing to price ranks amongst the most egregious of competition law infringements, the file was closed once the competition authority had received assurances that the information exchange had ceased and that the parties would provide their staff with training on com­ petition law compliance. Furthermore, whilst concentrations of private healthcare serv­ ice providers has resulted in divestment or prevented further healthcare facilities from being acquired, in England the competition authority is content to ignore concentration when the provider is state-­owned or to exempt such providers without explanation of why the concerns it would generally raise do not arise.130 As a further example, consider examination of the decision that each state-­owned healthcare provider would appoint a single provider of bedside telephony and entertainment.131 The grant of exclusivity; the duration of the contracts, up to 30 years, often with an option to extend for a further 5 years; and prohibiting the use of mobile telephones on hospital premises to reinforce exclusivity were seen as responsible for the high price charged to those wishing to call a hospital patient.132 Rather than making a formal finding of infringement, the competi­ tion authority was satisfied that the Department of Health would enter into discussion 127  The potential for regulatory intervention to address concerns is explicitly provided in Article 2(1) of 2013 ACM Policy Rules as applied to the Dutch healthcare service market. Such an approach may well be seen as an aspect of the state-­action doctrine. 128  Monopolies and Mergers Commission (1994, para. 3.10), Competition & Markets Authority (2015c, para. 3.45; 2015a; 2015b). 129  Office of Fair Trading (2012b). 130  Competition & Markets Authority (2015a, 2015b), neither of which apply to those engaged in the same conduct within the NHS. Most recently, see ME/6815/19 Anticipated acquisition by Aintree University Hospital Foundation Trust of Royal Liverpool and Broadgreen University Hospitals NHS Trust (22 August 2019). 131  Ofcom (2006a). Under section 25 of the Competition Act 1998 and section 370 of the Communications Act 2003, Ofcom can investigate potential infringements of the competition rules. On the decision to be reviewed see Department of Health (2000, para. 4.19–4.20) and Department of Health (2004, para. 1.21). 132  Ofcom (2006a, para. 40–42).

Health care Services and EU Competition Law   337 with providers to examine mechanisms by which the price for calls could be reduced.133 These examples raise a number of issues that remain to be resolved. The first is whether the identified difference in treatment arises because the competition authority is of the view that not-­for-­profit providers do not exercise market power or exercise power in a way that is not harmful.134 The second is whether the identified difference in treatment arises because the competition authority is of the view that market power, or its exercise in a harmful way, is not a concern when there is payment by the state (and the payor has the ability to set the price).135 Does competition law make the same demands of all mar­ ket participants? If not, when and why do the demands vary? Uncertainty about what competition law demands can also arise because, certainly from the patient’s perspective, price is very rarely an issue owing to third-­party pay­ ment. Early guidance on competition rules applicable to healthcare service providers, however, continued to articulate harm in terms of price. This creates a disconnect between the patients experience—treatment gratefully received—and of the laws’ con­ cern: that power has been exploited. How is harm to be identified and explained if not in terms of price? Drawing on clinical evidence, it is becoming more common to identify and articulate harm in terms of quality or healthcare outcomes.136 It remains the case however that whilst [e]conomists and health services researchers have become extremely sophisticated in analyzing the relationship between competition and healthcare qual­ ity . . . [r]esearch on healthcare quality has had almost no impact on competition law and policy.137

One thing much craved is for competition authorities to enumerate what may and may not be done by healthcare providers. May the same answer be given regardless of the element of the healthcare system matrix in which a healthcare provider operates, or should each element have a different answer? At the same time, because the economic and social characteristics of healthcare provision so differ from those experienced in other markets, or because regulators or courts have little experience examining market power problems in the healthcare sector, it is often argued that market power concerns be rigorously substantiated (empirically) in each specific instance that competition law– based action is proposed.138 Respecting a desire for such a case-­by-­case approach comes 133  Ofcom (2006b) and Ofcom (2006a, para. 43, 48). See also Office of Fair Trading (2006, paras. 2.31, 5.17–5.19) and, more generally, Office of Fair Trading (2009a). 134  Philipson and Posner (2001). 135  Cooperation & Competition Panel for NHS-­funded services (2010a, para. 7.64). 136  Department of Health (1994), Monitor (2014b), OECD (2013, para. 73–86). On the degree of deference paid to clinical professionals, see Havighurst and Brody (1995). 137  Hyman (2004, 159; emphasis added). 138  Greaney (2006, 521). Cf Minnesota Ass’n of Nurse Anaesthetists v. Unity Hosp., 208 F.3d 655, 662 (8th Cir. 2000) and the failure to prove an actual adverse effect on competition, such as increased prices for anaesthesia services, or a decline in either the quality or quantity of such services available to surgery patients.

338   Okeoghene Odudu at the expense of the ability to promulgate clear rules (rather than standards) of expected behaviour.

4.3  The Authority of the State A Member State may seek to reserve the provision of profitable healthcare services to providers also subject to obligations to provide healthcare services that are not profit­ able.139 For example it has been suggested that a restriction of competition may be justi­ fied ‘in connection with for instance an ambulance service, which may be obliged to be available at all times to provide services, some of which will be performed at a loss’ or ‘if the area of an ambulance service was enlarged and . . . it was not allowed to charge more for services carried out over longer distances.’140 As memorably explained by Advocate General Maduro, in seeking to apply competition law to healthcare services, we ‘must find a balance between the need to protect undistorted competition on the common market and respect for the powers of the Member States.’141 To what extent has a public authority authorised the conduct or measures subject to scrutiny, and to what extent can competition law (as opposed to public law, particularly judicial review) be used to scru­ tinise or challenge measures so authorised? On what authority should a competition authority intervene to ensure that competition law is being rigorously complied with in such a situation? This issue of mandate is at the heart of the debate in England about whether state-­ owned healthcare providers should be subject to the laws in merger control. The Department of Health, NHS Improvement, and the NHS Trust Development Authority have encouraged healthcare service providers to merge as an alternative to insolvency, which would be costly, disruptive, and politically damaging.142 For a variety of reasons, not least political, state-­owned healthcare providers have been seen as too important to fail, and state intervention of some sort is demanded by the citizenry.143 Not only is it difficult to see how to fit such political intervention into competition analysis, it also is contested whether law, or economics, can trump political expediency, particular when there is a strong democratic mandate for such action. If healthcare providers have a democratic mandate to engage in particular conduct with particular effects, does not that mandate trump the constraints that would otherwise be imposed by competition law?144 Is the nature of the power no longer market power, but political power? No clear 139  Blumstein (1998, 93). 140  Temple-­Lang (2005, 45–77, 62 and 64). Compare with Department of Health (2010, para. 3.4). 141  Case C-­205/03 P FENIN, AG Opinion para 26. 142  Dixon, Harrison, and Mundle (2011, 27–30), Goddard and Ferguson (1997, 35–36, 81–82), and Dawson (1995, 10). 143  Azgad-­Tromer (2017). 144  Democratic participation is a shibboleth of mutual organisations in general and of the public benefit corporation in particular: Department of Health (2005, para. 1.4 and 1.10), Collison (1995, 75), and Mills (2010).

Health care Services and EU Competition Law   339 answer has been given, but it is not yet clear that a democratic mandate can be claimed, it having be found that The notion that [certain forms of healthcare provider] are more democratic in nature and are more accountable to local people has not really been supported by the research evidence.145

5 Conclusion EU competition can be applied and has been applied to control the exercise of market power in healthcare service provision. Challenges have arisen not from the fact that the law is sought to be applied to healthcare service providers but instead from the fact that the law is being applied to not-­for-­profit or bodies controlled by the state. Implicit in antipathy towards competition law enforcement is an idea that within such ownership or control structures lies a self-­executing constraint on abuse of market power, so that to apply competition law would supplant or undermine a superior framework for address­ ing any identified concerns.146 Naturally, if market power is suitably controlled or pres­ ents no risk, competition law recedes into the background. And what may be seen is that the range of alternative tools available to control market power and its abuse is de­pend­ ent on the element in the matrix in which healthcare services are being provided. For example, in relation to state-­funded healthcare, the healthcare provider may be regarded as a contracting entity subject to rules on public procurement.147 Calls to exclude state-­ funded or -controlled healthcare service provision from the law on procurement are not, however, unheard. It has also been suggested that The principal alternative methods . . . [of] challenge . . . include direct approaches to the responsible Minister. . . . Other possibilities . . . include a complaint . . . or, possibly, a claim for judicial review.148

The task of fleshing out an alternative means of addressing or preventing market power problems in the healthcare service market remains. As has been said in a different context, however, the problem of market power ought not to be ‘cocooned from the attention of the courts.’149

145  Bojke and Goddard (2010, para. 2.3). Also Day (2005). 146  Case 1006/2/1/01 Bettercare II at para. 60. 147  Bovis (2004, 299–300) and Arrowsmith (2005, 265). 148  Case 1006/2/1/01 Bettercare II at para. 61 (also para. 261). On the ability to obtain legal redress for any perceived abuse, see McHale, Hughes, and Griffiths (1996) and Jacob (1991). 149  Ex Parte Datafin at 839.

340   Okeoghene Odudu

Acknowledgments A debt of gratitude is owed to Mary Guy, Albert Sanchez Graells, David Stewart, and the Handbook editors for comments on earlier drafts of this chapter.

Bibliography Ameringer, Carl  F. 2008. The Health Care Revolution: From Medical Monopoly to Market Competition. Berkeley: University of California Press. Arrowsmith, Sue. 2005. The Law of Public and Utilities Procurement. London: Sweet & Maxwell. Audit Commission. 1999. Critical to Success. London: Audit Commission. Azgad-Tromer, Shlomit. 2017. ‘Too Important to Fail: Bankruptcy versus Bailout of Socially Important Non-Financial Institutions.’ Harvard Business Law Review 7: 160–186. Baker, Jonathan B. 2007. ‘Market Definition: An Analytical Overview.’ Antitrust Law Journal 74: 129–173. Balan, David J. 2017. ‘Merger-Specificity of Quality and Cost Efficiencies in Hospital Merger Cases.’ In Healthcare Mergers: A Post-Mortem, edited by David  S.  Evans, 44–49. Boston, MA: Competition Policy International. Beckert, Walter, M.  Christensen, and Kate Collyer. 2012. ‘Choice of NHS-Funded Hospital Services in England.’ Economic Journal 122(560): 400–417. https://onlinelibrary.wiley.com/ doi/pdf/10.1111/j.1468-0297.2012.02496.x Blumstein, James  F. 1994. ‘Health Care Reform and Competing Visions of Medical Care: Antitrust and State Provider Cooperation Legislation.’ Cornell Law Review 79: 1459–1506. Blumstein, James F. 1998. ‘The Application of Antitrust Doctrine to the Healthcare Industry: The Interweaving of Empirical and Normative Issues.’ Indiana Law Review 31: 91–118. Bojke, Chris, and Maria Goddard. 2010. Foundation Trusts: A Retrospective Review. Research Paper 58. York, UK: University of York, Centre for Health Economics. Bovis, Christopher. 2004. ‘Public Procurement and the Internal Market of the Twenty-First Century: Economic Exercise Versus Policy Choice.’ In European Union Law for the TwentyFirst Century: Rethinking the New Legal Order: Volume 2, edited by Takis Tridimas, and Paolisa Nebbia, 291–310. Oxford: Hart. Boyle, Sean. 2011. ‘United Kingdom (England): Health System Review.’ Health Systems in Transition 13(1): 1–483. Brealey, Mark, and Kyla George. 2019. Competition Litigation: UK Practice and Procedure (2nd ed.). Oxford: Oxford University Press. Broers, Bart, and Ron Kemp. Summer. 2017. ‘Dutch Hospital Mergers: No Evidence for Improvement of Healthcare Quality.’ In Healthcare Megers: A Post-Mortem, edited by David S. Evans, 60–67. Boston, MA: Competition Policy International. Bundeskartellamt. 2006. ‘Bundeskartellamt Prohibits for the First Time Merger Between Public Hospitals.’ https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/ 2006/13_12_2006_Krankenhausuntersagung_eng.html Caplan, Bryan, and Edward Stringham. 2003. ‘Networks, Law, and the Paradox of Cooperation.’ Review of Austrian Economics 16: 309–326. Chalkley, Martin, and James  M.  Malcomson. 2000. ‘Government Purchasing of Health Services.’ In Handbooks in Economics: Handbook of Health Economics, edited by A. J. Culyer and J. P. Newhouse, 847–890. Netherlands: Elsevier.

Health care Services and EU Competition Law   341 Cicoria, Cristiana. 2006. ‘European Competition Law and Nonprofit Organizations: A Law and Economics Analysis.’ Global Jurist Topics 6(1): 1–54. http://www.bepress.com/gj/topics/ vol6/iss1/art3. Collins, Ben. March 2015. Procurement and Competition Rules: Can the NHS Be Exempted? London: The King’s Fund. Collison, Peter. 1995. ‘The Democratic Solution.’ In Patients or Customers: Are the NHS Reforms Working?, edited by Reginald Murley, 68–76. London: IEA Health and Welfare Unit. The Competition Authority (Ireland). 2010. ‘Competition in Professional Services: General Medical Practitioners.’ https://www.ccpc.ie/business/wp-content/uploads/sites/3/2017/03/ General-Medical-Practitioners-Report.pdf Competition & Markets Authority. 2014a. CMA Guidance on the Review of NHS Mergers. London: Competition and Markets Authority. Competition & Markets Authority. 2014b. Private Healthcare Market Investigation: Final Report. London: HMSO. Competition & Markets Authority. 2015a. ‘CMA Open Letter to Private Medical Practitioners.’ https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attach­ ment_data/file/482193/Open_Letter_to_private_medical_practioners.pdf Competition & Markets Authority. 2015b. ‘Private Medical Practitioners: Information on Competition Law.’ https://www.gov.uk/government/publications/medical-practitionersadvice-on-competition-law Competition & Markets Authority. 2015c. ‘Conduct in the Ophthalmology Sector.’ Case CE/9784-13. https://assets.publishing.service.gov.uk/media/55d5989f40f0b609ff000009/ Conduct_in_the_ophthalmology_sector_decision_v2.pdf Competition Commission. 2013a. ‘The Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust/Poole Hospital NHS Foundation Trust.’ https://assets.publishing.service. gov.uk/media/55194835ed915d1427000151/131017_final_report.pdf Competition Commission. 2013b. ‘Optimax Clinics Limited and Ultralase Limited: A Report on the Completed Acquisition by Optimax Clinics Limited of Ultralase Limited.’ https:// assets.publishing.service.gov.uk/media/551948ec40f0b61401000199/final_report.pdf?_ ga=2.126600585.584598033.1566983923-354537020.1566571685. Competition Commission. 2013c. ‘The Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust/Poole Hospital, NHS Foundation Trust Merger Inquiry Provisional Findings Report.’ https://webarchive.nationalarchives.gov.uk/20140402203524/http://www. competition-commission.org.uk/assets/competitioncommission/docs/2013/royalbournemouth-and-christchurch-poole/130711_rbch_merger_pfsreport_excised.pdf Cooperation & Competition Panel for NHS-Funded Services. 2010a. ‘Conduct Guidelines.’ https://webarchive.nationalarchives.gov.uk/20130513202849/http://www.ccpanel.org.uk/ reports-and-guidance/index.html Cooperation & Competition Panel for NHS-Funded Services. 2010b. ‘Merger Guidelines.’ https://webarchive.nationalarchives.gov.uk/20130513202849/http://www.ccpanel.org.uk/ reports-and-guidance/index.html Cooperation & Competition Panel for NHS-Funded Services. 2011. ‘NHS Wiltshire Conduct Complaint.’ http://webarchive.nationalarchives.gov.uk/20130513202829/http://www.ccpanel. org.uk/content/cases/NHS_Wiltshire_Conduct_Complaint/110728_FINAL_Circle_ Report.pdf Cooperation & Competition Panel for NHS-Funded Services. 2012a. ‘North Yorkshire and York PCT and York Hospitals NHS Foundation Trust Conduct Complaint.’ http://webarchive.

342   Okeoghene Odudu nationalarchives.gov.uk/20130513202829/http://www.ccpanel.org.uk/content/cases/North_ Yorkshire_and_York_PCT_and_York_Hospitals_NHS_Foundation_Trust_Conduct_ Complaint/120322_York_PCT_Assura_conduct_complaint_PUBLISHED.pdf Cooperation & Competition Panel for NHS-Funded Services. 2012b. ‘South Staffordshire and Shropshire NHS Foundation Trust Conduct Complaint.’ http://webarchive.nationalarchives. gov.uk/20130513202829/http://www.ccpanel.org.uk/content/cases/South_Staffordshire_ and_Shropshire_NHS_Foundation_Trust_Conduct_Complaint/1206029_Decision_to_ take_no_further_action_SSSFT.pdf Coscelli, Andrea, and Alan Overd. 2007. ‘Market Definition in the Pharmaceutical Sector.’ European Competition Law Review 28(5): 294–296. Cowen, Tyler, and Daniel Sutter. 1999. ‘The Costs of Cooperation.’ Review of Austrian Economics 12: 161–173. Cowen, Tyler, and Daniel Sutter. 2005. ‘Conflict, Cooperation and Competition in Anarchy.’ Review of Austrian Economics 18: 109–115. Currie, David 2013. ‘The New Competition and Markets Authority: How Will It Promote Competition?’ (Beesley Lecture 2013). https://www.gov.uk/government/speeches/ the-new-competition-and-markets-authority-how-will-it-promote-competition Davies, Paul. 1995. ‘Market Integration and Social Policy in the Court of Justice.’ Industrial Law Journal 24(1): 49–77. Dawson, Diane. 1995. ‘Regulating Competition in The NHS: The Department of Health Guide on Mergers and Anti-Competitive Behaviour.’ Centre for Health Economics: York Health Economics Consortium: NHS Centre for Reviews & Dissemination Discussion Paper 131. Dawson, Diane, and Lyndsay Mountford. September 2008. Health Care Services and the Single European Market. Office of Health Economics Briefing No. 44. London: Office of Health Economics. Day, M. 2005. ‘Foundation Hospitals Are Not Accountable to Their Communities.’ British Medical Journal 331(7522): 926. Demsetz, Harold. 1973. The Market Concentration Doctrine: An Examination of Evidence and a Discussion of Policy. AEI-Hoover Policy Studies; 7. Washington, DC: American Enterprise Institute for Public Policy Research. Department of Health. 1988/89. Working for Patients. Cm 555 London: HMSO. Department of Health. 1989. Self-Governing Hospitals: Working Paper 1. London: HMSO. Department of Health. 1994. The Operation of the NHS Internal Market: Local Freedoms, National Responsibilities. Leeds: NHS Executive. Department of Health. 2000. ‘The NHS Plan: A Plan for Investment, a Plan for Reform.’ Cm 4818-I. https://webarchive.nationalarchives.gov.uk/20130124064356/http://www.dh.gov.uk/ prod_consum_dh/groups/dh_digitalassets/@dh/@en/@ps/documents/digitalasset/ dh_118522.pdf Department of Health. 2004. ‘The NHS Improvement Plan: Putting People at the Heart of Public Services.’ Gateway Ref: 3398. https://webarchive.nationalarchives.gov.uk/ 20080728104527/http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/ PublicationsPolicyAndGuidance/DH_4084476 Department of Health. 2005. ‘A Short Guide to NHS Foundation Trusts.’ Gateway 5591. https:// www.wwl.nhs.uk/Library/Foundation_Trust/Foundation_Trust_Guide.pdf Department of Health. 2006. ‘The Future Regulation of Health and Adult Social Care in England.’ Gateway Ref: 7377. https://webarchive.nationalarchives.gov.uk/20070129123449/ http://www.dh.gov.uk/Consultations/LiveConsultations/LiveConsultationsArticle/fs/ en?CONTENT_ID=4140711&chk=cJg8L/

Health care Services and EU Competition Law   343 Department of Health. 2007a. ‘The NHS in England: The Operating Framework for 2008/09.’ Gateway reference: 9120. https://webarchive.nationalarchives.gov.uk/20091105160851/http:// www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/ DH_081094 Department of Health. 2007b. ‘Principles and Rules for Cooperation and Competition.’ Gateway Ref: 9244. https://webarchive.nationalarchives.gov.uk/20080709160443/http://www.integrat­ edcarenetwork.gov.uk/_library/Operating_framework_annex_on_competition.pdf Department of Health. 2010. ‘Liberating the NHS: Regulating the Healthcare Providers: A Consultation on Proposals.’ Dixon, Anna, Tony Harrison, and Claire Mundle. 2011. Economic Regulation in Health Care: What Can We Learn from Other Regulators? London: The King’s Fund. Dixon, Anna, Ruth Robertson, John Appleby, Peter Burge, Nancy Devlin, and Helen Magee. 2010. Patient Choice: How Patients Choose and How Providers Respond. London: The King’s Fund. Entin, Fredric J., Tracey L. Fletcher, and Jeffrey M. Teske. 1994. ‘Hospital Collaboration: The Need For an Appropriate Antitrust Policy.’ Wake Forest Law Review 29: 107–168. European Commission. 2007. Case No COMP/M.4367 APW/APSA/Nordic Capital/Capio. https://ec.europa.eu/competition/mergers/cases/decisions/m4367_20070316_20212_en.pdf Ferguson, Brian, and Marie Goddard. 1997. ‘The Case for and Against Mergers.’ In Concentration and Choice in Healthcare, edited by B. Ferguson, T. Sheldon, and J. Posnett, 67–82. London: Royal Society of Medicine. Finnish Competition Authority. 2003. ‘Yearbook.’ https://www.kkv.fi/globalassets/kkv-suomi/ julkaisut/vuosikirjat/kilpailuvirasto/2003/kivi-vuosikirja-2003-en.pdf (accessed 24 November 2019). Frantz, Roger S. 1988. X-Efficiency: Theory, Evidence and Applications. Norwell, MA: Kluwer. Frontier Economics, and Boston Consulting Group. 2012. ‘A Study Investigating the Extent to Which There Are Economies of Scale and Scope in Healthcare Markets and How These Can Be Measured by Monitor.’ www.gov.uk/government/uploads/system/uploads/attachment_ data/file/303160/Monit or_Economies_of_Scale_and_Scope_-_FINAL_REPORT_0_0.pdf. Gaynor, Martin, Mauro Laudicella, and Carol Propper. 2012. Can Governments Do It Better?: Merger Mania and Hospital Outcomes in the English NHS. London: Centre for Economic Policy Research. Gaynor, Martin, and William  B.  Vogt. 2000. ‘Antitrust and Competition in Health Care Markets.’ In Handbook of Health Economics: Volume 1B, edited by A.  J.  Culyer and J. P. Newhouse. Amsterdam: North Holland. Goddard, Maria, and Brian Ferguson. 1997. Mergers in the NHS: Made in Heaven or Marriages of Convenience? London: The Nuffield Trust. Greaney, Thomas  L. 1995. ‘Regulating for Efficiency in Health Care Through the Antitrust Laws.’ Utah Law Review 1995(2): 465–501. Greaney, Thomas L. 2006. ‘Antitrust and Hospital Mergers: Does the Nonprofit Form Affect Competitive Substance?’ Journal of Health Politics Policy and Law 31(3): 511–530. Guy, Mary. 2019. Competition Policy in Healthcare: Frontiers in Insurance-Based and TaxationFunded Systems. Cambridge: Intersentia. Hancher, Leigh, and Wolf Sauter. 2012. EU Competition and Internal Market Law in the Health Care Sector. Oxford: Oxford University Press. Hansmann, Henry B. 1980. ‘The Role of Nonprofit Enterprise.’ Yale Law Journal 89(5): 835–901. Hatzopoulos, Vassilis. 2002. ‘Killing National Health and Insurance Systems but Healing Patients? The European Market for Health Care Services After the Judgments of the ECJ in Vanbraekel and Peerbooms.’ Common Market Law Review 39: 683–729.

344   Okeoghene Odudu Hatzopoulos, Vassilis. 2005. ‘Health Law and Policy: The Impact of the EU.’ In EU Law and the Welfare State: In Search of Solidarity, edited by G.  De Búrca, 111–168. Oxford: Oxford University Press. Havighurst, Clark  C. 1977. ‘Controlling Health Care Costs: Strengthening Private Sectors Hand.’ Journal of Health Politics Policy and Law 1(4): 471–498. Havighurst, Clark C. 2001. ‘Health Care as a (Big) Business: The Antitrust Response.’ Journal of Health Politics, Policy and Law 26(5): 939–955. Havighurst, Clark C., and Peter M. Brody. 1995. ‘Accrediting and the Sherman Act.’ Law & Contemporary Problems 57(4): 199–242. Havighurst, Clark C., and Barak D. Richman. 2010–2011. ‘The Provider Monopoly Problem in Health Care.’ Oregon Law Review 89(3): 847–884. Hirschman, Albert  O. 1970. Exit, Voice, and Loyalty; Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press. House of Commons Health and Social Care Committee. 2019. ‘NHS Long-Term Plan: Legislative Proposals. Fifteenth Report of Session 2017–19.’ HC 2000. https://publications. parliament.uk/pa/cm201719/cmselect/cmhealth/2000/2000.pdf House of Commons Health Committee. 2014. ‘Public Expenditure on Health and Social Care: Seventh Report of Session 2013–14.’ HC 793. https://publications.parliament.uk/pa/ cm201314/cmselect/cmhealth/793/793.pdf Hurley, J. 2000. ‘An Overview of the Normative Economics of the Health Sector.’ In Handbook of Health Economics: Volume 1, edited by A. J. Culyer and J. P. Newhouse, 55–118. Amsterdam: Elsevier. Hyman, D.  A. 2004. ‘Five Reasons Why Health Care Quality Research Hasn’t Affected Competition Law and Policy.’ International Journal of Health Care Finance and Economics 4: 159–166. Jacob, Joseph M. 1991. ‘Lawyers Go to Hospital’ Public Law. Jung, Nelson, and Sophie Perrussel. 2014. ‘Merger Control in the NHS.’ Mlex AB EXTRA: 1–6. Kauper, Thomas  E. 1988. ‘The Role of Quality of Health Care Considerations in Antitrust Analysis.’ Law & Contemporary Problems 51(2): 273–340. Klein, Rudolf. 2010. The New Politics of the NHS: From Creation to Reinvention (6th ed.). Oxford: Radcliffe. Kleis, Maria, and Phedon Nicolaides. 2006. ‘The Concept of Undertaking in Education and Public Health Systems.’ European State Aid Law Quarterly 5(3): 505–508. Knox, Ron. 2010. ‘Health Care: Round Table.’ Global Competition Review 13: 7, 21. Lear, Julia, Elias Mossialos, and Beatrix Karl. 2010. ‘EU Competition Law and Health Policy.’ In Health Systems Governance in Europe: The Role of European Union Law and Policy, edited by E.  Mossialos, G.  Permanand, R.  Baeten, and T.  Hervey, Tamara, 346. Cambridge: Cambridge University Press. Lemley, Mark A., and Christopher R. Leslie. 2007–2008. ‘Categorical Analysis in Antitrust Jurisprudence.’ Iowa Law Review 93: 1207–1270. McHale, J.  V., D.  Hughes, and L.  Griffiths. 1996. ‘Disputes in the NHS Internal Market: Regulation and Relationships.’ Medical Law International 2:215–227. Miller, Frances. 1992. ‘Competition Law and Anticompetitive Professional Behaviour Affecting Health Care.’ Modern Law Review 55(4): 453–481. Mills, Cliff. 2010. Public Services: Made Mutual. Borehamwood: Mutuo. Monitor. 2013a. ‘Enforcement Guidance on the Procurement, Patient Choice and Competition Regulations.’ https://webarchive.nationalarchives.gov.uk/20140204104030/http://www.

Health care Services and EU Competition Law   345 monitor-nhsft.gov.uk/sites/default/files/publications/ToPublishEnforcementGuidance 20May2013.pdf Monitor. 2013b. ‘Substantive Guidance on the Procurement, Patient Choice and Competition Regulations.’ https://webarchive.nationalarchives.gov.uk/20140204104013/http://www. monitor-nhsft.gov.uk/sites/default/files/publications/ToPublishSubstantiveGuidance 20May2013.pdf Monitor. 2013c. ‘Briefing Note: The Respective Roles of Monitor, the Office of Fair Trading and the Competition Commission in Relation to Mergers Involving NHS Trusts and NHS Foundation Trusts.’ https://assets.publishing.service.gov.uk/government/uploads/system/ uploads/attachment_data/file/284760/Briefing_sheet_-_Mergers_involving_NHS_trusts_ and_NHS_foundation_trusts.pdf Monitor. 2013d. ‘Hypothetical Case Scenarios: Procurement, Patient Choice and Competition Regulations.’ Publication code: IRG 37/13. https://webarchive.nationalarchives.gov. uk/20140204155702/http://www.monitor-nhsft.gov.uk/sites/default/files/publications/ ToPublishIllustrativeCaseStudies20May2013.pdf Monitor. 2014a. ‘Supporting NHS Providers: Guidance on Merger Benefits.’ https://assets.pub­ lishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/340823/ Monitor_mergerbenefits_guidance.pdf Monitor. 2014b. ‘Consultation Response: Application of the Competition Act 1998; Market Investigation References; Choice and Competition Licence Conditions.’ https://assets.pub­ lishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/354067/ CC_consulation_responses.pdf Monitor and Competition & Markets Authority. 2014. ‘Supporting NHS Providers Considering Transactions and Mergers.’ https://www.gov.uk/government/publications/supportingnhs-providers-considering-transactions-and-mergers Monopolies and Mergers Commission. 1989. ‘Services of Medical Practitioners: A Report on the Supply of the Services of Registered Medical Practitioners in Relation to Restrictions on Advertising.’ Command 582. https://webarchive.nationalarchives.gov.uk/20111202171538/ http://www.competition-commission.org.uk/rep_pub/reports/1989/241medicalservices. htm#full Monopolies and Mergers Commission. 1990. ‘The British United Provident Association Ltd and HCA United Kingdom Ltd: A Report on the Merger Situation.’ https://webarchive. nationalarchives.gov.uk/20111202184139/http://www.competition-commission.org.uk/rep_ pub/reports/1990/267buprovident.htm#full Monopolies and Mergers Commission. 1994. ‘Private Medical Services: A Report on Agreements and Practices Relating to Charges for the Supply of Private Medical Services by NHS Consultants.’ Command 2452. https://webarchive.nationalarchives.gov.uk/20111202183914/ http://www.competition-commission.org.uk/rep_pub/reports/1994/348privmedical. htm#full Monopolies Commission. 1970. ‘Professional Services: A Report on the General Effect on the Public Interest of Certain Restrictive Practices so Far as They Prevail in Relation to the Supply of Professional Services.’ Command 4463. https://webarchive.nationalarchives.gov. uk/20111202171107/http://www.competition-commission.org.uk/rep_pub/reports/ 1970_1975/059services.htm The Netherlands Authority for Consumers & Markets. 2017. ‘Price and Volume Effects of Hospital Mergers: Investigation into Effects of Hospital Mergers 2007–2014.’ Case Number ACM/17/009041. https://www.acm.nl/sites/default/files/documents/2018-01/report-priceand-volume-effects-of-hospital-mergers.pdf

346   Okeoghene Odudu Newman, John  M. 2015. ‘Antitrust in Zero-Price Markets: Foundations.’ University Of Pennsylvania Law Review 164(1): 149–206. NHS England and NHS Improvement. 2019. ‘The NHS’s Recommendations to Government and Parliament for an NHS Bill.’ https://www.england.nhs.uk/wp-content/uploads/2019/09/ BM1917-NHS-recommendations-Government-Parliament-for-an-NHS-Bill.pdf (accessed 28 Sept 2019). NHS Modernisation Agency. 2007. ‘10 High Impact Changes for Service Improvement and Delivery: A Guide for NHS Leaders.’ Gateway Ref: 3483. https://webarchive.nationalarchives. gov.uk/20081112125404/http://www.nodelaysachiever.nhs.uk/NR/rdonlyres/D942C9A2FCAA-454B-84B1-68D4A4959177/0/RG000610highimpactchanges.pdf NMa. 2003a. ‘NMa Suspects Pharmacies in Assen of Restricting Competition.’ https://www. acm.nl/en/publications/publication/5994/NMa-Suspects-Pharmacies-in-Assen-ofRestricting-Competition NMa. 2003b. ‘NMa Suspects Pharmacies in Breda of Restricting Competition.’ https://www. acm.nl/en/publications/publication/5862/NMa-Suspects-Pharmacies-in-Breda-of-RestrictingCompetition NMa. 2004. ‘Pharmacies in Assen and Tilburg Must Lift Restrictions on Competition.’ https:// www.acm.nl/en/publications/publication/5859/NMa-Pharmacies-in-Assen-and-TilburgMust-Lift-Restrictions-on-Competition NMa. 2012. ‘NMa Fines Dutch National Association of General Practitioners for Illegal Establishment Recommendations.’ https://www.acm.nl/en/publications/publication/6719/ NMa-fines-Dutch-National-Association-of-General-Practitioners-for-illegal-establishmentrecommendations Odudu, Okeoghene. 2009. ‘Economic Activity as a Limit to Community Law.’ In The Outer Limits of European Union Law, edited by C. Barnard and O. Odudu, 225–243. London: Hart. Odudu, Okeoghene. 2011. ‘Are State-Owned Health-Care Providers Undertakings Subject to Competition Law?’ European Competition Law Review 32(5): 231–241. OECD. 2013. ‘Roundtable on the Role and Measurement of Quality in Competition Analysis.’ DAF/COMP(2013)8 SSRN-Id2288043. http://www.oecd.org/competition/Quality-incompetition-analysis-2013.pdf Ofcom. 2006a. CW/00844/06/05. ‘Own-Initiative Investigation into the Price of Making Telephone Calls to Hospital Patients: A Case Closure Document Issued by the Office of Communications.’ https://www.ofcom.org.uk/__data/assets/pdf_file/0023/81149/cw_844_ case.pdf Ofcom. 2006b. ‘Ofcom Refers Price of Hospital Calls for Review by Department of Health.’ https://www.ofcom.org.uk/about-ofcom/latest/media/media-releases/2006/ofcomrefers-price-of-hospital-calls-for-review-by-department-of-health Office of Fair Trading. 2003a. ‘The Control of Entry Regulations and Retail Pharmacy Services in the UK.’ Office of Fair Trading. 2003b. ‘Anaesthetists’ Groups.’ http://webarchive.nationalarchives.gov. uk/20140402142426/http://www.oft.gov.uk/OFTwork/competition-act-and-cartels/ca98/ decisions/anaesthetists-groups Office of Fair Trading. 2004. ‘The Competition Act 1998 and Public Bodies.’ OFT 443 Policy Note 1/2004 Office of Fair Trading. 2006. ‘Market Investigation References: Guidance About the Making of References Under Part 4 of the Enterprise Act.’ OFT 511. Office of Fair Trading. 2008. ‘Restatement of OFT’s Position Regarding Acquisitions of Failing Firms.’ OFT 1047.

Health care Services and EU Competition Law   347 Office of Fair Trading. 2009a. ‘Government in Markets: Why Competition Matters—A Guide for Policy Makers.’ OFT1113. Office of Fair Trading. 2009b. ‘Mergers: Jurisdictional and Procedural Guidance.’ OFT 527. Office of Fair Trading. 2010a. ‘Merger Assessment Guidelines.’ OFT 1254 CC2 (Revised). Office of Fair Trading. 2010b. ‘Anticipated Acquisition by Priory New Investments No 3 Limited of Affinity Healthcare Holdings Limited: The OFT’s Decision on Reference Under Section 33(1).’ Office of Fair Trading. 2010c. ‘Competition in Mixed Markets: Ensuring Competitive Neutrality: A Working Paper.’ OFT 1242. Office of Fair Trading. 2011a. ‘Private Healthcare Market Study: Report on the Market Study and Proposed Decision to Make a Market Investigation Reference.’ OFT 1396. Office of Fair Trading. 2011b. ‘Anticipated National Address Gazetteer Joint Venture Between Ordnance Survey and Local Government Improvement and Development.’ ME/4802/10. Office of Fair Trading. 2011c. ‘Public Bodies and Competition Law.’ OFT 1389. Office of Fair Trading. 2012a. ‘OFT Welcomes Action by NHS Trusts to Ensure Compliance with Competition Law.’ http://webarchive.nationalarchives.gov.uk/20140402161850/http:// www.oft.gov.uk/news-and-updates/press/2012/71-12 Office of Fair Trading. 2012b. ‘Competition Law and Private Patient Units.’ http://webarchive. nationalarchives.gov.uk/20140402161850/http://www.oft.gov.uk/shared_oft/publicmarkets/PPUs.pdf Office of Fair Trading. 2013. ‘The OFT’s Role in Reviewing NHS Mergers: Frequently Asked Questions.’ https://assets.publishing.service.gov.uk/government/uploads/system/uploads/ attachment_data/file/284395/NHS_FT_FAQs.pdf Office of Fair Trading, Competition Commission, and Monitor. 2013. ‘Ensuring That Patients’ Interests Are at the Heart of Assessing Public Hospital Mergers.’ https://assets.publishing. service.gov.uk/government/uploads/system/uploads/attachment_data/file/283809/131017_ oft_cc_monitor_merger_statement_final.pdf Office of Health Economics. 2012. ‘Competition in the NHS.’ https://www.ohe.org/publica­ tions/report-office-health-economics-commission-competition-nhs Office of National Statistics. 2004. ‘Public Service Productivity: Health.’ Economic Trends 613(December): 38–59. Office of National Statistics. 2019. ‘Healthcare Expenditure, UK Health Accounts: 2017.’ Newport: Office of National Statistics. OXERA. 2011. ‘Techniques for Defining Markets for Private Healthcare in the UK: Literature Review.’ Prepared for Office of Fair Trading. London: OXERA. Philipson, T. J., and R. A. Posner. 2001. ‘Antitrust and the Not-for-Profit Sector.’ Journal of Law and Economics 52: 1. Pieters, Danny, and Stefaan van den Bogaert. 1997. The Consequences of European Competition Law for National Health Policies. Antwerpen: Maklu. Posner, Richard A. 1975. ‘The Social Costs of Monopoly and Regulation.’ The Journal of Political Economy 83(4): 807–828. Procurement Lawyers’ Association. 2016. ‘The Procurement and Competition Regimes Applicable to National Health Service Commissioners and Providers in England.’ https:// www.procurementlawyers.org.uk/wp-content/uploads/2017/07/PLA_NHS-ProcurementWorking-Group-Paper.pdf Prosser, Tony. 2010. The Regulatory Enterprise Government, Regulation, and Legitimacy. Oxford: Oxford University Press. Rossi, Cecilia, Russell Whitehouse, and Alexander Moore. 2019. ‘Estimating Diversion Ratios in Hospital Mergers.’ Journal of Competition Law & Economics 15(1): 639–663.

348   Okeoghene Odudu Sage, William M. 2014. ‘Getting the Product Right: How Competition Policy Can Improve Health Care Markets.’ Health Affairs 33(6): 1076–1082. Sanchez-Graells, Albert. 2014. ‘Monitor and the Competition and Markets Authority.’ University of Leicester School of Law Research Paper No. 14-32. http://dx.doi.org/10.2139/ ssrn.2528569. Sanchez-Graells, Albert. 2015. ‘New Rules for Health Care Procurement in the UK. A Critical Assessment from the Perspective of EU Economic Law.’ Public Procurement Law Review 24(1): 16–30. Scherer, F.  M. 1992. ‘Schumpeter and Plausible Capitalism.’ Journal of Economic Literature 30(3): 1416–1433. Scherer, F. M., and David Ross. 1990. Industrial Market Structure and Economic Performance (3rd ed.). Boston, MA: Houghton Mifflin. Schmid, Andreas, and Ulrich Varkevisser. 2016. ‘Hospital Merger Control in Germany, the Netherlands and England: Experiences and Challenges.’ Health Policy 120(1):16–25. Schmid, Andreas, and Ulrich Volker. 2013. ‘Consolidation and Concentration in the German Hospital Market: The Two Sides of the Coin.’ Health Policy (Amsterdam, Netherlands) 109(3): 301–310. Sinclair, Duncan. 2014. ‘“Undertakings” in Competition Law at the Public-Private Interface: An Unhealthy Situation.’ European Competition Law Review 35(4): 167–171. Spencelay, Emma, and Jennifer Dixon. 2014. Mergers in the NHS: Lessons from the Decision to Block the Proposed Merger of Hospitals in Bournemouth and Poole. London: Health Foundation. Stiglitz, Joseph E. 2000. Economics of the Public Sector (3rd ed.). New York, NY: W. W. Norton. Temple Lang, John. 2005. ‘Privatisation of Social Welfare: European Union Competition Law Rules.’ In Social Welfare and EU Law: Essays in European Law, edited by Michael Dougan and Eleanor Spaventa, 45–77. Oxford: Hart. Timmins, Nicholas. 1996. The Five Giants: A Biography of the Welfare State. London: Fontana Press. van de Gronden, Johan Willem, Erika Szyszczak, Ulla Neergaard, and M. Krajewski eds. 2011. Health Care and EU Law, Legal Issues of Services of General Interest. The Hague: T. M. C. Asser Press. Wanless, Derek, John Appleby, Anthony Harrison, and Darshan Patel. 2007. Our Future Health Secured?: A Review of NHS Funding and Performance. London: King’s Fund. Webster, Charles. 1998. The National Health Service: A Political History. Oxford: Oxford University Press. West, Peter A. 1998. ‘Market: What Market? A Review of Health Authority Purchasing in the NHS Internal Market.’ Health Policy 44(2): 167–183.

CHAPTER 18

Hea lthca r e A n titrust i n the U n ited States The Intersection of Regulation and Competition Thomas Greaney

1  Introduction Antitrust law has played a pivotal role in shaping institutional and professional arrangements in American healthcare. It has touched every sector including delivery, payment, medical equipment and technology, intermediaries, pharmaceuticals, and professional organizations. Historically, much of what can be broadly classified as “­competition policy” in healthcare is found in the application of traditional antitrust principles to the conduct and structure of provider and payer organizations rather than in any sweeping statutory enactments. Although some landmark legislation, such as the repeal of health planning statutes and adoption of the 1973 Health Maintenance Organization (HMO) law,1 removed some important barriers to the growth of managed care, the task of dealing with unacceptable practices and problematic market structures has been left to stand­ard antitrust law. Federal courts shoulder the task of interpreting and applying the extraordinarily broad commands of the principal federal antitrust statutes, the Sherman and Clayton Acts. Almost all states have enacted antitrust statutes that closely mirror the federal laws and are applied in state courts. Most significantly, private rights of action established 1  Health Maintenance Organization Act of 1973, 42 USC. §§ 300e-­300e-­17 (2006). The HMO Act established a federal qualifying system that provided seed money for new HMOs, required employers to offer an HMO option if the employer provided traditional indemnity coverage to its employees, and pre-­empted state laws and policies that inhibited HMO formation.

350   Thomas Greaney under the Clayton Act enable “any person . . . injured in his business or property by reason of anything forbidden in the antitrust laws” to obtain treble damages or obtain injunctive relief.2 Litigation by private plaintiffs, whom the Supreme Court has called “private attorneys general,”3 greatly outnumbers cases brought by public agencies. However, in some areas such as merger litigation, government actions are most significant because of restrictive standing requirements for private plaintiffs. In sum, antitrust law enforcement is highly pluralistic: litigation may be brought by the US Department of Justice (DoJ), the Federal Trade Commission (FTC), state attorneys general, and private plaintiffs. While subject to what are generally the same statutory actions, each is subject to somewhat different incentives in choosing their targets. One can discern, for example, differences based on the federal versus state focus of the FTC and DoJ compared to that of state attorneys general; private plaintiffs (and to some extent state attorneys general) are strongly motivated by the potential damage awards; government enforcers are keenly aware of the longer term effects of judicial precedents. As a general matter, antitrust litigation focuses primarily on conduct and structure. Conduct issues are typically covered by the Sherman Act’s prohibition on restraints of trade and acts of monopolization or attempted monopolization.4 Prominent among the myriad agreements in healthcare reached by antitrust cases are (1) horizontal restraints of trade including boycotts, price fixing, and market allocations involving competing hospitals, payers, physicians, and pharmaceutical companies; (2) exclusionary agreements involving the interactions of those groups; and, less commonly, (3) the acquisition or maintenance of market power by single entities and foreclosure of competition by vertical arrangements involving intersector combinations.5 Attention to concerns about market structure is the focus of antitrust challenges to mergers, joint ventures, affiliations, and networks that create market power or facilitate its exercise.6 Applying antitrust law to the healthcare industry entails some special problems.7 In particular, the peculiarities and distortions of healthcare markets often necessitate a sophisticated analysis in order to reach economically sound results. Among the many issues raised are whether market failures inherent in healthcare—particularly imperfect information and agency relationships—suggest more restrained approaches to applying antitrust law. whether the behavior of nonprofit healthcare providers conforms to 2  15 USC. §15(a). 3  Zenith Radio Corp. v. Hazeltine Res., Inc., 395 US 100, 130–131 (1969). 4  15 USC. §§1–2. 5  Detailed compendia of the large number of antitrust healthcare cases can be found in John J. Miles, Health Care and Antitrust Law (Thomson Reuters 2018 ed. 1992) and Barry Furrow, Thomas Greaney, Sandra Johnson, Timothy Stoltfuz Jost, & Robert Schwartz, Health Law Hornbook (West Academic Publishing 3d ed. 2014). 6  Id.; see also US Dept. of Justice & Federal Trade Commission, Horizontal Merger Guidelines (2010). 7  Classic discussions of the peculiarities of markets in medical care are Kenneth J. Arrow, Uncertainty and the Welfare Economics of Medical Care, 53 Am. Econ. Rev. 941 (1963) and Paul Starr, The Social Transformation of American Medicine 226–230 (1982) (arguing that professionalism is a major cause of market problems).

Healthcare Antitrust in the United States   351 t­raditional economic assumptions about competitors, and what impact widespread interventions by state and federal government have on the application of federal antitrust law.8 These complexities are the source of uncertainty in applying antitrust law to healthcare sectors. Moreover, many observers note that litigation before judges lacking expertise in economics or sophistication about healthcare compounds the risk of error. To some extent, the American system is aided by expertise housed in the FTC and the US DoJ Antitrust Division. Both agencies have devoted considerable resources to healthcare antitrust and have issued extensive policy pronouncements, guidelines, advisory opinions, and speeches to clarify enforcement policy and nudge legal doctrine toward developing appropriate standards. Notably, however, although the FTC Act vests the agency with broader statutory scope than does the Sherman Act (“unfair” methods of competition vs. “unreasonable restraints of trade”) and regulatory tools such as rule-­making and authority to conduct investigatory hearings, it has been quite guarded in exercising those powers.9 That reticence is emblematic of the American wariness of interfering with markets even when actions are justifiable as improving competition. Rather than survey the myriad issues arising from the application of American antitrust law to healthcare, this section will focus on two central themes. First, it will address the interplay of administrative regulation with antitrust law enforcement, noting particularly those areas where the goals of antitrust are in tension with regulation and discussing the doctrinal accommodations made to deal with that problem. Next, it will discuss the critical and delicate balance that antitrust law must strike in promoting competitively structured markets while taking into account market imperfections and hard-­ to-­measure outcome variables.

2  The Intersection of Regulation and Competition A wide spectrum of regulations affect healthcare payment and delivery in the United States: physician licensure, accreditation of hospitals and other health facilities, supervision of provider behavior by government payers, regulation of insurance industry practices, and certificate of need (CON) regulation requiring governmental approval to

8  A large academic literature addresses these issues. See, e.g., Peter J. Hammer & William M. Sage, Antitrust, Health Care Quality, and the Courts, 102 Colum. L. Rev. 545 (2002); Thomas L. Greaney, Chicago’s Procrustean Bed: Applying Antitrust Law in Health Care, 71 Antitrust L. J. 857 (2004) (courts’ pervasive neglect of market failures in healthcare antitrust cases systematically biases outcomes); Thomas Rice, The Economics of Health Care Reconsidered (1998) (questioning whether market forces will produce efficient or socially desirable outcomes). 9  Sandeep Vaheesan, Resurrecting “A Competitive Charter of Economic Liberty”: The Latent Power of the Federal Trade Commission, 19:3 U. Pa. J. Bus. L. 645 (2017).

352   Thomas Greaney open or add to healthcare facilities, to name just a few.10 Moreover, because federal, state, and local governments purchase an enormous share, perhaps as much as 45%, of all of health services, their influence over the industry is obviously substantial. Not surprisingly, as a condition of payment, these programs impose an immense number of regulatory requirements. While some of the most intrusive command and control forms of regulation, such as state rate-­setting and CON laws (requiring prior administrative approval for entry and large capital investments) have been removed or modified, a large body of law continues to govern the payment and delivery of healthcare.11 Yet despite the pervasiveness of regulation, market competition is the cornerstone of the American healthcare system. Indeed, a major theme of government policy over the past 30 years has been to increase reliance on market competition, even in sectors in which the government is the payer, such as Medicare and Medicaid. Major legislative adjustments enacted by Congress over the past 15 years, the Medicare Modernization Act, the Affordable Care Act, and the Medicare Access and CHIP Reauthorization Act (MACRA) are all fundamentally directed at encouraging the development of competitive delivery and payment systems, enabling consumer choice, and removing regulatory barriers to competition.12 Hence, rather than viewing health policy as facing dichotomous choices—competition or regulation—it is more accurate to see the two as complementary.13

2.1  Antitrust Law’s Attack on Self-­Regulation Private industry self-­regulation in healthcare has long been in the cross-­hairs of competition policy advocates. For many years provider self-­regulation and norms dictated the terms of market interactions. What Clark Havighurst called a “professional paradigm” dominated the policy and law of healthcare, giving physicians wide discretion over their professional practice and direct control over most institutional arrangements affecting their practice.14 After all, providing medical care entails a high level of expertise, and policymakers felt that physicians were in the best position to determine how care should 10  Robert Field, Government as the Crucible for Free Market Health Care: Regulation, Reimbursement and Reform, 159 U. Pa. L. Rev.1660 (2011). See also the other chapters in this section of the Handbook. 11  See generally Robert Field, Mother of Invention: How the Government Created Free-­Market Health Care (2013). 12  See Thomas L. Greaney, Competition Policy and Organizational Fragmentation in Health Care, 71 U. Pitt. L. Rev. 217 (2009) (discussing measures contained in the ACA and other legislation designed to address healthcare market imperfections). 13  See generally Kristin Madison, Peter D. Jacobson, & Gary Young, Health Policy and Regulation, in Health Care Management: Organization, Design and Behavior (6th ed. 2011); Thomas L. Greaney, The Affordable Care Act and Competition Policy: Antidote or Placebo?, 89 Or. L. Rev. 811 (2011); Robert Field, Government as the Crucible for Free Market Health Care: Regulation, Reimbursement and Reform, 159 U. Pa. L. Rev. 1660–1726 (2011). 14  See Clark C. Havighurst, The Professional Model of Care: Obstacle to Decentralization, 30 Jurimetrics J. 415 (1990).

Healthcare Antitrust in the United States   353 be delivered. A significant turning point, begun by antitrust litigation, was the direct challenge to private mechanisms that impeded competition. Cases brought by private plaintiffs and the FTC resulted in courts striking down numerous professional restraints. Most notable of the early cases affecting professional arrangements in healthcare was Goldfarb v. Virginia State Bar, which held that “learned professions” were not implicitly exempt from the antitrust laws and found the Sherman Act’s interstate commerce requirement satisfied with regard to legal services.15 Subsequent antitrust litigation removing a variety of private restraints of trade that had long inhibited competition spurred significant changes in the healthcare industry. “Ethical” restrictions imposed by professional associations on the pricing or contracting practices of their members were among the first norms to be challenged under the antitrust laws. Courts rejected a host of justifications offered for these practices (e.g., that professionals should be regarded as engaged in something other than a traditional commercial enterprise;16 that professional control over pricing was necessary to avoid degradation in quality,17 that desirable cross-­subsidies to low-­income individuals resulted from these practices,18 that contractual links to lay organizations would undermine professional judgments,19 and that consumers were ill-­equipped to evaluate professional services).20 Of particular concern were professional norms and organizational arrangements directly interfering with the price mechanism, including professional societies setting maximum or minimum prices through fee schedules, which constitutes illegal price fixing,21 and collective efforts to deny the cooperation necessary to develop innovative, cost-­reducing payment plans.22 Perhaps the most important such case was the FTC’s challenge to the American Medical Association’s restrictions on “contract practice.” These ethical standards opposed physicians agreeing to provide medical services for a specified price or on a salaried basis; companion norms prohibited physicians from accepting “inadequate” compensation or from “underbidding” other physicians.23 However, other, less direct restraints of trade have been found to require closer

15  Goldfarb v. Virginia State Bar, 421 US 773 (1975). 16  Id.; American Med. Ass’n v. F.T.C., 638 F.2d 443 (2d Cir. 1980), aff ’d, 455 US 676 (1982). 17  Nat’l Soc’y of Prof. Eng’rs v. United States, 435 US 679 (1978). 18  American Med. Ass’n v. United States, 317 US 519 (1943); see generally Reuben A. Kessel, Price Discrimination in Medicine, 1 J. L. Econ. 20 (1958). 19  American Med. Ass’n, 94 F.T.C. 701 (1979) (consent order). 20  F.T.C. v. Indiana Fed’n of Dentists, 476 US 447 (1986). 21  Goldfarb v. Virginia State Bar, 421 US 773 (1975); Arizona v. Maricopa Cty. Med. Soc’y, 457 US 332 (1982). 22  See, e.g., Group Health Co-­op. of Puget Sound v. King Cty. Med. Soc’y, 237 P.2d 737 (Wash. 1951); United States v. Halifax Hosp. Med. Ctr., 1981–1 Trade Cas. (CCH) 64,151 (M.D.Fla. 1981) (consent decree); Forbes Health Sys. Med. Staff, 94 F.T.C. 1042 (1979) (consent order). 23  American Med. Ass’n, 94 F.T.C. 701, 1011–1012 (1979), aff ’d as modified on other grounds, 638 F.2d 443 (2d Cir.1980), aff ’d on an equally divided court, 452 US 960 (1981). See also American Soc’y of Anesthesiologists, 93 F.T.C. 101 (1979) (consent order) (medical society rule restricting members from rendering services other than on a fee-­for-­service basis).

354   Thomas Greaney e­ xamination under antitrust law’s “Rule of Reason.”24 For example, in California Dental Association v. F.T.C.25 the Supreme Court found a professional society’s restrictions on price and quality advertising merited closer examination than lower courts had applied in view of plausible claims that such reviews prevented confusion and deceptive practices in a market beset with information imperfections. Likewise, plausible efficiency justifications, such as those associated with ethical bans on acceptance of contingent fees and referral fees by professionals, have been discussed by the courts.26 In such cases, courts recognize the potential pro-­competitive effects of informing customers and reducing search costs. The net impact of these changes, which has been characterized as a move from “provider-­driven” healthcare to “payer-­driven” healthcare,27 was to eliminate formal institutional and professional arrangements that formed barriers to competition and insulated both providers and patients from financial responsibility for healthcare services. As a result of these institutional changes and public policy initiatives, both public and private payment systems demanded that providers engage in active competition for contracts. This led to increasing patient cost-­sharing and assumption of risk by providers. The shift was not without tension, however. Physicians bridled under changes affecting their ability to make referrals, pressures to adopt payer-­directed processes and standards, and new requirements to compete on price. As a result, the 1990s saw the emergence of a new set of antitrust issues: development of countless physician networks that were little more than price-­fixing arrangements,28 lawsuits claiming that denials of hospital staff privileges29 and exclusions from payer networks30 were the product of anticompetitive conspiracies, and extensive consolidation in provider and payer markets.31 In this way, restraints rooted in professional norms were replaced by collusion and structural impediments that continue to evolve and challenge competitive conditions. More recently, attention has shifted to dealing with the problem of market power resulting from pervasive concentration, a topic discussed later. To be sure, the move away from the professional paradigm entailed a reduction in provider autonomy. “Organized medicine” no longer dictated the parameters of practice arrangements which became more subject to marketplace forces. But, at the same 24 The rule of reason refers to the methodology for judicial review of restraints of trade that are not so inherently anticompetitive and have some plausible pro-­competitive properties so as to require summary condemnation under the per se rule. Proof of market power or actual harmful effects is required. See Broadcast Music, Inc. v. Columbia Broadcasting System, Inc. 441 US 1 (1979). 25  Cal. Dental Ass’n v. F.T.C., 526 US 756 (1999). 26  The FTC has accepted consent decrees prohibiting the American Institute of Certified Public Accountants from barring contingent and referral fees. In re American Institute of Certified Public Accountants, Dkt. No. C-­3297, 55 Fed.Reg. 40944–02 (October 5, 1990). 27  David Dranove, Mark Shanley, & William D. White, Price and Concentration in Hospital Markets: The Switch from Patient-­Driven to Payer-­Driven Competition, 36 J. L. Econ. 179, 179–204 (1993). 28  See Furrow et al. supra note 5 at 743–749. 29  Id. at 732–743. See Thomas L. Greaney, Thirty Years of Solicitude: Antitrust Law and Physician Cartels, 7 Hous. J. Health L. & Pol’y. 189 (2007). 30  Id. at 762–763. 31  Id. at 782–803.

Healthcare Antitrust in the United States   355 time, an unconstrained market gives providers significant freedom to make clinical decisions and structure their business relationships. Indeed, physicians enjoy complete freedom of movement to new geographical sites and to choose among employment, partnerships, and a host of other affiliations, and they are relatively unconstrained in their ability to exercise clinical judgment in choosing among therapies, pharmaceutical interventions, and so forth. That said, government regulation and payment policies for government healthcare programs are highly influential in shaping business arrangements. As the following sections indicate, in some cases, regulations limit competition in order to advance other governmental interests and, in others, seek to nudge providers and payers to adopt more cost-­effective and clinically beneficial arrangements.

2.2  Regulatory Impediments to Competition Although antitrust enforcement has been largely successful in eliminating ­self-­regulatory practices that restrain trade, competition is inhibited by a large number of state and ­federal laws, including licensing regulations and reimbursement rules. These laws, ­discussed here, are largely beyond the reach of antitrust law because of interpretive ­principles confining its enforcement to private agreements. However, considerable law has developed for determining whether an action of a state governmental body is truly the act of the government and hence immune from antitrust challenge. Under the so-­called state action doctrine, a two-­pronged analysis applies. First, the action must be “one clearly articulated and affirmatively expressed as state policy”; second, the policy must be “actively supervised by the state.”32 In the case of political subdivisions such as municipalities or other local governmental entities, the latter requirement need not be met.33 While the Supreme Court has denied state action doctrine protection to professional boards composed of a majority of active practitioners,34 and other cases have limited the scope of the doctrine where the state fails to actively supervise the decisions of state agencies,35 there is considerable federal and state regulation that is beyond the reach of antitrust enforcement, as discussed in the next section.

2.3  Supply-­Side Regulations State law governs professional licensure and defines the scope of practice for complementary and alternative providers, such as nurse midwives, nurse practitioners, physician assistants, and many others. Considerable literature suggests that these laws and 32  City of Lafayette v. Louisiana Power & Light Co., 435 US 389 (1978); Crosby v. Hosp. Auth. Of Valdosta and Lowndes Counties, 93 F.3d 1515 (11th Cir. 1996). 33  Hallie v. Eau Claire, 471 US 34, 46 (1985). 34  N.C. State Bd. of Dental Exam’r v. F.T.C., 574 US 494 (2015). 35  F.T.C. v. Ticor Ins. Co., 504 US 621 (1992).

356   Thomas Greaney regulations impose unnecessary costs and limit competition in healthcare markets while lacking sound evidence of preventing harm to patients. For example, restrictive scope-­of-­practice regulations for Advanced Practice Registered Nurses (APRNs) have been found to reduce access and quality and increase costs, while more even-­handed regulation can allow nurses to help meet the growing primary care shortage.36 State medical boards may also inhibit rivalry by unnecessarily limiting the conditions under which practitioners can compete. In a recent case, a state medical board issued rules requiring in-­person visits before practitioners could conduct examinations remotely and render diagnoses via telemedicine,37 requirements widely perceived as clinically unwarranted and serving the interests of the traditional providers represented on the state board. Although the rule was subsequently withdrawn after Teledoc sued the medical board under the Sherman Act, such rules would gain antitrust immunity if adopted by an independent board.38 Finally, state CON laws which reduce the possibility of new entry or competitive growth by rival hospitals have been criticized by the FTC and academic commentators but are still in force in more than half the states.39

2.4  Reimbursement and Provider Network Regulation Certain government and private reimbursement policies create incentives for consolidation that are not justified by quality or administrative rationales. For example, the Medicare payment system reimburses hospitals for “facility fees” in outpatient settings at a higher rate than it reimburses physicians for the same services provided in private practice. This creates a strong incentive for hospitals to acquire physician practices, a phenomenon discussed later that increases the cost of medical services without enhancing quality of care. Similarly, a reimbursement policy known as the 340B program ­creates an artificial incentive for physicians who administer very expensive drugs, such as oncologists, to become employed by hospitals.40 Other payment and reimbursement policies also inhibit competition. The Affordable Care Act amended federal anti-­referral law to restrict the ability of physicians to own 36  The FTC has vigorously opposed state legislation and regulations inhibiting the ability of APRNs to practice to the full extent of their licensure. See FTC Staff Comment Before the Mass. House of Representatives Regarding House Bill 2009 Concerning Supervisory Requirements for Nurse Practitioners and Nurse Anesthetists (Jan. 2014); Comment from FTC Staff to the Hon. Theresa W. Conroy, Conn. House of Representatives (Mar. 19, 2013); Written Testimony from FTC Staff to Subcomm. A of the Joint Comm. on Health of the State of W. Va. Legislature (Sept. 10–12, 2012); 37  Teladoc, Inc. v. Texas Med. Bd., 453 S.W.3d 606 (Tex. App. 2014); Abby Goodnough, Texas Medical Panel Votes to Limit Telemedicine Practices in State, N.Y. Times, Apr. 10, 2015. 38  Katie Dvorak, Court Ruling Spells Possible Victory for Teladoc Against Texas Medical Board, FierceHealthIT (2015), http://www.fiercehealthit.com/story/ court-­ruling-­spells-­possible-­victory-­teladoc-­against-­texas-­medical-­board/2015-­06-­01. 39  ftc, Improving Health Care: A Dose of Competition ch. 1, at 22; ch. 8 at 1–5 (2004). 40  See Thomas L. Greaney & Barak D. Richman, Promoting Competition in Healthcare Enforcement and Policy: Framing an Active Policy Agenda (Am. Antitrust Inst., White Paper Series on Competition in the Delivery and Payment of Healthcare Services), https://ssrn.com/abstract=3267556.

Healthcare Antitrust in the United States   357 hospitals. While concerns about “cherry picking” of healthy, insured patients are ­warranted, an outright ban was unnecessary and eliminated a much-­needed source of competition for dominant hospitals. Other laws interfere with competitive contracting by inhibiting the ability of payers to contract selectively and thereby induce rivalry among payers. For example, state “any willing provider” laws force payers to contract with physicians and thereby undermine incentives for price and quality competition. Likewise, overly restrictive network adequacy laws can reduce the ability of payers to design networks composed of providers they believe will provide the best quality at an affordable price.41 In sum, the regulatory context in which providers and payers operate is a critical determinant of the effectiveness of market competition in healthcare. In some instances, antitrust enforcement is constrained by state and federal regulations that impede market competition. Also, as discussed in the following section, antitrust doctrine offers only limited avenues for addressing markets that have been allowed to become overly concentrated.

3  Market Structure A large body of economic research indicates that concentration in hospital and physician markets translates into higher prices with neutral or even negative effects on qual­ity.42 For example, studies show price increases ranging from 20% to 40% occur where hospitals merge in concentrated markets.43 Indeed the pricing power of dominant providers has been shown to be a leading driver of the persistently high cost of healthcare in America.44 Despite decades of antitrust litigation and regulatory interventions, the problem has worsened and threatens to undermine the benefits of market-­based ­policies. While antitrust enforcement has been reinvigorated in recent years and has established legal precedent that constrains anticompetitive horizontal mergers, market concentration remains a serious impediment to effective competition. Since at least the early 1980s, the structure of hospital markets has been a top concern of antitrust enforcers under both Republican and Democratic administrations. The agencies have focused primarily on hospital markets, bringing 22 challenges to acute care hospital mergers in federal court and causing numerous others to abandon plans 41  Mark A. Hall & Paul B. Ginsburg, A Better Approach to Regulating Provider Adequacy (2017), https://www.brookings.edu/research/a-­better-­approach-­to-­regulating-­provider-­network-­adequacy/. 42  Martin Gaynor, Examining the Impact of Health Care Consolidation, testimony before Subcomm. on Oversight and Investigations of the H. Comm. on Energy and Commerce (2018). 43  Martin Gaynor & Robert Town, The Impact of Hospital Consolidation-­Update. Robert Wood Johnson, The Synthesis Project (2012). 44  Zach Cooper, Stuart V. Craig, Martin Gaynor, & John Van Reenen, The Price Ain’t Right? Hospital Prices and Health Spending on the Privately Insured (Nat’l. Bureau of Econ. Res., Working Paper No. 21815, 2015).

358   Thomas Greaney after government inquiries began. However, hospital merger enforcement has experienced something of a roller-­coaster ride in court over the years which is partly responsible for the levels of concentration experienced today. Hospital markets have become increasingly concentrated, with the average hospital market in the mid-­2000s having only three equal-­size systems and many large markets being dominated by a single, “must have” hospital system.45 After initially winning a number of suits in federal court in the 1980s and early 1990s, state and federal antitrust enforcers suffered seven defeats in litigating hospital merger cases.46 What followed was a 7-­year period of quietude during which no cases were filed, and hospital merger activity accelerated. As mergers went unchallenged, hospital market concentration grew considerably. However, following a series of retrospective studies to evaluate the outcomes in several consolidations, the FTC and state attorneys general have been successful in obtaining federal court injunctions blocking hospital and physician mergers.47 Importantly, these decisions and others have clarified legal standards and established what are likely to be enduring precedents. Among other things, they clarified that provider and insurer markets are highly localized and thereby insulated from competition from nearby markets, rejected arguments that market power will be checked by the countervailing power of large or sophisticated buyers (employers or insurers), and declined to accept arguments that uncertainties arising from rapidly changing market conditions undermine inferences of market power or that consolidation is essential to achieve beneficial integrative efficiencies.48 Moreover, the federal antitrust agencies and state attorneys general expanded their focus on structural conditions to other areas. In 2016, the US Justice Department obtained injunctions that blocked two mergers which involved four of the nation’s five largest health insurance firms.49 In addition, the FTC has successfully challenged several horizontal mergers involving physician practices.50 That said, American healthcare nevertheless has a serious concentration problem. Consider the following data on market structure: Commercial Insurance: • 69% of insurance markets are highly concentrated, and the four largest insurers have 83% of commercial insurance nationally.51 45  See Gaynor testimony supra note 42 (noting that “many areas of the country are dominated by one or two hospitals with no close competitors” and citing Boston, Cleveland, Pittsburg, and San Francisco as examples). 46  See Miles supra note 5. 47  See e.g., F.T.C. v. Advocate Health Care Network, 841 F.3d 460 (7th Cir. 2016); F.T.C. v. Penn State Hershey Med. Ctr., 838 F.3d 327 (3d Cir. 2016); ProMedica Health Sys., Inc. v. F.T.C., 749 F.3d 559, 571 (6th Cir. 2014). 48  See Greaney & Richman, AAI White Paper supra note 40. 49  United States v. Anthem, Inc., 855 F. 3d 345 (D.C. Cir. 2017); United States v. Aetna, Inc., 240 F. Supp. 3d 1 (D.D.C. 2017). 50  St. Alphonsus Med. Ctr.-Nampa & F.T.C. v. St. Luke’s Health Sys., 778 F.3d 775, 793 (9th Cir. 2015); F.T.C. & N. D. v. Sanford Health, Case No. 1;17-­cv-­00133 (D.N.D. Dec. 13 2017); see also In the Matter of Renown Health, Fed. Trade. Comm’n. Dkt. No. C-­4366 (Dec. 4, 2012) (consent order). 51  Am. Med. Ass’n, Competition in Health Insurance: A Comprehensive Study of US Markets (2017).

Healthcare Antitrust in the United States   359 • In half of all markets, two of the largest insurers have more than 70% of the market.52 Physician Services • 65% of metropolitan areas (MSAs) have highly concentrated specialty markets; 39% have concentrated primary care markets.53 Hospitals • 90% of inpatient acute care hospital markets are highly concentrated.54 • Many large metropolitan markets (e.g., Boston, Pittsburgh, San Francisco) are dominated by one or two hospitals.55 Pharmacy Benefit Management • The three largest pharmacy benefit managers control more than 70% of the national market.56 While payer and provider market concentration has undoubtedly created pricing power in each sector, antitrust law is unlikely to provide a remedy. As a general matter, antitrust has little to say about market power achieved by growth or accruing from past mergers that are too stale to tackle despite the fact that such firms may charge supra-­ competitive prices or otherwise injure consumers. Challenges to consummated transactions are notoriously difficult to mount, particularly because of difficulties inherent in fashioning relief or “unscrambling the egg.”57 Moreover, antitrust law only condemns monopolists that inappropriately obtain or maintain market power, and, even in those cases, plaintiffs may settle for promises to change their behavior58 rather than divestiture of assets. Remedies are to be found elsewhere, such as price caps on dominant providers.59

52  Martin Gaynor, Examining the Impact of Health Care Consolidation, statement before the Energy and Commerce Oversight Comm., US House of Representatives, 7 (February 14, 2018). 53 Brent D. Fulton, Health Insurance Market Concentration in the United States: Trends in the United States and Policy Responses, 36 Health Aff., no. 9, 2017, at 1530–1538. 54  Id. 55 Gaynor supra note 52 at 6. 56  Id. Pharmacy benefit managers are firms that manage prescription drug benefits on behalf of health insurers, Medicare Part D drug plans, large employers, and other payers. 57  F.T.C. v. Penn State Hershey, 838 F.3d 327 (3d Cir. 2016); F.T.C. v. University Health Inc., 938 F.2d 1206, 1217 n. 2 (11th Cir. 1991). 58  See Steven C. Salop, Reinvigorating Vertical Merger Policy, 127 Yale L. J. 1962, 1962–1994 (2018). 59  See Thomas L. Greaney, Coping with Concentration, 36 Health Aff., no. 9, 2017, at 1564; Aaron Baum et al., Health Care Spending Slowed After Rhode Island Applied Affordability Standards to Commercial Insurers, 38 Health Aff., no. 2, 2018.

360   Thomas Greaney Another area in need of close attention is vertical consolidation (e.g., when a hospital combines with a physician practice). As a general matter, antitrust law has been excessively tolerant of vertical mergers, and case law and guidance is sparse. A number of prominent economists have sharply criticized enforcers’ laissez-­faire approach, noting that the preconditions underlying the Chicago School’s60 analysis of vertical combinations “rarely hold” and “can obscure how a particular merger may raise real competitive concerns.”61 Notably, the FTC has recently conducted hearings regarding the need to update its vertical merger guidelines to better align its guidance and enforcement policy with modern economic thinking. In several health markets, the number of vertical acquisitions has grown rapidly, but antitrust enforcement has been lacking. For example, the government has not challenged a single hospital acquisition of a physician practice based on a vertical foreclosure theory and recently has approved the combination of the nation’s largest pharmacy benefit management company with the country’s largest health insurer. The core concern with hospital acquisitions of physician practices is that they may foreclose rival hospitals and potential entrants in the hospital services market, preventing such competitors from obtaining a sufficient base of patients and depriving access to physicians to admit, treat, or refer to their facilities. The economic harm visited on consumers may flow from either (1) elimination of rival hospitals from competing in a ­hospital service market so as to increase the market power of the integrated hospital in a service market or (2) foreclosure of effective competition by non–vertically consolidated hospitals, causing higher average costs, higher prices, and less formidable ­competitive challenges to vertically integrated hospitals with market power. A number of studies document the harms arising from hospital acquisition of physician practices. Not only are hospital acquisitions associated with higher prices for physician services among commercial insurers,62 but higher costs spill over to government programs as well.63

60  The Chicago school of antitrust law refers to a generally conservative approach to applying antitrust law that is based on a faith in the efficiency of markets. For the classic explication by Chicago’s leading voice, see Richard Posner, The Chicago School of Antitrust Analysis, 127 U. Penn L. Rev. 925 (1979); for a critical analysis see Herbert Hovenkamp, Antitrust After Chicago, 84 U. Mich. L. Rev. 213 (1985). 61  Steven C. Salop & D. P. Culley, Potential Competitive Effects of Vertical Mergers: A How-­To Guide for Practitioners (Dec. 8, 2014), https://ssrn.com/abstract=2522179. 62  Laurence C. Baker, M. Kate Bundorf, & Daniel P. Kessler, Vertical Integration: Hospital Ownership of Physician Practices Is Associated with Higher Prices and Spending, 33 Health Aff., no. 5, 2014, at 756; James Robinson & Kelly Miller, Total Expenditures per Patient in Hospital-­Owned and Physician-­Owned Physician Organizations in California, 312 jama, no. 16, 2014, at 1663 (finding hospital-­owned physician organizations had 10–20% higher total expenditures/patient than physician-­ owned organizations); Cory Capps, David Dranove, & Christopher Ody, The Effect of Hospital Acquisitions of Physician Practices on Prices and Spending 1 (Inst. for Pol’y Res., Northwestern University, Working Paper No. WP-­15-­02, 2015) (finding that vertical integration was associated with a 13.7% increase in physician prices). 63  Medicare Payment Advisory Commission (MedPAC), Report to the Congress: Medicare and the Health Care Delivery System, ch. 10 at 299 (2017).

Healthcare Antitrust in the United States   361

4  Future Directions: The Case for Invigorating Antitrust Enforcement in Healthcare American antitrust has left several important avenues of analysis untouched. For example, the effect of monopoly on the distribution of wealth plays no part in doctrine and is largely ignored even by those advocating a “populist” approach to revising the law. While market power is readily seen as a mechanism for transferring wealth from the working and middle classes to the top of the income distribution or, as one commentator put it, turning “disposable income of the many into capital gains, dividends and executive compensation for the few,”64 conventional antitrust analysis does not explicitly take redistribution into account. Indeed, Chicago School welfare analysis counsels indifference to wealth transfers from consumers to buyers. Although rarely acknowledged, a strong case can be made that market power in healthcare has greater redistributive effects than elsewhere in the economy.65 The factors contributing to this are the combined effects of imperfect agency, information asymmetry, moral hazard, and market power. Insurance dulls patient sensitivity to price increases, provider market power and information deficits enable physicians and hospitals to induce demand for unnecessary services, and conflicting incentives cloud the physician–patient agency relationship. As Clark Havighurst and Barak Richman summarized this effect, The extraordinary profits that health insurance makes available to powerful sellers are earned mostly at the expense not of direct purchasers—insurers or patients— but of consumers bearing the cost of insurance.66

Scholars and policy experts have long noted the strong association of market power and favorable regulatory policies.67 This is certainly the case in the healthcare sector. Moreover, market participants have benefited not only from the variety of regulatory favors described earlier, but from assumptions in antitrust and other legal doctrines that assume a well-­functioning market. The persistent failure to adapt legal analyses to the 64  Lina Kahn & Sandeep Vaheesan, Market Power and Inequality: The Antitrust Counterrevolution and Its Discontents, 11 Harv. L. & Pol’y Rev. 235 (2017). 65  See, e.g., Clark C. Havighurst & B. D. Richman, Who Pays—Who Benefits—Unfairness in American Health Care, 25 Notre Dame J. L. Ethics & Pub. Pol’y 493 (2014). 66  Clark C. Havighurst & Barak D. Richman, The Provider Monopoly Problem in Health Care, 89 Ore. L. Rev. 847, 862 (2011). 67  See Robert Pitofsky, The Political Content of Antitrust, 127 U. Pa. L. Rev. 1051 (1979); for a more recent analysis of the interplay of market power and regulation see Lina Kahn & Sandeep Vaheesan, Market Power and Inequality: The Antitrust Counterrevolution and its Discontents, 11 Harv. L. & Pol’y Rev. 235 (2017) (firms in concentrated markets obtain political influence to support favorable policy).

362   Thomas Greaney peculiar economics of competition in the healthcare sector has had important adverse consequences for consumers. As I have described elsewhere, courts have frequently oversimplified antitrust analysis by adopting plain vanilla, Chicago School assumptions about markets while failing to incorporate the effects of market imperfections in their analyses of health markets.68 As a result, they have rendered decisions finding extraordinarily large geographic markets for basic acute care hospital services by failing to appreciate the heterogeneity of demand for care and the fact that consumers exhibit different preferences for travel for healthcare than for other services or products.69 Other cases refused to recognize supply-­side heterogeneity, failing to appreciate that mergers of “must have” hospitals may create risks of anticompetitive effects.70 While recent decisions have corrected those errors, extraordinary market concentration has resulted.71 That brings us to what is the most challenging issue facing American healthcare policy: how to contain the potentially ruinous escalation in healthcare costs while moving toward universal coverage. As this book goes to press, America is once again debating (as it has during every presidential campaign over at least two decades) potential reforms of the healthcare system. If, as seems likely, the nation is going to persist in relying on market competition rather than a single-­payer framework, the central question will be whether government can develop tools that curb cost escalation. Two pertinent facts should be kept in mind. First, the principal source of America’s high healthcare costs is not excessive utilization, demographic change, or malpractice law. The real cause was well captured by the title of an article written 15 years ago by the much esteemed economist, the late Uwe Reinhart, “It’s the Prices Stupid.”72 The point is that the prices charged at every level—hospitals, physicians, diagnostic and outpatient centers, equipment suppliers, intermediaries, administrators—are much higher (often by multiples of two to four) than in other nations. The second critical point is that excessive prices are, for the most part, the product of market power. Economic studies ­confirm that market concentration, especially in hospital and physician markets, creates 68 Thomas L. Greaney, Chicago’s Procrustean Bed: Applying Antitrust in Health Care, 71 Antitrust L. J. 857 (2004). An emerging critique of conventional antitrust wisdom applies the teaching of behavioral economics. See Maurice E. Stucke, Behavioral Economics at the Gate: Antitrust in the Twenty-­First Century, 38 Loy. U. Chi. L. Rev. 513, 527–528 (2007); see generally Korobkin & Ulen, Law & Behavioral Science: Removing the Rationality Assumption from Law & Economics, 88 Cal. L. Rev. 1051, 1051–1102 (2000); Christine Jolls et al., A Behavioral Approach to Law & Economics, 50 Stan. L. Rev. 1471 (1998). For applications in several areas of healthcare law see, Thomas L. Greaney, Economic Regulation of Physicians: A Behavioral Economics Perspective, 53 St. Louis U. L. J. 1189 (2009). 69  Cory S. Capps et al., The Silent Majority Fallacy of the Elzinga-­Hogarty Criteria: A Critiques and New Approach to Analyzing Hospital Mergers (Nat’l. Bureau of Econ. Res., Working Paper No. w8216, 2001). 70  United States v. Long Island Jewish Med. Ctr., 983 F. Supp. 121 (E.D.N.Y. 1997). 71  See e.g., Kenneth L. Danger & H. E. Frech, Critical Thinking About “Critical Loss” in Antitrust, 46 Antitrust Bull. 339 (2001); James Langenfeld & Wenqing Li, Critical Loss in Evaluating Mergers, 46 Antitrust Bull. 299 (2001); Gregory Vistnes, Hospitals, Mergers, and Two-­Stage Hospital Competition, 67 Antitrust L. J. 671 (2000). 72  Gerard Anderson & Uwe Reinhardt, It’s the Prices, Stupid: Why the United States Is So Different from Other Countries, 22 Health Aff., no. 3, 2003; see also Uwe E. Reinhardt, Priced Out: The Economic and Ethical Costs of American Health Care (2019).

Healthcare Antitrust in the United States   363 bargaining leverage that results in monopoly returns and ultimately escalating ­insurance premiums and government program costs.73 Is more effective enforcement of existing antitrust law the answer? Sadly, no. While antitrust law can and should prevent excessive concentration, as discussed earlier, it has failed to do so in provider markets. As noted earlier, significant practical difficulties in fashioning relief make antitrust challenges to consummated mergers almost impossible. Furthermore, antitrust law offers only limited relief against the extant power of monopolies and oligopolies: it only acts—and then infrequently—where those with market power have obtained or expanded their power by improper means.74 While occasionally such challenges are mounted—the state of California and private plaintiffs have initiated a case against a dominant hospital system’s use of “all or nothing contracts” and other measures to stifle competition75—they rarely seek to break up the dominant firm, relying instead on often ineffective behavioral commitments to curb future abuses. Thus, we close with this seeming paradox: the prospects for efficient competition in American healthcare markets rest on adopting effective new regulations. Regulation targeted at controlling the pricing of local provider monopolies and encouraging competition among payers and other intermediaries is what is needed. Whether America is capable of reaching a consensus on reaching a middle ground between competition and regulation is, to say the least, an open question. A brief postscript: this chapter was written before COVID-­19 struck the United States. Enormous adjustments to the American delivery and payment systems have occurred, many of which are likely to continue after the nation emerges from the crisis. An important question for American policymakers will be the impact of the likely surge of hospital bankruptcies, acquisitions by large systems of smaller hospitals and physician practices, and changes in the competitive structure of insurance markets.76 With provider and payer markets already highly concentrated, the challenge for ­antitrust enforcement will be the task of preserving market structures conducive to competition while allowing efficiency-­improving consolidation to go forward. This is a problem antitrust law has faced in prior years, and antitrust doctrine affords judges sufficient flexibility to interpret the law in light of changing market conditions.77 At the same time, antitrust is relatively powerless to deal with extreme market concentration absent improper conduct to obtain or maintain monopoly power. Finally, to hazard a prediction: the next round of health reform in the United States will include a heavy dose of price regulation even if the market-­oriented health system survives.

73  See e.g., Office of the Att’y General of Mass., Examination of Health Care Cost Trends and Drivers (March 16, 2010); Zack Cooper et al., supra note 44; Gaynor supra note 42. 74  See Greaney, Coping with Concentration, supra note 59. 75  State of Cal., ex rel. Xavier Becerra v. Sutter Health, CGC-­18-­565398, 2018 WL 1584066 (Cal. Super. Ct. Mar. 29, 2018). 76  See Jesse Drucker et al., Wealthiest Hospitals Got Billions in Bailout for Struggling Health Providers, N. Y. Times (May 25, 2020). 77  See Darren Bush, Too Big to Bail: The Role of Antitrust in Distressed Industries, 77 Antitrust L. J. 277 (2010).

chapter 19

I n troduction to Fr au d a n d A buse L aw Tracey A. Elliott and Joan H. Krause

Fraud is universal. As economists have long known, “[f]or every clever person who goes to the trouble of creating an incentive scheme, there is an army of people, clever and otherwise, who will inevitably spend even more time trying to beat it.”1 Where there is an opportunity to profit by manipulating the rules, it should come as no surprise that people will do so. This is equally true in healthcare. The sheer magnitude of dollars flowing through the global healthcare system—estimated at $7.35 trillion in 2013—creates an attractive target.2 The myriad interactions among patients, healthcare providers, and payers create ample opportunity to siphon money out of the system, while the opacity of the system makes such activities difficult to detect. These factors give rise to several interrelated observations about healthcare fraud: it is universal, it is dynamic, and it is persistent.

1  Healthcare Fraud: Universal, Dynamic, and Persistent First, healthcare fraud is universal. Experience has demonstrated that fraud can, and does, flourish in virtually any healthcare system.3 One analysis estimated that 1  Steven D. Levitt & Stephen J. Dubner, Freakonomics: A Rogue Economist Explores the Hidden Side of Everything 24–25 (2005). 2  Jim Gee & Mark Button, The Financial Cost of Healthcare Fraud 2015: What Data from Around the World Shows (2015), § 4.5, https://www.pkf.com/media/31640/PKF-­The-­financial-­cost-­of-­fraud-­2015.pdf 3  European Commission, Study on Cross-­Border Cooperation: Capitalising on existing initiatives for cooperation in cross-­border regions: Final Report (2018); European Commission, Study on Corruption in the Healthcare Sector, HOME/2011/ISEC/PR/047-­A2 (2013), https://ec.europa.eu/home-­affairs/sites/homeaffairs/ files/what-­is-­new/news/news/docs/20131219_study_on_corruption_in_the_healthcare_sector_en.pdf

366   Tracey A. Elliott and Joan H. Krause fraud-­related losses in the United Kingdom, the United States, France, Belgium, the Netherlands, Australia, and New Zealand from 1997 to 2013 ranged between 0.6% and 12.5%, with an average loss rate of 6.19%.4 That is not to say, however, that healthcare fraud occurs identically in every country. At its core, healthcare fraud is opportunistic: the locus of fraud depends, first and foremost, on how healthcare is financed. In other words, “the way we pay people affects the way they cheat.”5 As the structure of healthcare finance differs among countries, fraud will be expressed in different ways. Second, healthcare fraud is dynamic, responding and adapting to changes in the healthcare system. For example, there has been long-­standing concern that paying providers a fee for each service rendered encourages overtreatment or at least overbilling of care. One way to counteract this incentive is to pay for care on a more global basis, such as a per-­patient fee or a lump-­sum payment for a bundle of related services. But global payment creates its own incentives: the incentive to provide less (and less expensive) care, to avoid patients likely to need extensive or expensive treatment, to falsify the number of patients for whom payment should be made, or to submit additional claims for services that fall outside the bundled rate. As the United States experience with managed care in the 1980s and 1990s demonstrated, changing the payment system does not eliminate healthcare fraud: it simply changes the way in which fraud occurs.6 Third, healthcare fraud is persistent. If different payment systems invite different types of fraud and fraud is dynamic enough to adapt to payment changes, it is no wonder that fraud enforcement always seems to be one step behind. Despite increasingly more sophisticated anti-­fraud enforcement efforts, government entities often find themselves chasing after those who already have taken money out of the system. While discussions increasingly are framed by the goal of fraud prevention, anti-­ fraud strategies largely have relied as much on increasingly harsh forms of ex post punishment as they have on improving ex ante detection—often without noticeable effect.7

4  Gee & Button, supra note 2 at § 4.1. 5 Pamela S. Bucy, Health Care Reform and Fraud by Health Care Providers, 38 Vill. L. Rev. 1003, 1049 (1993). 6  See Sharon L. Davies & Timothy Stoltzfus Jost, Managed Care: Placebo or Wonder Drug for Health Care Fraud and Abuse?, 31 Ga. L. Rev. 373, 389 (1997). Similarly, changing patterns of fraudulent claims emerged following changes to National Health Service (NHS) dentistry contracts in the UK in 2006, which led to dentists being paid in accordance with three categories of treatment rather than for each procedure. NHS, “NHS dental services in England” (2009); Department of Health (DoH), “Over £70 Million Wasted on Fraud Under Current Dental Contract” (2012) May 8, https://www.gov.uk/ government/news/over-­70-­million-­wasted-­on-­fraud-­under-­current-­dental-­contract (last accessed Jan. 7, 2018). 7  Graham Brooks et al., The Scale of Health-­Care Fraud: A Global Evaluation, 25 Security J. 76, 79–80 (2012).

Introduction to Fraud and Abuse Law   367

2  Addressing Healthcare Fraud The situation is further complicated by the variability in the ways in which countries address the problem of healthcare fraud. This variability makes it difficult even to describe the full universe of fraudulent healthcare activities, let alone to conceptualize a common model. The problems coalesce into three types of considerations: definitional, operational, and ethical. First and foremost, healthcare fraud suffers from a basic definitional problem: What, exactly, do we mean by the term? There is wide variation in the types of activities identified as fraudulent. In the United States, for example, it is common to describe this area of law as “healthcare fraud and abuse.”8 In this context, “fraud” involves actions taken with the intent to deceive, coupled with another’s detrimental reliance on the misrepresentation; “abusive” behavior lacks the intent to deceive, instead involving activities such as creative interpretation of payment rules in self-­interested ways that may violate the spirit (if not actually the letter) of the law.9 Thus, “abuse” functions as a financial concept, rather than the more colloquial meaning of the term to describe the physical abuse or neglect of vulnerable patients. Unless such mistreatment is accompanied by improper requests for payment, the US legal system addresses these concerns through laws outside of the “fraud and abuse” context.10 In Europe, the differences between state laws means that there is no single “European” legal or regulatory approach, with terms such as “fraud” and “abuse” not always being used consistently.11 For example, the European Healthcare Fraud and Corruption Network (EHFCN)12 found that there was semantic inexactitude and confusion in the use of the term “fraud,” with there being a “compulsive tendency” for members instead to use euphemisms such as “wrongful billing” and “inappropriate use” in counter-­fraud reports and when states were exchanging information.13 To encourage the use of a common language between states and counter-­fraud organizations in Europe, the EHFCN has attempted to clarify terminology by creating the EHFCN Waste Typology Matrix for

8  See, e.g., Health Insurance Portability and Accountability Act of 1996, Pub. L. No. 104–191, tit. II (“Preventing Health Care Fraud and Abuse; Administrative Simplification”), 110 Stat. 1936, 1991 (1996). 9  Jerry L. Mashaw & Theodore R. Marmor, Conceptualizing, Estimating, and Reforming Fraud, Waste, and Abuse in Healthcare Spending, 11 Yale J. Reg. 455, 463 (1994). 10  See, e.g., N.C. Gen. Stat. §§ 108A.99 et seq. (Protection of the Abused, Neglected, or Exploited Disabled Adult Act). 11 EHFCN, The EHFCN Waste Typology Matrix, http://www.ehfcn.org/what-­is-­fraud/ ehfcn-­waste-­typology-­matrix/ 12  The EHFCN was created in 2005 as a not-­for-­profit organization dedicated to combatting fraud, corruption, and waste in the healthcare sector across Europe. Its members are healthcare and counter-­ fraud organizations in Europe. See www.ehfcn.org/who-­we-­are/ 13 EHFCN, supra note 12.

368   Tracey A. Elliott and Joan H. Krause the use of its members, which defines regulatory infringements according to a scale that reflects the degree of intention involved. At the most serious end of the scale, “corruption” is defined as “illegally obtaining a benefit of any nature by abuse of power with third party involvement,” followed by “fraud” (“illegally obtaining a benefit of any nature by intentionally breaking a rule”) and “abuses” (“unjustly obtaining a benefit of any nature by knowingly stretching a rule or guideline or by taking advantage of an absence of rule or guideline”), with “errors” (“unjustly obtaining a benefit of any nature by unintentionally breaking a rule or a guideline”) being the least serious infractions.14 The problem of incorrect billing for healthcare may be used to illustrate these distinctions in practice. For example, corruption would be committed by a doctor who billed for more medical products than were actually used whilst receiving bribes from a pharmaceutical company. On the other hand, the erroneous billing by a doctor of a daytime consultation as a nighttime consultation would amount to fraud if intentional but error if unintentional. Abuse would include such behaviour as taking advantage of a lacuna in billing rules to assimilate noninsured care into the payment system by using a code for insured care.15 Similar distinctions between fraud and lesser financial contraventions, based on the level of intention or knowledge of the party concerned, can be found in the anti-­fraud policy of Assurance Maladie (the French universal sickness insurance scheme), which uses the term “fraud” to denote illegal acts of deception intentionally committed to wrongfully obtain monetary gain or an unjustified benefit from the state or health insurance organization or pension and occupational health insurance fund; “abusive activities” (“d’activités abusives”) to denote the repeated use of an item, service, or prerogative contrary to acceptable levels; and “negligent activities” (“d’activités fautives”) to indicate unintentional activities which, although not amounting to fraud and abuse, are legally irregular and repeatedly committed.16 In the context of billing for treatment, billing for care not provided would amount to fraud, whilst prescription of treatment in excess of the patient’s treatment needs would amount to abusive activity, and the unintentional infringement of billing rules would amount to a wrongful activity.17 By contrast, in the UK, the discourse focuses on healthcare “fraud,” with the term “abuse” tending to be used to describe a form of fraud where an individual’s rights in relation to their financial affairs or assets are violated, particularly where that individual may be regarded as belonging to a vulnerable group such as the elderly or those with cognitive impairments.18 This approach may be found in the Fraud Act 2006, which 14 Id. 15  EHFCN, “Lexicon of Infringements by Healthcare Providers in Healthcare Provision Considered to Be ‘WASTE’ and Suggested Enforcement Actions,” www.ehfcn.org/what-­is-­fraud/ ehfcn-­waste-­typology-­matrix/ 16  L’Assurance Maladie, CIR:1/2012, www.mediam.ext.cnamts.fr/ameli/cons/CIRCC/2012/ CIR-­1-­2012.PDF. Cf. Direction de L’audit, du contrôle contentieux et de la repression des frauds (DACCRF), “France,” in Healthcare Fraud, Corruption and Waste in Europe: National and Academic Perspectives (M. Mikkers, W. Sauter, P. Vineke, & J. Boertjens eds. 2017), ch. 8, 159. 17  L’Assurance Maladie, supra note 16. 18  Cf. The Law Society, Practice Notes: Financial Abuse (2013), https://www.lawsociety.org.uk/ support-­services/advice/practice-­notes/financial-­abuse/ (last Accessed Jan. 7, 2018).

Introduction to Fraud and Abuse Law   369 created a very broad general fraud offence that may be committed in three ways: by making a false representation (section 2), by failing to disclose information (section 3), and by abuse of position (section 4). This codified offence criminalizes the making of dishonest false representations to secure an actual or potential financial advantage to oneself or disadvantage to another, but it is broader than the US definition of fraud because there is no requirement for detrimental reliance.19 In addition, the inclusion of certain dishonest failures to disclose information within the definition of fraud means that both passive and active conduct is criminalized. For example, the Fraud Act has been used to prosecute those within the healthcare sector who have abused their position of trust for monetary gain,20 but it has also been used to prosecute medical practitioners for lying or simply failing to disclose relevant information when making work applications to agencies or hospital trusts.21 Definitional issues extend beyond to the scope of fraudulent activities to questions regarding the type of penalty, if any, that is merited. For example, in the United States, a physician’s misbehavior might be cognizable as a variety of legal wrongs: a criminal prosecution by the state or federal government, a civil prosecution brought by a governmental or other public entity, a civil action (such as a malpractice suit) filed by an injured individual, an administrative action involving participation in public healthcare programs, or a disciplinary proceeding threatening the physician’s licensure or ability to practice at a particular hospital or clinic. The punishment for these different wrongs ranges from years of imprisonment to monetary fines to loss of medical licensure. A similar range of legal approaches to sanction fraudulent conduct in the healthcare sector may be found in Europe: from the imposition of fixed administrative fines or penalty charges to punish minor fraud (e.g., in obtaining free National Health Service prescription drugs and dental and optical treatment in the UK22 or the fraudulent

19  Fraud Act 2006, s.2. 20  E.g., R v. Moss [2013] EWCA Crim 1554 (practice manager writing out checks on practice accounts in her own favor, forging accounts ledger to cover her dishonesty); R v. Martin [2009] EWCA Crim 2076 (manager of residential care home using position to obtain money from account of vulnerable resident); Onyeaso v. General Medical Council [2013] EWHC 4787 (Admin) (forging prescription forms to obtain medication). Section 2 of the Fraud Act might also be used to prosecute healthcare professionals who claim for payment to which they are not entitled as there is an implied representation when the claim is made that the claimer is entitled to the sum claimed: cf. the pre-­Fraud Act case of R v. Firth (1990) 91 2 Cr App R 217. 21  E.g., R v. Rasoq [2012] EWCA Crim 674 (doctor lied on job application form and failed to disclose that he had been suspended on full pay pending disciplinary proceedings in relation to his conduct and clinical performance); The Professional Standard Authority v. Ghaffar [2014] EWHC 2723 (Admin) (biomedical scientist forged MSc and lied in application for NHS job for which he was not qualified); R v. Madu [2016] EWCA Crim (doctor failed to disclose to his NHS employer that he was applying to medical employment agencies for locum work, failed to disclose his health status to his NHS employer or to the agencies, and failed to disclose to the agencies that he had been suspended from his NHS post). 22  The National Health Service (Penalty Charge) Regulations 1999 (SI 1999/2794); The National Health Service (Penalty Charge) (Scotland) Regulations 1999 (SI 1999/121); The National Health Service (Penalty Charge) (Wales) Regulations 2001 (SI 2001/1300).

370   Tracey A. Elliott and Joan H. Krause granting of sick leave or disability certificates by health professionals in Portugal),23 a criminal prosecution, civil actions (in tort or contract) instituted by aggrieved individuals or institutions, and disciplinary actions brought by employers or regulatory bodies. In addition, the UK and France have in recent years followed the example of the United States and passed legislation to add Deferred Prosecution Agreements (DPAs, known as Convention judiciaire d’interêt public (CJIP) in France)24 to the options available to prosecutors dealing with corporate fraud, avoiding the expense of a lengthy criminal trial by suspending prosecution if the defendant agrees to abide by specified conditions. The range of possible penalties varies enormously, from minor fines to the imposition of huge financial penalties, from the imposition of lengthy custodial sentences to dismissal by an employer, or the suspension or revocation of a healthcare professional’s registration to practice. As a normative matter, should healthcare fraud be addressed only by criminal laws, or are there circumstances in which lesser penalties such as civil fixed penalties or alternative methods of disposal such as DPAs are sufficient? Is the United Kingdom too quick to employ the criminal justice system in healthcare? Are the United States and Central and Eastern European states (in particular, Hungary, Greece, Bulgaria, and Romania) unduly reluctant to do so? The more serious the penalties applicable to healthcare fraud, the more important it is to have a comprehensive, coherent definition that can distinguish truly fraudulent behaviors from lesser improprieties on both ethical and operational levels. In particular, because of the imposition of state punishment following a criminal conviction, widely regarded as “society’s most intrusive and degrading sanction,”25 the legal definition of offences should fairly represent the nature of the criminal conduct committed26 so that the degree and nature of wrongdoing is accurately communicated and a clear signal is sent to society of the level of culpability and how the defendant is to be regarded.27 However, although definitional clarity and consistency in legal enforcement are important, pragmatic operational considerations may prevail over optimal counter-­ fraud activities. The very aggressive policing of fraud has the potential to deter even 23  Inspeção-­Geral des Actividades em Saúde (IGAS), https://www.sns.gov.pt/entidades-­de-­saude/ inspecao-­geral-­das-­atividades-­em-­saude/. For discussion of the power of IGAS to impose administrative fines, see IGAS, “Portugal,” in Healthcare Fraud, Corruption and Waste in Europe: National and Academic Perspectives (M. Mikkers, W. Sauter, P. Vineke and J. Boertjens, eds. 2017), ch. 12. 24  DPAs were introduced in the United Kingdom by the Crimes and Courts Act 2013, Schedule 17 (in force from 24 February 2014). CJIPs were introduced in France by the Sapin II law (2016, in force 1 June 2017). 25  N. Jareborg, “What Kind of Criminal Law,” in Beware of Punishment: On the Utility and Futility of the Criminal Law (A. Snare, ed. 1995), 95. 26  This is known as the principle of fair labelling. See A. Ashworth, “The Elasticity of Mens Rea,” in Crime, Proof and Punishment: Essays in Memory of Sir Rupert Cross (C. F. H. Tapper, ed. 1981), 56; J. Chalmers & F. Leverick, “Fair Labelling in Criminal Law” 71(2) MLR 217 (2008). 27 Ashworth, supra note 26; Chalmers & Leverick, supra note 26, 226–229; A. P. Simester, J. R. Spencer, F. Stark, G. R. Sullivan, & G. J. Virgo, Simester and Sullivan’s Criminal Law: Theory and Doctrine, 7th edition (Hart 2019), 32–33.

Introduction to Fraud and Abuse Law   371 legitimate practices, causing harm to patients. For example, in countries where the health system is underfunded and medical practitioners relatively poorly paid, attempts to crack down on fraud and abuse may encourage physicians and other healthcare practitioners to emigrate to wealthier Western European states (such as Germany, the United Kingdom, or Sweden) where levels of remuneration are higher,28 thus causing or exacerbating a shortage of healthcare professionals to treat patients within the home state. Healthcare fraud also suffers from an identity problem: Are the fraud prohibitions designed only to protect the financial resources earmarked for healthcare, or do they reflect broader ethical principles? When healthcare is publicly funded—as with Medicare in the United States, the National Health Fund in Poland, or the National Health Services in the United Kingdom and Italy—does fraud merely cause financial harm to taxpayers and patients, or does it threaten a more serious violation of public trust? While fraud enforcement focuses on the financial impact of the behavior and the violation of discrete legal prohibitions, anti-­fraud rhetoric often invokes ethical considerations as well. In the words of one former US prosecutor, “What conduct do the fraud laws prohibit? They prohibit a breach of good faith and fair dealing as understood in the community involving a deception or breach of trust. . . . The assumption is that our concept of fraud comes from somewhere other than a positive rule.”29 Recognizing that healthcare fraud may have an ethical as well as a financial component may be useful in setting enforcement priorities. Activities with a high potential for both ethical and financial harm, such as providing unnecessary care to unsuspecting patients,30 are prime targets for fraud prevention and detection efforts. Misbehavior that lacks these hallmarks, such as low-­level fraud by patients to receive free prescription medicines in England, or NHS optical services or dental care in the United Kingdom, or to claim entitlement under European Union Directive 2011/24/EU31 for treatment covered by the European Health Insurance Card (EHIC) or by insurance, seems less compelling as an enforcement priority. Yet even low-­level financial fraud may in the aggregate have significant economic consequences for a public healthcare system,32 thereby implicating broader ethical concerns over the just allocation of limited 28  European Commission, Updated Study on Corruption in the Health Care Sector: Final Report (2017). This report identifies this as being a particular issue in Hungary (pp. 66, 70,75), Romania (pp.100–101), and Bulgaria (p. 204). 29  James Sheehan, Address, Bio-­Tech Fraud: Reality or Fantasy?, 2 Hous. J. Health L. & Pol'y 11, 18 (2002). 30  See e.g., R v. Paterson (Ian) [2017] EWCA Crim 1625, [2018] 1 Cr App R (S) 14, and the case of Louisiana cardiologist Dr. Mehmood Patel. M. O’Riordan, “Cardiologist Gets 10 Years for Performing Unnecessary Interventions” (2009) Medscape, June 26, https://www.medscape.com/viewarticle/705029 31  Directive 2011/24/EU of the European Parliament and of the Council of 9 March 2011 on the application of patients’ rights in cross-­border healthcare, https://eur-­lex.europa.eu/legal-­content/EN/ TXT/?uri=CELEX%3A32011L0024 32  For example, the prescription charge in England for the year 2018–2019 was £8.80 (approximately $11.74) for each medicine or appliance dispensed, yet it has been estimated that prescription fraud costs the NHS £217 million per year; see NHS Counter Fraud Authority, The NHS Counter Fraud Authority: A New Organisation Fighting Fraud in the NHS (2017) p. 4.

372   Tracey A. Elliott and Joan H. Krause resources. Tension arises at times because ethical and financial considerations may suggest different priorities and drive fraud enforcement in different directions. The debate may be theoretical, but it is one with practical consequences.

3  Operational Manifestations of Healthcare Fraud In combination, differences in definition and payment structure dictate that fraud will manifest in different ways in different countries. As noted earlier, how healthcare is paid largely determines the ways in which fraud will occur. To the extent fraud is opportunistic, it will be concentrated at the weakest points in the system, those with the greatest opportunity for profit and the lowest chance of detection. For example, in countries where dual practice in both state and private health facilities is permitted, there is the temptation for dishonest practitioners to “double bill” for services provided or to seek to divert patients to the private sector. In underfunded systems (such as Greece, Lithuania, and Hungary),33 where doctors are paid low salaries and informal payments are poorly regulated, there is the temptation to boost remuneration by inflating the number of hours worked or asking patients for under-­the-­table additional payments or gifts. Moreover, the ubiquity of the internet has provided an opportunity for fraudsters everywhere to set up websites selling counterfeit or “fake” medicines directly to the public, bypassing the regulation of online pharmacies.34 The fragmented, decentralized model of healthcare insurance in the United States, combined with the heavy reliance on easily falsified documentation to assure the propriety of services provided, offers similar opportunities for fraud. In virtually any system, the ability of third parties to influence the delivery of care will be of concern; a recent example is the potential for the pharmaceutical industry to exert influence over physician prescribing habits. And any changes in payment models—including the ongoing move towards “value-­based” models of care in the United States and the shift in the UK and the Netherlands to a more market-­based approach with greater involvement of private-­sector providers in the provision of health services—are likely to shift both the locus and the form of healthcare fraud, requiring recalibration of the enforcement response. 33  European Commission, supra note 30, pp. 34, 56, 70. 34  Falsified Medicines Directive 2011/62/EU; Regulation EU/699/2014); and in the UK: Human Medicines (Amendment) Regulations 2013 (SI 2013/1855); MHRA, MHRA Guidance. Selling Human Medicines Online (Distance Selling) to the Public (2017), www.gov.uk/government/uploads/system/ uploads/attachment_data/file/600339/FMD_guidance_document_revised_March_2017.pdf); General Pharmaceutical Council, Guidance for Registered Pharmacies Providing Pharmacy Services at a Distance, Including on the Internet (2017), www.pharmacyregulation.org/sites/default/files/guidance_ for_registered_pharmacies_providing_pharmacy_services_at_a_distance_including_on_the_internet_ april_2015.pdf

Introduction to Fraud and Abuse Law   373 In these chapters, we will focus on healthcare fraud and the range of legal solutions provided to address fraud in the United States (primarily at the federal level) and Europe. Given the sheer number of European states, the different ways in which their healthcare systems are organized and funded, and the differences in state legal regulation, it is not possible to identify a single “European approach” to healthcare fraud. Instead, the European chapter will seek to provide key examples of the range of approaches adopted within Europe. First, because fraud is dynamic and dependent on the mechanisms for funding in the relevant system, we examine—with a broad brush— the manner in which healthcare systems in the United States and Europe are funded and organized. Second, we examine the issue of what qualifies as healthcare fraud, considering both the definitional problem and the interplay between the moral and ethical content of fraud and legal definitions of fraudulent conduct. Third, we address the criminal and civil laws used to combat fraud in these jurisdictions. Fourth, we consider operational issues in the identification, prevention, and punishment of healthcare fraud in the United States and Europe, examining common examples of fraud occurring within these states’ healthcare systems and the problems encountered by law enforcement agencies in seeking to tackle such criminality. Finally, we conclude by considering the common themes that emerge when the issue of healthcare fraud in the United States and Europe is examined, and key points of difference among states. Whilst it is true to say that healthcare fraud is universal and persistent, nevertheless it is historically, socially, and politically constructed. Thus, social and political conditions (such as the reforms in former communist Central and Eastern European countries since the collapse of the Soviet Bloc and differences among states in levels of corruption) may lead to healthcare fraud manifesting itself in different forms from one state or geographical area to another.

chapter 20

Fr au d a n d A buse L aw i n the U n ited States Joan H. Krause

1  Introduction and Context Despite years of increasing enforcement, healthcare fraud remains a significant problem in the United States. The federal government has taken the lead in anti-­fraud activities, not only increasing the resources devoted to fraud enforcement but also expanding the range and severity of laws targeting healthcare fraud. Both individually and collectively, governmental and private payors have developed increasingly sophisticated analytical tools to enable earlier fraud detection. Yet healthcare fraud persists, seemingly able to adapt to each change. The United States experience exemplifies the ethical, definitional, and operational concerns that make healthcare fraud so difficult to identify let alone to prevent. Because the United States lacks a centralized, universal program of health insurance, healthcare is financed by a variety of public, private, and hybrid sources. In 2017, more than half of the population received health insurance from private sources, with employer-­based insurance covering nearly half and roughly 36% receiving public healthcare benefits.1 National healthcare expenditures reached an estimated $3.5 trillion in 2017, accounting for 17.9% of gross domestic product. The federal government was responsible for the largest share of total healthcare spending (28.1%), followed by households (28%), private businesses (19.9%), and state and local governments (17.1%).2 1  Kaiser Family Foundation, State Health Facts: Health Insurance Coverage of the Total Population, https://www.kff.org/other/state-­indicator/total-­population/?currentTimeframe=0&selectedDistributio ns=employer-­-­non-­group&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D (last visited Aug. 30, 2019). Approximately 14% were covered by the federal Medicare program for the elderly and long-­term disabled, 21% were covered by the joint federal-­state Medicaid program, and 1% were covered by other public programs including those for veterans and the active military. 2  Centers for Medicare and Medicaid Services (CMS), National Health Expenditure Fact Sheet (last modified Apr. 26, 2019), https://www.cms.gov/Research-­Statistics-­Data-­and-­Systems/Statistics-­Trends-­ and-­Reports/NationalHealthExpendData/NHE-­Fact-­Sheet.html

376   Joan H. Krause Healthcare reimbursement in the United States is in the midst of significant change. Historically, healthcare was reimbursed under a fee-­for-­service system in which providers were paid for each separate service rendered to a patient. In addition to encouraging overtreatment, inefficiency, and the fragmentation of care, fee-­for-­service reimbursement creates opportunity for fraud: the more care provided (or at least billed), the more payment made, regardless of whether that care is necessary or of high quality.3 In response, payors have experimented with numerous strategies to control healthcare costs, initially under the moniker of “managed care.” The strictest mechanism for managing care was through the strategy of capitation, in which a primary care physician received a fixed per-­patient payment in return for meeting the patient’s healthcare needs during a set period of time. To the extent capitation reduced costs, however, it did so largely by imposing barriers to certain treatments and limiting patients to seeing providers who were part of the organization’s network and had agreed to abide by various financial and caregiving conditions, leading to both patient and physician dissatisfaction. Moreover, while capitation blunted the temptation to provide excessive services, it moved too far in the other direction, incentivizing providers to undertreat and to target patients less likely to need extensive or expensive care.4 More recently, there has been a push—spearheaded by changes made to Medicare reimbursement under the Patient Protection and Affordable Care Act (ACA)—to reward providers for the value rather than the volume of services provided.5 Anti-­fraud efforts are undertaken by not only the federal government, but also by state governments and private insurers, creating a multifaceted and at times in­con­sist­ ent approach to fraud prevention and enforcement in the United States. This decentralized model makes it difficult to ascertain the true extent of fraudulent activities. Moreover, we only can measure the fraud we can identify; by definition, the most successful fraud schemes will remain hidden. As Professor Malcolm Sparrow succinctly observed, “What you see is not the problem. It’s what we don’t see that really does the damage.”6 The Federal Bureau of Investigation (FBI) estimates that healthcare fraud costs “tens of billions of dollars” every year, while other recent estimates range from $75 billion to an astounding $270 billion a year.7 While there simply is no way to tell 3  See Joan H. Krause, Integration, Fragmentation, and Human Nature: The Role of the Fraud and Abuse Laws in a Changing Health Care System, in Oxford Handbook of U.S. Healthcare Law 852 (I. Glenn Cohen, Allison Hoffman, & William M. Sage, eds., 2017). 4  Id.; Michael E. Porter & Robert S. Kaplan, How to Pay for Health Care, Harv. Bus. Rev. 88 (July–Aug. 2016). 5  Porter & Kaplan, supra note 4; 42 U.S.C. §§ 1395w-­4(p) (value-­based payment modifier for physicians) & 1395ww(o) (hospital value-­based purchasing program); Medicare Access and CHIP Reauthorization Act of 2015, Pub. L. No. 114–141, 129 Stat. 87 (2015) (changing physician payment). 6  Malcolm K. Sparrow, License to Steal: How Fraud Bleeds America’s Health Care System 2 (updated ed. 2000). 7  Federal Bureau of Investigation (FBI), What We Investigate: Health Care Fraud, https://www.fbi. gov/investigate/white-­collar-­crime/health-­care-­fraud (last visited Aug. 1, 2018); National Health Care Anti-­Fraud Association, The U.S. Health Care System and the Challenges of Fraud 9–10 (Sept. 2017) (listing various estimates), https://www.nhcaa.org/media/127538/nhcaa_ushealthcaresystem_2017.pdf

Fraud and Abuse Law in the United States   377 exactly how much money is lost to healthcare fraud in the United States, the amount undoubtedly is substantial. Given the stakes, it is no surprise that the federal government expends significant resources every year to combat healthcare fraud and abuse.

2  Ethical, Definitional, and Operational Aspects of Healthcare Fraud in the United States The complexity of the healthcare system in the United States, combined with the wide range of available legal remedies, has resulted in anti-­fraud efforts that appear dis­or­gan­ ized and at times inconsistent. Despite general agreement about the goal of eliminating healthcare fraud, there is little consensus as to the most effective means of doing so or sometimes even whether particular conduct constitutes fraud at all. As a result, definitional, operational, and ethical concerns permeate the healthcare fraud debate.

2.1  Definitional Concerns At the most basic level, healthcare fraud and abuse suffers from a basic definitional problem: What, exactly, does the term encompass? Even within a seemingly unified federal scheme, there is wide variation in the types of activities that are considered to constitute healthcare fraud and abuse. Fraud in this context denotes activities undertaken with the intent to deceive, coupled with another’s detrimental reliance on the misrepresentation.8 Abuse, in contrast, refers to self-­interested behavior done without the intent to deceive, such as creative interpretation of payment rules or “other instances of profligate spending on medical care that is at the margin of fraudulent practice.”9 The lines between these categories are by no means clear: for example, a doctor who submits a bill for a more expensive category of services than was actually provided to a patient, such as a longer or more complex examination, might have made an honest mistake, might have engaged in abusive behavior by opting for a self-­serving interpretation of an ambiguous billing code, or might well have committed fraud by intentionally submitting a false bill for more lucrative services. The difference may lie not only in behavior, but also in what a prosecutor is able to prove regarding the doctor’s state of mind. For comparative purposes, however, one clarification is necessary: in the United States, “abuse” is almost 8  Jerry L. Mashaw & Theodore R. Marmor, Conceptualizing, Estimating, and Reforming Fraud, Waste, and Abuse in Healthcare Spending, 11 Yale J. Reg. 455, 463 (1994). 9  Id. at 463–464 (“Abuse lies in the motivation imputed to the provider, in the suspicion that costs were increased for purposes related, at best, only tangentially to producing a better medical result”); see also Robert Fabrikant et al., Health Care Fraud: Enforcement and Compliance § 1.01 (2018, Rel. 43) (“Abuse involves actions inconsistent with acceptable business or medical practices”).

378   Joan H. Krause entirely a financial concept; it does not encompass the physical abuse or neglect of vulnerable patients. Unless such mistreatment violates the requirements imposed by the federal healthcare programs or otherwise is accompanied by an improper request for payment, it will not fall within the definition of healthcare fraud and abuse.10 Those cases are left primarily to the states to address under separate abuse and neglect statutes.11 Another category of undesirable behavior falls under the label of waste. Rather than an intention to deceive or to take advantage of uncertainty in billing rules, waste refers to “medical interventions . . . that produce, on average, no better medical results than some cheaper treatment . . . [or that] fail to produce benefits that on average exceed their costs.”12 Many forms of waste may be systemic problems rather than suboptimal choices by individual physicians, such as the documented higher utilization rates seen when physicians own diagnostic equipment compared to when they do not.13 Whether waste constitutes a form of fraud rather than simply a form of inefficiency depends on the specific facts at issue. The lines between the three concepts are blurry at best, and a particular behavior often could be characterized as fitting within multiple categories. Indeed, under many of the fraud laws the characterization may not matter. The Civil False Claims Act (FCA), for example, reaches not only misrepresentations that the claimant knows to be false, but also situations in which the claimant acts in reckless disregard or deliberate ignorance of truth or falsity—a definition that easily encompasses many situations in which providers creatively interpret payment rules in ways that increase their reimbursement.14 The situation is quite different, however, under a criminal fraud statute requiring proof of knowing and willful intent, such as the Medicare and Medicaid Anti-­Kickback Statute.15 The definitional uncertainty may make it difficult to know in advance whether a particular activity will fall within the statutes prohibiting healthcare fraud and abuse.

2.2  Operational Issues Concerns about fraud extend beyond the definitional, encompassing not only the scope of fraudulent activities but also the ways in which the prohibitions are operationalized. Healthcare fraud enforcement may take a variety of forms: criminal prosecution, civil 10  See, e.g., OIG, Early Alert: The Centers for Medicare & Medicaid Services Has Inadequate Procedures to Ensure that Incidents of Potential Abuse or Neglect at Skilled Nursing Facilities are Identified and Reported in Accordance with Applicable Requirements, 01-­17-­00504, Aug. 24, 2017, https://oig.hhs.gov/oas/reports/region1/11700504.pdf 11  See, e.g., N.C. Gen. Stat. §§ 108A.99 et seq. (Protection of the Abused, Neglected, or Exploited Disabled Adult Act). 12  Mashaw & Marmor, supra note 8 at 464–465. 13  See id. at 455; OIG, Financial Arrangements Between Physicians and Health Care Businesses, OAI-­12-­88-­01410, May 1989, https://oig.hhs.gov/oei/reports/oai-­12-­88-­01410.pdf (concluding that patients of doctors with financial ties to clinical laboratories received more clinical laboratory services than other Medicare patients). 14  31 U.S.C. § 3729(b)(1). 15  42 U.S.C. § 1320a-­7b(b).

Fraud and Abuse Law in the United States   379 FCA suits by the government or private actors, or administrative actions under the public healthcare programs—not to mention ancillary civil actions such as medical malpractice and disciplinary proceedings involving licensure or hospital staff privileges. The choice of enforcement pathway is driven not only by practical concerns but also by normative ones: Should healthcare fraud be addressed only by the most serious criminal sanctions or are there circumstances in which lesser penalties suffice? If the latter, what factors should distinguish the cases: the extent of wrongdoing, the level of intentionality, the financial costs, the potential for physical or financial harm to patients in addition to harm to the government, or some combination thereof? And how do we weigh these factors against other policy concerns, such as the possibility that excessive enforcement might chill legitimate behaviors by risk-­averse healthcare providers, thereby leading to unanticipated patient harms? Calibrating the proper pathway for fraud enforcement requires a balancing of concerns to assure we do not inadvertently cause more harm than we seek to prevent. In the United States that choice often is left to prosecutorial discretion, which may be influenced as much by political and policy concerns as by the severity of the misconduct.16 Because of the significant overlap in the conduct covered by the civil, administrative, and criminal fraud laws, prosecutors retain a great deal of control over the disposition of cases. A doctor who intentionally “upcodes” a claim by billing for more complex and lucrative services than what was given to a patient, for example, could face “permissive” exclusion from the federal healthcare programs at the government’s discretion, civil monetary penalties for filing an improper claim, criminal penalties for making a false statement or representation in an application for federal healthcare program benefits, civil liability under the FCA, or criminal penalties under the criminal false claims provisions—just to name a few.17 Whether that case will be pursued as a criminal or civil matter, or perhaps declined altogether, also depends on enforcement priorities. Governing policies include prosecution guidelines, enforcement initiatives, and both formal and informal interagency understandings regarding which agency has primary jurisdiction in different types of cases—all of which may be subject to change depending not only on evolving definitions of fraud, but also on the priorities of different presidential administrations. For example, while civil fraud prosecutions were prominent for much of the Bush Administration, the early years of the Obama Administration saw a greater focus on criminal investigations.18 That change was also evident in an announcement in 2014 that the Department of Justice (DOJ) Civil Division would share all new private FCA 16  See Joan H. Krause, Following the Money in Health Care Fraud: Reflections on a Modern-­Day Yellow Brick Road, 36 Am. J. L. & Med. 343, 358 (2010) (describing choice of enforcement pathways). 17  See 42 U.S.C. §§ 1320a-­7(b)(7) (exclusion), 1320a-­7a (civil penalties), and 1320a-­7b(a) (criminal penalties); 31 U.S.C. § 3729(a)(1)(A) (FCA); 18 U.S.C. § 287 (criminal false claims). 18  See Krause, Integration, Fragmentation, and Human Nature, supra note 16, table 1 (comparing civil and criminal cases from 2004–2013). In the first years of the Trump Administration, the number of criminal fraud investigations declined from the Obama-­era high, although both civil and criminal investigations appeared to exceed the pace during the Bush Administration. See id.; Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2017, Apr. 2018, at 1, https://oig.hhs.gov/ publications/docs/hcfac/FY2017-­hcfac.pdf.

380   Joan H. Krause c­ omplaints with the Criminal Division as soon as they were filed, significantly increasing the chances that FCA suits would be accompanied by parallel criminal proceedings.19 A similar evolution has occurred with regard to the DOJ approach to prosecuting corporate fraud, an area of major concern in healthcare given the proliferation of entities—from hospitals to home health organizations to pharmaceutical manufacturers— that are involved in the US healthcare system. During the Clinton Administration, the DOJ issued a memorandum on Federal Prosecution of Corporations setting forth the factors that would guide the decision of whether to bring federal charges against a corporation; this included the corporation’s willingness to waive the attorney–client and work product privileges in order to demonstrate cooperation.20 In the wake of the Enron and WorldCom scandals, the Bush Administration revised the document to include consideration of whether the responsible individuals could be prosecuted but appeared to limit the ability of prosecutors to demand that corporations waive the privileges.21 In 2015, the Obama Administration replaced these documents with the “Yates Memorandum,” in which the DOJ vowed to pursue action against both the corporation and any individuals involved in the wrongdoing.22 Although the Yates Memorandum was not withdrawn by the Trump Administration, the DOJ issued more defendant-­ friendly guidance, including a memorandum encouraging the dismissal of many private FCA cases.23 The choice of enforcement pathway obviously will depend heavily on the particular facts. Where the primary harm is financial, the goal may be the return of those assets to the government; criminal sanctions may be reserved for cases in which financial compensation is inadequate, such as physical injury to patients. Yet even financial penalties are designed to deter future misconduct as well as to make the victim whole, and they must be calibrated correctly if they are to achieve the desired effect.24 In the United States, the process is further complicated by the specter of enormous civil penalties and 19  See HHS, Office of the Inspector General, Medicare Fraud Strike Force, https://oig.hhs.gov/fraud/ strike-­force/ (last visited Aug. 10, 2018); Remarks by Assistant Attorney General for the Criminal Division Leslie R. Caldwell at the Taxpayers Against Fraud Education Fund Conference (Sept. 17, 2014), https://www.justice.gov/opa/speech/remarks-­assistant-­attorney-­general-­criminal-­division-­leslie-­r-­caldwell­taxpayers-­against 20  Memorandum from Eric H. Holder, Jr., Deputy Attorney Gen., U.S. Dep’t of Justice, to All Component Heads and United States Attorneys (June 16, 1999), https://www.justice.gov/sites/default/ files/criminal-­fraud/legacy/2010/04/11/charging-­corps.pdf 21  See Gideon Mark, The Yates Memorandum, 51 U.C. Davis L. Rev. 1589, 1596–1602 (2018) (discussing the memoranda). 22  Memorandum from Sally Quillian Yates, Deputy Attorney Gen. on Individual Accountability for Corporate Wrongdoing (Sept. 9, 2015), https://www.justice.gov/archives/dag/file/769036/download 23  Memorandum from Michael D. Granston, Director, Commercial Litigation Branch, Fraud Section, on Factors for Evaluating Dismissal Pursuant to 31 U.S.C. 3730(c)(2)(A) (Jan. 10, 2018), https:// assets.documentcloud.org/documents/4358602/Memo-­for-­Evaluating-­Dismissal-­Pursuant-­ to-­31-­U-­S.pdf 24  See, e.g., Katrice Bridges Copeland, Enforcing Integrity, 87 Ind. L.J. 1033 (2012) (arguing that civil and administrative settlements have proved ineffective in deterring misconduct); William M. Sage, Fraud and Abuse Law, 282 JAMA 1179, 1180 (1999) (arguing that the tendency to settle fraud allegations makes “much of fraud control resemble a rebate program more than a law enforcement exercise”).

Fraud and Abuse Law in the United States   381 the threat of exclusion from federal healthcare programs, which create strong incentives to settle rather than risk a trial. These penalties are so large that some consider them an example of “punitive civil sanction[s] . . . in which the sanction’s purpose is punishment, but its procedure is drawn primarily from the civil law.”25 To the extent the healthcare fraud laws impose arguably punitive penalties, they elide the traditional distinctions between criminal and civil enforcement mechanisms. Moreover, even if current government policy discourages prosecution, the FCA permits private “relators” to file their own suits, independent of the governmental chain of command.26 As a result, there remains significant unpredictability regarding which activities will be considered fraudulent and what remedies will be sought.

2.3  The Ethical Dimensions of Fraud Finally, healthcare fraud suffers from an identity problem: Are the fraud prohibitions designed to protect the financial resources earmarked for healthcare, or do they reflect broader ethical principles? Is the primary concern the potential to divert resources from their intended uses—resources which, in the case of publicly funded healthcare programs, are contributed by taxpayers? Are these activities not only financially undesirable but also inherently unethical? Many definitions of fraud, at least implicitly, contain both financial and ethical components. According to one leading text, “[f]raud involves intentional acts of deception with knowledge that the action or representation could result in an inappropriate gain.”27 Fraud also requires “the detrimental reliance on that misrepresentation by another party to the transaction,”28 meaning that the financial gain comes at another’s expense. Thus, the ethical component of fraud is described as a breach of trust in interpersonal relationships: “deceit, corruption, shame, cheating, and duplicity toward others.”29 In healthcare, those “others” may include not only public and private payors, but also disadvantaged rule-­abiding competitors as well as patients who suffer physical injuries, incur additional expenses, or face intangible dignitary harms from interference with their ability to make free and informed healthcare choices. That harm might even 25  Kenneth Mann, Punitive Civil Sanctions: The Middleground Between Criminal and Civil Law, 101 Yale L.J. 1795, 1799 (1992). See also Vermont Agency of Natural Resources v. U.S. ex rel. Stevens, 529 U.S. 765, 784 (2000) (noting that Civil False Claims Act “imposes damages that are essentially punitive in nature”); Joan H. Krause, Health Care Providers and the Public Fisc: Paradigms of Government Harm Under the Civil False Claims Act, 36 Ga. L. Rev. 121, 202–210 (2001) (describing problems with FCA settlements). 26  31 U.S.C. § 3730(b). 27  Fabrikant et al., supra note 9, at § 1.01. 28  Mashaw & Marmor, supra note 8, at 463. 29  Tamar Frankel, Trust and Honesty: America’s Business Culture at a Crossroad 4 (2006) (defining fraud as a synonym of dishonesty, distrust, and deception). See also Stuart P. Green, Why It’s a Crime to Tear the Tag Off a Mattress: Overcriminalization and the Moral Content of Regulatory Offenses, 46 Emory L.J. 1533, 1586–1590 (describing certain regulatory crimes as violations of the duty of fair play).

382   Joan H. Krause accrue to the broader population of taxpayers, whose tax dollars are taken with the understanding that they will be used to pay for legitimate healthcare services through government programs such as Medicare and Medicaid. At a rhetorical level, the federal government has embraced the message that healthcare fraud is a breach of both ethical and financial expectations. As one former federal healthcare fraud prosecutor explained, because “our concept of fraud comes from somewhere other than a positive rule,” our laws are designed to “prohibit a breach of good faith and fair dealing as understood in the community involving a deception or breach of trust.”30 That rhetoric is heightened when fraudulent activities have the potential to harm patients rather than simply causing losses for the federal Treasury.31 Faced with these threats, the federal government has positioned itself as the protector of vulnerable citizens and the guardian of the ethical values of the medical profession. Yet when it comes to fraud enforcement, rather than acting as a protector, the government more clearly plays the role of a defrauded payor. In a FCA suit brought by a private party, for example, roughly 15–30% of the proceeds will be awarded to the relator who initiated the suit.32 Under current law, most of the remaining funds, as well as civil penalties and healthcare-­related criminal fines and forfeitures, must be deposited into the Medicare Part A Trust Fund; from there, those funds are available for appropriation to the Health Care Fraud and Abuse Control Account to fund future anti-­fraud efforts.33 Thus, the money recovered in fraud cases is available to the Medicare program, whistleblowers, and the investigating and prosecuting agencies themselves but rarely is shared with patient-­victims or otherwise directed to address deeper distributional or access concerns. While the rhetoric of fraud enforcement has embraced the ethical basis of fraud, the reality of fraud recovery focuses primarily on the financial. Even if not fully reflected in recoveries, however, recognizing that healthcare fraud may have an ethical as well as a financial component may be useful in setting enforcement priorities. Activities with a high potential for both ethical and financial harm—particularly those with the potential for causing physical harm to patients, such as unnecessary surgeries—are prime targets for fraud prevention and detection efforts. In one case, federal prosecutors alleged that cardiac surgeons had performed unnecessary heart ­surgeries on as many as 700 patients, including unnecessary open heart and coronary bypass surgeries.34 In contrast, activities that involve small financial improprieties with little potential for physical injury may be less compelling enforcement targets. Yet even low-­level financial frauds may, in the aggregate, have significant economic consequences—consequences that are magnified when they implicate broader ethical debates 30  James Sheehan, Address, Bio-­Tech Fraud: Reality or Fantasy?, 2 Hous. J. Health L. & Pol’y 11, 18 (2002). 31  See Joan H. Krause, Healthcare Fraud and Quality of Care: A Patient-­Centered Approach, 37 J. Health L. 161, 173–175 (2004) (describing rhetoric of patient protection in healthcare fraud cases). 32  31 U.S.C. § 3730(d). 33  42 U.S.C. § 1395i(k). 34  See Press Release, US Attorney, E. Dist. of Cal., Doctors Accused of Performing Unnecessary Heart Surgeries at Redding Medical Center Agree to Pay Millions to Settle Fraud Allegations and Accept Restrictions on Their Medical Practice (Nov. 15, 2005).

Fraud and Abuse Law in the United States   383 over the just allocation of limited resources in a publicly funded healthcare program such as Medicare.35 Thus, these considerations may at times drive fraud enforcement in different directions.

3  Legal Context of Fraud Enforcement in the United States 3.1  Laws Commonly Used to Address Health Care Fraud in the United States In the United States, the decentralized nature of healthcare leads to corresponding complexity in fraud enforcement authority. A variety of laws may be used to pursue healthcare fraud. Some of these laws—such as the Medicare and Medicaid Anti-­Kickback Statute (AKS), the federal Health Care Fraud crime, and state Medicaid Fraud statutes— directly target fraud in health insurance and benefits programs.36 Other statutes provide civil and administrative penalties for specific activities involving publicly funded programs, such as “exclusion” of the healthcare provider from participation.37 The FCA offers another model, applying broadly to fraudulent financial transactions with the federal government.38 Healthcare fraud also may be prosecuted under traditional federal criminal laws such as Mail and Wire Fraud, as well as general state fraud and consumer protection statutes.39 Together, these statutes create a web of overlapping federal and state jurisdiction imposing a variety of civil, criminal, and administrative penalties. Of these myriad statutes, three federal laws draw the most attention: the FCA, the AKS, and the “Stark” physician self-­referral prohibition. Despite its roots as a statute designed to prevent fraud on the Union Army during the Civil War, the FCA has evolved into a potent weapon against modern healthcare fraud. The FCA prohibits a variety of activities, including submitting false or fraudulent claims to the government and engaging in conspiracies to submit false claims. The basic false claims provision prohibits a defendant from presenting or causing to be presented a claim for payment or approval where the claim is false or fraudulent and the acts are undertaken “knowingly”—a broad intent standard that includes not only 35  See, e.g., United States v. Krizek, 859 F. Supp. 5 (D.D.C. 1994) (alleging that Medicaid psychiatrist “upcoded” more than 8,000 claims by approximately $30 each, for total damages of $245,392 and potential FCA exposure of almost $81 million). 36  See, e.g., 18 U.S.C. § 1847 (Health Care Fraud); 42 U.S.C. §1320a-­7b(b) (Medicare & Medicaid Anti-­Kickback Statute); N.C. Gen. Stat. §§ 108A-­70.10 et seq. (North Carolina Medical Assistance Provider False Claims Act). 37  See 42 U.S.C. §1320a-­7 (exclusion). 38  31 U.S.C. §§ 3729 et seq. 39  See, e.g., 18 U.S.C. §§1341 & 1343 (Mail and Wire Fraud); Tex. Stat. Ann. §§ 17.46 (Deceptive Trade Practices).

384   Joan H. Krause actual knowledge of falsity but also deliberate ignorance and reckless disregard.40 The FCA is a civil statute, carrying no prison term but imposing massive financial penalties: as of June 19, 2020, violations are subject to civil penalties of $11,665 to $23,331 per claim, plus three times the government’s damages.41 FCA cases may be filed either by federal prosecutors or by qui tam “relators,” private parties (such as competitors or ­former employees) who sue on the government’s behalf in exchange for a portion of the proceeds of a successful suit.42 While the FCA governs the submission of claims to the federal government, the AKS is the major federal law affecting financial relationships within the healthcare market. The AKS prohibits offering, paying, soliciting, or receiving any “remuneration” to induce someone to refer patients or to purchase, order, or recommend any item or serv­ice that may be paid for under a federal healthcare program.43 This broad criminal statute is designed to limit the influence of financial incentives on healthcare decision-­making, reaching both those who offer or pay as well as those who solicit or receive remuneration. Remuneration is defined more broadly than a direct bribe or kickback, reaching any payments made “directly or indirectly, overtly or covertly, in cash or in kind.”44 For example, offering below-­market rent in a hospital-­owned building to a doctor who refers a significant number of patients to that hospital would be a form of remuneration because it relieves the physician of a financial burden—higher rent—she otherwise would have to bear. The statute applies to referrals made in connection with any of the federal healthcare programs, although Medicare and Medicaid cases are most common. Violation is a felony subject to up to 10 years imprisonment and a fine of $100,000, as well as exclusion from the federal healthcare programs and civil penalties of $100,000; a claim for serv­ ices provided in violation of the AKS also constitutes a false or fraudulent claim under the FCA.45 Finally, the Ethics in Patient Referrals Act of 1989, often called the “Stark Law” after its sponsor Representative Fortney “Pete” Stark, is a civil statute that prohibits the referral of Medicare and Medicaid patients to entities with which the referring physician has a financial relationship.46 Although both the Stark Law and the AKS are designed to limit the effect of financial incentives on healthcare decision-­making, Stark reflects the specific concern that self-­referral may lead to overutilization. The law prohibits referrals of patients for designated health services—including clinical laboratory services, inpatient and outpatient hospital services, outpatient prescription drugs, and home health serv­ices—if the referring physician or an immediate family member has a “financial relationship” with the entity providing those services unless the arrangement meets a limited number of exceptions.47 Prohibited financial relationships include both ownership/ investment (via equity, debt, or any other means) and compensation, incorporating a 40  31 U.S.C. § 3729. 41  28 C.F.R. § 85.5. 42  31 U.S.C. § 3730(d). 43  42 U.S.C. § 1320a-­7b(b). 44  Id. at § 1320a-­7b(b)(1) & (2). 45 Id. See also id. at §§ 1320a-­7(a) (mandatory exclusion), 1320a-­7(b)(7) (permissive exclusion) §1320a–7a (civil penalties), 1320a-­7b(g) (FCA). 46  42 U.S.C. §1395nn. 47  Id. at § 1395nn(a), (b) & (h)(6).

Fraud and Abuse Law in the United States   385 broad definition of remuneration.48 The statute prohibits payment for services furnished pursuant to prohibited referrals, and any such claims will be denied; the knowing submission of a bill for prohibited services is subject to a civil penalty of up to $15,000 for each service and $100,000 for each arrangement or scheme to violate the statute, as well as potential exclusion from the federal healthcare programs and a potential FCA suit.49

3.2  Representative Examples of Healthcare Fraud The definitional, operational, and ethical debates permeating healthcare fraud are illustrated by recent examples of fraud enforcement in the United States. Fraud occurs in every corner of the healthcare market, although different sectors have been singled out for scrutiny over time. The 2017 Health Care Fraud and Abuse Control Program Annual Report, for example, highlighted more than twenty different areas of recent enforcement, ranging from ambulance services to psychiatric and psychological testing serv­ices.50 According to a Government Accountability Office report based on 2010 data, the majority (60.6%) of civil fraud investigations and almost half (49.2%) of criminal investigations focused on entities such as hospitals, durable medical equipment suppliers, home health providers, and medical facilities. Physicians comprised the largest category of individuals, targeted in an estimated 15.4% of criminal and 12.1% of civil investigations.51 To the extent healthcare fraud is an opportunistic crime, it occurs in different ways depending on how care is reimbursed. Where payment is based on the volume of serv­ ices provided, such as the complexity of a physician appointment or the patient’s diagnosis at the time of hospital admission, providers may be tempted to “upcode” to more lucrative categories of care.52 Where payment is conditioned on a physician’s certification that services are necessary, such as for Medicare coverage of home health services or durable medical equipment, providers may falsify paperwork to make it appear that care has been authorized.53 Where entities are reimbursed on the basis of complicated rules that permit only certain costs to be charged to the federal healthcare programs, as in the case of Medicare hospital cost reports, providers may be tempted to shift costs

48  Id. at 1395nn(a) & (h). 49  42 U.S.C, §1395nn(g)(3)-(4); United States ex rel. Thompson v Columbia/HCA Healthcare Corp, 125 F.3d 899 (5th Cir. 1997) (permitting FCA suits). 50  Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2017, supra note 18, at 13–32. 51  Government Accountability Office, Types of Providers Involved in Medicare, Medicaid, and the Children’s Health Insurance Program Cases, GAO-­12-­820, at 16–18, 25–27 (Sept. 2012), https://www.gao. gov/assets/650/647849.pdf 52  See, e.g., United States v. Krizek, 859 F. Supp. 5 (D.D.C. 1994) (alleged billing for lengthier psychiatric treatment sessions than actually occurred). 53  Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2017, supra note 18, at 18–23 (describing recent cases).

386   Joan H. Krause into categories that receive favorable payment.54 In short, payment mechanisms drive the way in which fraud will be expressed in any particular healthcare sector. Despite the wide array of cases, several categories of providers have long been under scrutiny. By virtue of their roles at the core of medical care and research, hospitals are at the center of a web of complex financial relationships. Hospital reimbursement is multifaceted, with institutions receiving funding not only for services provided in inpatient and outpatient facilities, but also for the training of medical students and residents. This raises concerns not only regarding whether care was appropriately and accurately billed, but also which category of medical professional rendered that care.55 To the extent that hospitals must rely heavily on physicians to send them patients, they may be tempted to violate the AKS by paying physicians for referrals—if not directly, then perhaps in the guise of perks such as reduced space or equipment rental fees.56 Similarly, due to the fact that many designated health services are provided in the hospital setting, hospitals have been the defendants in the highest profile Stark cases.57 In short, nearly all hospital activities, from surgery to laboratory services to outpatient facilities, have the potential to create fraud liability. While hospitals remain perpetually under scrutiny, perhaps the most visible enforcement target in recent years in the United States has been the pharmaceutical industry. Between 1991 and 2015, pharmaceutical manufacturers entered into 373 fraud settlements totaling $35.7 billion.58 Common allegations have included the unlawful promotion of pharmaceutical products and the payment of various forms of kickbacks to physicians. Nearly every facet of research, development, and sales has been the subject of industry-­wide investigations. Fraud concerns may arise throughout the lifespan of a pharmaceutical product. In the research stage, growing attention has been paid to the legitimacy and transparency of research study design, conduct, and publication. For example, critics have raised concerns over so-­called seeding trials, which are designed to “seed” the market for a new 54 DOJ, Largest Health Care Fraud Case in U.S. History Settled: HCA Investigation Nets Record Total of $1.7 Billion (June 26, 2003) (describing allegations that hospital chain engaged in cost reporting fraud schemes), https://www.justice.gov/archive/opa/pr/2003/June/03_civ_386.htm 55  See, e.g., United States ex rel. Riley v. St. Luke's Episcopal Hosp., 355 F.3d 370 (5th Cir. 2004) (alleging hospital filed false claims for services that were medically unnecessary or rendered by an unlicensed physician); Pamela Bucy Pierson, The Path from Regulator to Hunter: The Exercise of Prosecutorial Discretion in the Investigation of Physicians at Teaching Hospitals, 44 St. Louis U. L.J. 3 (2000) (describing Physicians at Teaching Hospitals [PATH] initiative investigating academic medical centers for improper billing of resident services when teaching physicians were not present). 56  See, e.g., United States ex rel. Thompson v. Columbia/HCA Corp., 938 F. Supp. 399 (S.D. Tex. 1996), aff ’d in part, vacated in part, 125 F.3d 899 (5th Cir. 1997) (alleging hospital violated FCA by submitting claims for services rendered in violation of the AKS and Stark law based on inducements offered to physicians in exchange for referrals). 57  See, e.g., United States ex rel. Drakeford v. Tuomey, 792 F.3d 364 (4th Cir. 2015) (qui tam action alleging violation of Stark law with regard to part-­time physician employment contracts). 58  Sammy Almashat et al., Twenty-­Five Years of Pharmaceutical Industry Criminal and Civil Penalties: 1991 Through 2015, Mar. 31, 2016, at 4, 10, available at https://www.citizen.org/sites/default/ files/publiccitizen-­pharmasettlements1991-­2015-­chartbook.pdf

Fraud and Abuse Law in the United States   387 drug under the guise of studying a legitimate scientific research question. Ethicists have also questioned studies that are “ghost-­managed” by a manufacturer and study results that are “ghost-­written” by company staff while the research publicly is attributed to independent medical experts.59 Pharmaceutical fraud, in particular, illustrates the definitional, operational, and ethical debates. Intentional misrepresentations of a product’s safety and effectiveness, such as those made to secure Food and Drug Administration (FDA) approval and Medicare payment, are easily characterized as fraud, while behaviors that take advantage of ambiguity in payment rules may be considered abusive. Operationally, however, these disputes are often resolved by large-­scale global settlements in which the company pays a large penalty but does not admit any liability.60 Looming over many of these investigations is a concern that improprieties involving prescription drugs may cause real physical harm to patients or, at the very least, permit drug manufacturers to leverage the knowledge imbalance between doctor and patient in a way that erodes the ethical underpinnings of care. Many straightforward fraud issues arise in manufacturers’ financial relationships with physicians. Because manufacturers cannot make money unless their drugs are prescribed to patients, manufacturers compete with each other to convince physicians to prescribe their products rather than those of their competitors. In the United States, critics have long been concerned that manufacturers may attempt to increase product sales not only by emphasizing clinical superiority, but also by offering physicians financial inducements—ranging from gifts both small (free lunches) and lavish (all-­ expense-­paid trips) to lucrative consultancy or fake research agreements. In 1994, the Office of the Inspector General (OIG) issued a Special Fraud Alert on Prescription Drug Marketing Schemes identifying a number of pharmaceutical marketing tactics that might violate the AKS, including a “frequent flier” ’ scheme in which physicians were given airline miles for completing questionnaires when new patients were prescribed the company’s product, and a sham “research grant’ ” program in which physicians earned substantial payments for minimal patient recordkeeping tasks.61 Nearly a decade later, those concerns were again reflected in the OIG’s Compliance Program Guidance for Pharmaceutical Manufacturers, which warned against offering kickbacks and other financial inducements to purchasers and prescribers and highlighted the

59  See, e.g., Kevin P. Hill et al., The ADVANTAGE Seeding Trial: A Review of Internal Documents, 149 Annals Int. Med. 251 (2008) (describing seeding trial for Vioxx); Sergio Sismondo, Ghost Management: How Much of the Medical Literature Is Shaped Behind the Scenes by the Pharmaceutical Industry?, 4 PLOS Med. 1429 (2007). For a discussion of other issues arising in clinical research, see the chapter by Coleman in this volume. 60  See, e.g., DOJ, Warner-­Lambert to Pay $430 Million to Resolve Criminal & Civil Health Care Liability Relating to Off-­Label Promotion (May 13, 2004), http://www.justice.gov/archive/opa/pr/2004/ May/04_civ_322.htm 61  OIG, Publication of OIG Special Fraud Alerts, 59 Fed. Reg. 65,372, 65,376 (Dec. 19, 1994) (originally issued Aug. 1994).

388   Joan H. Krause particular dangers of mischaracterizing such payments as discounts, educational and research grants, and product support services.62 These concerns have been reflected not only in governmental fraud and abuse guidance, but also in voluntary codes adopted by the American Medical Association (AMA) and the major pharmaceutical trade associations. The AMA Guidelines on Gifts to Physicians from Industry, adopted in 1990 and revised in 2001, warned that gifts “should primarily entail a benefit to patients and should not be of substantial value” and offered strategies to minimize industry influence over research and medical conferences.63 Similarly, the Pharmaceutical Research and Manufacturers of America (PhRMA), the primary name-­brand industry trade association, issued a Code on Interactions with Healthcare Professionals in 2002 (updated in 2008) warning that industry–physician “[i]nteractions should be focused on informing healthcare professionals about products, providing scientific and educational information, and supporting medical education.”64 Concerns regarding the pharmaceutical industry also encompass behaviors that take advantage of legal and ethical gray areas to permit manufacturers to increase their sales through interactions with physicians even without offering illegal inducements. Presuming that manufacturers would be unlikely to spend billions of dollars on ineffective sales efforts, for example, critics have chronicled the ways in which the industry has adapted social psychology research to craft the most effective marketing interventions, such as recruiting medical and scientific professionals as “key opinion leaders” to influence their colleagues’ prescribing behavior.65 The ACA addressed this topic by incorporating the Physician Payments Sunshine Act, which requires manufacturers to report payments to physicians and teaching hospitals—including research payments, consulting and speaker fees, entertainment expenses, and gifts—with much of the reported information available in a publicly accessible database.66 Thus far, however, these efforts have done little to quell concerns about the untoward effects of pharmaceutical industry influence on physician decision-­making. 62  OIG, Compliance Program Guidance for Pharmaceutical Manufacturers, 68 Fed. Reg. 23,731, 23,735–23,736 (May 5, 2003). 63  Council on Ethical and Judicial Affairs (CEJA), Gifts to Physicians from Industry, 265 JAMA 501 (1991); CEJA, Guidelines on Gifts to Physicians From Industry: An Update, 47 Food & Drug L. J. 445 (1992); CEJA, Guidelines on Gifts to Physicians from Industry: An Update, 56 Food & Drug L. J. 27, 33, 35–56 (2001). 64 PhRMA, Code on Interactions with Healthcare Professionals (2008) at 4, http://phrma-­docs. phrma.org/sites/default/files/pdf/phrma_marketing_code_2008.pdf (last accessed Oct. 18, 2018). 65  See, e.g., Sunita Sah & Adriane Fugh-­Berman, Physicians Under the Influence: Social Psychology and Industry Marketing Strategies, 41 J.L. Med. & Ethics 665 (2013); Sergio Sismondo, Key Opinion Leaders and the Corruption of Medical Knowledge: What the Sunshine Act Will and Won’t Cast Light On, 41 J.L. Med. & Ethics 635 (2013); Howard Brody, Hooked: Ethics, the Medical Profession, and the Pharmaceutical Industry (2007). 66  Patient Protection and Affordable Care Act of 2010, Pub. L. No. 111–148, § 6002, 124 Stat. 119, 689 (2010); Open Payments, https://www.cms.gov/openpayments/(last visited Oct. 18, 2018); Richard Saver, Shadows Amid Sunshine: Regulating Financial Conflicts of Interest in Medical Research 145 Chest 379 (2014). For a broader discussion of physician conflicts of interest, see the chapter by Saver in this volume.

Fraud and Abuse Law in the United States   389 Most recently, enforcement efforts have focused on the ways in which manufacturers may promote drugs “off-­label” for indications beyond those approved by the FDA.67 An approved drug may be considered illegally misbranded if its label is false, misleading, or fails to contain adequate directions for use; by definition, labels do not contain directions for uses that have not been approved.68 While physicians are free to prescribe drugs for both approved and unapproved uses, the FDA has long taken the position that manufacturers may not promote their products for unapproved purposes—behavior that, if not explicitly fraudulent, certainly could be characterized as an abusive end-­run around the FDA drug approval process.69 The agency’s efforts to limit off-­label promotion have come under attack in recent years, however, with critics arguing that the policy is an unconstitutional restriction on free speech.70 At present, the issue remains unresolved. Misbranding, in turn, implicates the fraud and abuse laws because FDA approval often forms the basis for Medicare and Medicaid reimbursement. In general, coverage is restricted to “medically accepted” indications, which include not only those for which the drug has been approved by the FDA but also those recognized in specific medical research compendia.71 Absent an exception, other uses of the drug will not be covered, and claims submitted for those uses are requests for payment to which the claimants are not entitled under the FCA. While manufacturers do not usually submit claims to Medicare or Medicaid, the FCA also applies to defendants who “cause” false or fraudulent claims to be submitted.72 When a manufacturer’s off-­label promotion of a drug convinces a physician to prescribe the drug for a non-­covered use and claims are submitted to Medicare or Medicaid, those claims may be considered false and the manufacturer may have violated the FCA by “causing” the submission of false claims.73 Whether the manufacturer’s behavior is motivated by a clear intention to defraud the federal healthcare programs or merely an abusive attempt to use ambiguity to its own advantage, the behavior is subject to suit—and likely a multimillion dollar settlement—under the FCA.74 67  See Joan H. Krause, Truth, Falsity, and Fraud: Off-­Label Drug Settlements and the Future of the Civil False Claims Act, 71 Food & Drug L.J. 401 (2016). 68  21 U.S.C. §§ 331(a), 352(a), 352(f) (2016). An approved drug also may be considered new (and unapproved) if it is promoted for an unapproved use. 21 C.F.R. § 310.3(h)(4) (2018). 69  See United States v. Caronia, 703 F.3d 149, 153 (2d Cir. 2012). 70  Id. at 160. In 2012, the Second Circuit held that the FDA could not prohibit a pharmaceutical sales representative’s truthful statements about off-­label uses for one of his company’s drugs, although the case left open the possibility of prosecutions based on a theory that promotional speech provides evidence that a drug is intended for an off-­label use. United States ex rel. Polansky v. Pfizer, Inc., 822 F.3d 613, 615 n.2 (2d Cir. 2016) (describing Caronia holding). 71  See, e.g., 42 U.S.C. §§ 1396b(i)(1), 1396r-­8(k)(3), 1396r-­8(k)(6) (2018) (defining a medically accepted indication for Medicaid as a use approved by the FDA or “which is supported by one or more citations included or approved for inclusion in” specified compendia); id. § 1395w-­102(e)(4)(A)(ii) (applying same definitions to Medicare Part D). 72  See 31 U.S.C. § 3729(a)(1)(A)-(B). 73  United States ex rel. Franklin v. Parke-­Davis, 147 F. Supp. 2d 39, 52 (D. Mass. 2001) (explaining that “the alleged FCA violations arises—not from unlawful off-­label marketing activity itself—but from the submission of Medicaid claims for uncovered off-­label uses induced by Defendant’s fraudulent conduct”). 74  See Almashat et al., supra note 58 (describing settlements).

390   Joan H. Krause Despite the lingering First Amendment debate, many manufacturers have settled off-­label FCA allegations in recent years. Numerous factors continue to drive these cases, including incentives for both the government and manufacturers to settle FCA litigation, the fact that qui tam relators have many incentives to file these suits, and the economics of FCA recovery, which funnel money to the federal Treasury even where the government declines to intervene.75 In short, the pharmaceutical industry remains a prime target for healthcare fraud enforcement on a wide array of theories.

3.3  The Evolving Enforcement Context Fraud control in the United States is not static: as healthcare fraud has evolved to encompass new types of activities, the enforcement context has evolved as well. Beginning with the enactment of the Health Insurance Portability and Accountability Act (HIPAA) of 1996, the typical legislative response to fraud has been to increase funding and authority for enforcement.76 HIPAA created a new federal Health Care Fraud crime prohibiting defrauding a healthcare benefit program, as well as additional crimes punishing theft or embezzlement, false statements, and obstruction of criminal investigations related to healthcare offenses and expansion of exclusions and civil penalties.77 A key component of HIPAA was the creation of a Health Care Fraud and Abuse Control Program to coordinate federal, state, and local enforcement efforts. The centerpiece of the Control Program is the Health Care Fraud and Abuse Control Account, which uses fraud recoveries to fund new inspections, investigations, and prosecutions—allowing the federal government to allocate more than $1 billion dollars to anti-­fraud activities in fiscal year 2017 alone.78 While an encouraging sign, the amount recovered still pales in comparison to improper payments each year—let alone compared to the unknown true amount lost to fraud. While the logic of ever-­expanding enforcement authority is clear, greater resources and new crimes have not necessarily led to a decrease in healthcare fraud. The federal government has taken active measures to reduce fraud since at least the 1980s, but the amounts recovered have rarely exceeded about $2 billion a year, and estimates of the amount of healthcare expenditures lost to fraud continue to grow. This disconnect may 75  See Krause, Truth, Falsity, and Fraud, supra note 67 at 427–430. 76  Health Insurance Portability and Accountability Act of 1996, Pub. L. No. 104–191, 110 Stat. 1936 (1996). 77  18 U.S.C. §§ 247, 669, 1035, 1347, 1518. 78  Id. § 201. As a result, the DOJ opened 967 new criminal and 948 new civil and healthcare fraud investigations, and it won or negotiated more than $2.4 billion in settlements and judgments in healthcare fraud cases; the government’s estimated return on investment between 2015 and 2017 was $4.20 for every $1.00 spent. See Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2017, supra note 18, at 7–8. After HIPAA, subsequent federal legislation continued this expansion of fraud sanctions and increased the anti-­fraud budget. See, e.g.¸ Balanced Budget Act of 1997, Pub. L. No. 105–133, 111 Stat. 251, §§ 4304(b), 4314, 4331(c); Deficit Reduction Act of 2005, Pub. L. No. 109–171, § 6031, 120 Stat. 4 (2006).

Fraud and Abuse Law in the United States   391 in part be due to the fact that, traditionally, anti-­fraud measures have focused almost exclusively on ex post enforcement, identifying improper claims and then attempting to reclaim the diverted funds—a strategy that may generate repayment from legitimate providers who are constant presences in the market, such as long-­standing hospitals, but one that has little chance of success against a fraudulent enterprise such as a sham “pharmacy” whose owners have long since pulled up stakes and disappeared. As one critic notes, “[a] strategy that relies on the capacity of the justice system is necessarily a limited one. . . . Many fraud problems can be eliminated most efficiently by using methods other than enforcement post factum.”79 Attacking healthcare fraud by dedicating increasing funds to enforcement is a politically palatable response, but preventing healthcare fraud over the long run also requires better ex ante control over who is allowed to file claims in the first place and whether those claims are paid.80 Congress appeared to heed that message with the enactment of major health reform legislation in 2010. The ACA and the accompanying Health Care and Education Reconciliation Act represented one of the first committed attempts to prevent the fraudulent diversion of program money at both the front and back ends of the funding stream.81 While the statutes continued the trend of increasing the fraud enforcement budget and amending the major fraud laws to close perceived loopholes, they also contained a variety of provisions designed to provide better control over persons who submit bills to the federal healthcare programs and more timely assessment of claims: for example, mandating additional background screening for certain categories of high-­ risk providers, reducing the time frame for the submission of Medicare claims, and requiring additional documentation for referrals for services posing a high risk of fraud.82 The reliance on both fraud prevention and interagency cooperation was also reflected in the creation of the Obama Administration’s Health Care Prevention and Enforcement Action Team (HEAT) in 2009, a cabinet-­level joint Department of Health and Human Services (HHS) and DOJ initiative targeting major areas of healthcare fraud risk. HEAT was designed to build on the Medicare Strike Force model, with targeted criminal investigations focusing on high-­risk items and services (such as durable medical equipment) in regions with high rates of healthcare fraud (such as South Florida). By using state-­of-­the-­art technology to analyze electronic claims data for patterns that might indicate fraud in as close to real-­time as possible, HEAT exemplified the focus on ex ante enforcement.83 Over time, these latter changes may well provide greater success in the fight against healthcare fraud than increased penalties alone.

79  Sparrow, supra note 6, at 213. 80  Id. at 222–224 (noting importance of early detection and use of prepayment controls). 81  Patient Protection and Affordable Care Act of 2010, Pub. L. No. 111–148, 124 Stat. 119 (2010); Health Care and Education Reconciliation Act of 2010, Pub. L. No. 111–152, 124 Stat. 1029 (2010). 82  See Krause, Following the Money in Health Care Fraud, supra note 16. 83  See Health Care Fraud and Abuse Control Program Annual Report for Fiscal Year 2017, supra note 18, at 9–12 (recent HEAT activity); Medicare Fraud Strike Force, supra note 19.

392   Joan H. Krause

4 Conclusion Healthcare fraud is historically, socially, and politically constructed, manifesting in distinct ways that reflect differences not only in healthcare financing but also in a nation’s priorities, values, and cultural norms. Nevertheless, healthcare fraud remains a worldwide concern—perhaps simply because, as one commentator wryly noted, “[t]hat’s where the money is.”84 Regardless of the structure of the healthcare system, there will be opportunities to divert funds for personal advantage. Healthcare fraud has proven remarkably dynamic and persistent, adapting to changes in both fraud detection and reimbursement models. That is as true in the United States as anywhere else in the world—perhaps even more so, given the complexity and fragmentation of the country’s healthcare system. Adapting the traditional ex post fraud enforcement model to these ever-­changing realities will remain the challenge for the foreseeable future.

84 Pamela H. Bucy, Crimes by Health Care Providers, 1996 U. Ill. L. Rev. 589.

chapter 21

Hea lthca r e Fr au d a n d A buse i n Eu rope Tracey A. Elliott

1  Introduction and Context In common with the United States, healthcare fraud presents a significant problem in Europe, with it being estimated that, in the UK alone, fraud costs the National Health Service (NHS) £1.27 billion a year1 and that the level of fraud could amount to more than 30% of European Union (EU) healthcare budgets.2 However, the continent of Europe consists of at least 50 recognised sovereign states, 27 of these being members of the EU, each with its own healthcare system, legal system, policing organisations, and regulatory frameworks. If ascertaining a clear picture of the problem of healthcare fraud and abuse is a complex and difficult task in relation to the United States, it is even more so in relation to Europe. Even a cursory examination of European geography, societies, economies, healthcare systems, and legal regimes highlights that it is impossible to identify an overarching ‘European approach’ to the ethical, definitional, and operational challenges that make healthcare fraud difficult to define, identify, police, and prevent. Whilst healthcare fraud is a problem in all European states, differences in historical, cultural, and political circumstances; the level and organisation of funding; legal regulation; and enforcement mechanisms mean that the scope of the problem varies among countries.3 Fraud, in common with other forms of criminal offending, is not a crime that exists in a vacuum. It is historically, politically, and socially situated: the ‘fraud problem’ in a state may be seen as reflecting the society within which it exists. The annual corruption index 1  NHS Counter Fraud Authority (NHSCFA), https://cfa.nhs.uk/ 2  J Bobek et al., Study on Cross-­Border Cooperation: Capitalising on Existing Initiatives for Cooperation in Cross-­Border Regions, European Commission (2018), http://dx.doi. org/10.2875/825256, p. 8. 3  Cf. European Commission (EC), Study on Corruption in the Healthcare Sector, HOME/2011/ISEC/ PR/047-­A2(2013), p. 9.

394   Tracey A. Elliott produced by Transparency International,4 which measures the extent to which corruption is perceived to exist among a country’s public officials and politicians, indicates that, although corrupt practices exist in all states, levels vary from country to country. The 2018 index5 indicates that Scandinavian and Western European countries have generally fairly low levels of public corruption.6 By contrast, Mediterranean, Central, and Eastern European states have higher levels of corruption,7 with Ukraine and Russia emerging as the most corrupt.8 As Savedoff and Hussman have observed, the extent of corrupt practices in healthcare systems reflects the societies within such systems operate, being ‘less likely in societies where there is broad adherence to the rule of law, transparency and trust, and where the public sector is ruled by effective civil service codes and strong accountability mechanisms’.9

2  Healthcare Funding in Europe and Fraud Whilst all healthcare systems are vulnerable to fraud and abuse, the manner in which states manage the financing of their healthcare systems and the payment of healthcare professionals is significant because it may affect the forms of abusive conduct which occur. Transparency International has identified tax-­funded systems, as in the UK, Ireland, Italy, Spain, Portugal, and the Nordic Countries, as being vulnerable to procurement corruption, the large-­scale diversion of funds at the ministerial level, and theft by 4  Transparency International is an organisation that works with governments to stop abuses of power, bribery, and corrupt secret deals: https://www.transparency.org 5  The annual Corruption Perceptions Index ranks 180 countries and territories by their perceived levels of public-­sector corruption according to experts and businesspeople, using a scale of 0 to 100, where 0 is highly corrupt and 100 is very clean: https://www.transparency.org/cpi2018 6  Thirteen of the top twenty countries being from these geographical areas. The positions and scores were as follows: Denmark (1: 88/100), Finland, Sweden and Switzerland (3: 85/100), Norway (7: 84/100), the Netherlands (8: 82/100), Luxembourg (9: 81/100), Germany and United Kingdom (11: 80/100), Austria and Iceland (14: 76/100), Belgium (17: 75/100), Ireland (18: 73/100), and France (21: 72/100): https://www.transparency.org/cpi2018 7  Of these states, central and southeastern European countries emerge from the index as having markedly more public corruption than the majority of their Western European neighbours: Portugal (30: 64/100); Poland and Slovenia (36: 60/100); Cyprus, Czech Republic, and Lithuania (38; 59/100); Latvia and Spain (41: 58/100); Malta (51: 54/100); Italy (53: 57/100); Slovakia (57: 50/100); Croatia (60: 48/100); Romania (61: 47/100); Hungary (64: 46/100); Greece and Montenegro (67: 45/100); Belarus (70: 44/100); Bulgaria (77: 42/100); Serbia (87: 39/100); Bosnia and Hercegovina (89: 38/100); Kosovo and Macedonia (93: 37/100) and Albania (99: 36/100). Of the former Soviet Bloc states, only Estonia was placed in the top twenty (18: 73/100). 8  With Ukraine placed 120th with a score of 32/100 and Russia 144th with 27/100: https://www. transparency.org/cpi2018 9  W Savedoff and K Hussman, ‘Why Are Healthcare Systems Prone to Corruption?’, Transparency International, Global Corruption Report 2006, 4, https://www.transparency.org/whatwedo/publication/ global_corruption_report_2006_corruption_and_health

Healthcare Fraud and Abuse in Europe   395 employees. These systems also suffer from doctors wrongfully charging patients, diverting them to private practice, or using public facilities for private practice.10 By contrast, in public or private insurance-­based systems, such as in Germany, the Czech and Slovak Republics, Austria, France, the Netherlands, and Switzerland, where the body financing the services is distinct from the provider of the services, there is greater vulnerability to fraud in billing for services, overtreatment, and the diversion of funds.11 Furthermore, whilst arrangements made by states to pay healthcare providers may be intended to reward and incentivise good practice, the different payment systems (salary,12 fee-­for-­service,13 capitation,14 prospective payments) are all vulnerable to abuse by healthcare providers, albeit via different methods. The payment of low salaries may encourage healthcare professionals to seek informal or ‘under-­the-­table’ payments or to otherwise boost their income by dishonest means. Fee-­for-­service systems may lead to overbilling or the overtreatment of patients. Capitation, prospective payments, and other forms of lump-­sum reimbursement may provide an incentive to reduce the amount and quality of service provided to patients: providing shorter consultations, ‘cherry-­picking’ healthier patients who require fewer consultations, and overreferring to other parts of the system those patients who require more physician time and attention.15

3  The Legal Regulation of Fraud and Abuse and Definitional Issues 3.1  An Outline of Laws Commonly Used to Regulate Fraud Whilst fraudulent conduct is criminalised in all European countries, and fraud is recognised as an offence that gives rise to surrender for extradition under a European Arrest 10  Id., 11–12. 11  Id., 12–13; EC, supra n. 3, 25–26. 12  Greece, Iceland, Portugal, and Sweden. 13  Belgium, France, Germany, Ireland, Luxembourg, and Switzerland. 14  Hungary, Poland, and the Slovak Republic. 15  For discussion of the potential risks involved in the different payment systems, see V Paris, M Devaux, and L Wei (2010), ‘Health Systems Institutional Characteristics: A Survey of 29 OECD Countries’, OECD Health Working Papers, no. 50, OECD Publishing, Paris; http://dx.doi. org/10.1787/5kmfxfq9qbnr-­en, p. 33; L Rochaix, ‘Performance-­Tied Payment Systems for Physicians’, in RB Saltman, JM Figueras, and C Sakellarides, Critical Challenges for Health Care Reform in Europe (Open University Press 1998), 196; M Grignon and D Polton, Influence of Physician Payment Methods on the Efficiency of the Health Care System (2002) Commission on the Future of Health Care in Canada, Working Paper No.35, https://www.researchgate.net/profile/Michel_Grignon/publication/268044081_ Influence_of_Physician_Payment_Methods_on_the_Efficiency_of_the_Health_Care_System/ links/54db96040cf28d3de65bbd99.pdf; S Simoens and A Guiffrida, ‘The Impact of Physician Payment Methods on Raising the Efficiency of the Healthcare System’ (2004) 3 Appl Health Economics Health Pol 39; S Simoens and J Hurst, The Supply of Physician Services in OECD Countries (2006) OECD Working Paper 21, DELSA/HEA/WD/HWP(2006)1, www.oecd.org/health/health-­systems/35987490. pdf, pp. 44–49.

396   Tracey A. Elliott Warrant without the need for proof of double criminality,16 there is no common template for the structure of fraud offences in Europe, and legal definitions of fraud vary between countries. Nearly all European States have dedicated fraud offences incorporated into their Criminal Codes, but the structure of these offences varies. For example, the French Criminal Code contains a specific offence of ‘fraudulent obtaining’,17 but the offence of theft may also be used to prosecute the fraudulent appropriation of property,18 and there is a separate offence concerning fraud in relation to tenders.19 The German Criminal Code contains a dedicated fraud offence20 but also a range of offences relating to specific types of fraud, for example: computer fraud, benefit or subsidy fraud, and insurance fraud.21 It has been suggested that the criminal codes of the Netherlands22 and Lithuania23 ‘do not qualify fraud as an offence’,24 although both codes contain general deception offences25 which, together with an array of alternative offences, some of which are highly specific, may potentially be used to prosecute fraudulent or dishonest behaviour. For example, the Dutch Criminal Code contains offences relating to forgery,26 the intentional provision of information to a person granting a benefit payment or allowance;27 theft; and aggravated theft.28 16  Provided the offence is punishable in the Member State issuing the warrant by a custodial sentence or a detention order for a minimum period of at least three years and is defined by the law of that state: European Council Framework Decision 2002/JHA, Article 2. Cf. Extradition Act 2003, Schedule 2. 17  Article 313–1. The French Criminal Code (updated 24th November 2019) may be accessed at https:// www.legislationline.org/download/id/8442/file/France_CC_am24112019_fr.pdf. This offence requires that the defendant intentionally deceives the victim, causing them ‘to transfer funds, valuables or any property, to provide a service or to consent to an act incurring or discharging an obligation’. Article 313–2 provides for increased sentences in certain circumstances of aggravation, including where the fraudulent obtaining was committed to the prejudice of a person whose particular vulnerability due to age, sickness, infirmity, or a physical or psychological disability or to pregnancy, is apparent or known to the perpetrator. 18  Article 1: ‘Theft is the fraudulent appropriation of a thing belonging to another person’ (‘Le vol est la soustraction frauduleuse de la chose d'autrui’). 19  Article 313–316. 20  German Criminal Code (2013), Section 263. This criminalises damaging another’s property by causing or maintaining an error by misrepresentation or by the distortion or suppression of the truth with intent to obtain for oneself or a third person an unlawful material benefit. An English translation of the Code may be accessed at https://www.gesetze-­im-­internet.de/englisch_stgb/englisch_stgb.html#p0026 21  Sections 263a, 264, and 265. 22  An English translation of the Dutch Criminal Code (2012) may be accessed at www.ejtn.eu/ PageFiles/6533/2014 seminars/Omsenie/WetboekvanStrafrecht_ENG_PV.pdf 23  An English translation of the Lithuanian Criminal Code may be accessed at www.lithuanialaw. com/lithuanian-­criminal-­code-­495 24  J Boertjens, M Mikkers, W Sauter, and P Vincke, ‘Introduction’, in M Mikkers, W Sauter, P Vincke and J Boertjens, eds., Healthcare Fraud, Corruption and Waste in Europe (Eleven 2017) 6. 25  Section 326, Dutch Criminal Code; Article 182, Lithuanian Criminal Code. 26  Section 227(1) (providing a false statement in relation to a fact recorded in an authentic document which serves to verify the truthfulness of the statement with the intention that the document be used as if the statement were true), section 228 (intentionally issuing a false certificate of a birth or of the cause of death or of the existence or nonexistence of diseases, infirmities, or disabilities by a doctor). The provision of false certificates by other persons is criminalized by section 229. 27  Section 227a. 28  Sections 310–312. Other offences which might potentially be used to prosecute dishonest, corrupt, and fraudulent behaviour are bribery and corruption (Sections 177, 177a and 178); making, falsifying or

Healthcare Fraud and Abuse in Europe   397 In most European states, the substantive offence of fraud is constructed as a result crime, which requires the prosecution to prove that the specified result or consequence (e.g., the obtaining of property or of a service or discharging of an obligation) occurred.29 By contrast, the offence of fraud by false representation created by the UK’s Fraud Act 2006 (s.2),30 is an ‘incredibly broad’,31 conduct-­based offence, which is committed as soon as the false representation32 is made dishonestly33 with intent to make gain for oneself or another or to cause loss or expose another to a risk of loss. The prosecution does not need to establish that there was any gain as a result of the representation or that the representation was acted upon or even believed.34 The offence is essentially inchoate in nature35 and therefore relatively easy to prove. More controversially,36 in cases not using a false certificate of good conduct, competence, poverty, disabilities, or other circumstances with the intention that it be used in order to obtain employment or to evoke goodwill and assistance (Section 239); intentionally making, falsifying, or using a false cash card, stored value card, any other card available to the public or an identity data carrier available to the public that is intended for making or obtaining automated payments or other services with the intention of benefiting oneself or another (Section 232); and embezzlement (Sections 321–322). The Lithuanian Criminal Code contains offences of swindling (Article 182), theft (Article 178), extortion of property (Article 181), misappropriation of property (Article 183), and causing property damage by deceit (Article 186). 29  AP Simester, JR Spencer, F Stark, GR Sullivan, and GJ Virgo, Simester and Sullivan’s Criminal Law, 7th edition (Oxford 2019) 73. See, e.g., French Criminal Code, Article 313-­1; German Criminal Code, section 263; Italian Criminal Code, Article 640; Spanish Criminal Code, Article 248; Greek Criminal Code, Article 386; Polish Criminal Code, Article 286. 30  The Fraud Act 2006 does not apply in Scotland save for s.10(1): s.15(2). In Scotland, the main offences used to prosecute fraud are common law fraud (which requires that a practical result be achieved by means of a false pretence, uttering (the use of a forged document), embezzlement (which usually involves the appropriation of money), and a variety of statutory provisions, including the Criminal Justice and Licensing (Scotland) Act 2010, s.49 (articles for use in fraud). 31  D Ormerod and K Laird, Smith, Hogan and Ormerod’s Criminal Law, 15th edition (Oxford University Press 2018) 936. 32  A representation is false if it is untrue and misleading and the maker know that it is, or may be, untrue or misleading: s.2(2). 33  Dishonesty is not defined in the Act. Following the decision of the Supreme Court in Ivey v Genting Casinos UK Ltd (t/a Crockfords Club) [2017] UKSC 67, [2018] AC 391, the test for dishonesty is objective. There is no requirement that a defendant appreciate that he is acting dishonestly: R v Bolton (David) and Booth (Rosemary) [2020] EWCA Crim 578. This decision has been subject to criticism: D Ormerod and K Laird, ‘Ivey v Genting Casinos: Much Ado About Nothing?’ [2017–2018] 9 The UK Supreme Court Yearbook 380; M Dyson and P Jarvis, ‘Poison Ivey or Herbal Tea Leaf?’ (2018) 134 LQR 198; G Virgo, ‘Case Comment: Cheating and Dishonesty’ (2018) 77(1) Cambridge L J 18. 34  Ormerod and Laird, supra n. 31, 937; D Ormerod, ‘The Fraud Act 2006: Criminalising Lying?’ (2007) Crim LR 193. 35  J Horder, ‘Bribery as a Form of Criminal Wrongdoing’ (2011) 127 LQR. 37, 41–42; A Ashworth and L Zedner, Preventive Justice (Oxford University Press 2014) 97–98. 36  The offence has been criticised as being ‘excessively broad, vague’ and offending ‘against the principles of legality, certainty and fair warning’: Ormerod and Laird, supra n. 31, 461. The Law Commission recommended that the offence be abolished (Fraud, Law Com. No. 276 [2002] para 9.5), but the offence was retained because of fears that unforeseen gaps in the new statutory scheme might emerge (Hansard [HC], 13 March 2006 col.1110). In 2012, the Ministry of Justice concluded that the offence should be retained because it ‘continues to be an effective and essential tool in combating fraud’: Post-­legislative Assessment of the Fraud Act 2006, Cm.8372 (2012), para. 42.

398   Tracey A. Elliott covered by the Fraud Act 2006,37 prosecutors in England and Wales may still have recourse to the common law offence of conspiracy to defraud, which may be used to prosecute dishonest agreements by two or more ‘to deprive a person of something which is his or to which he is or would be or might be entitled’.38 For example, in R v Barton and Booth,39 the offence was used to prosecute the operator of a luxury nursing home who, with the assistance of his general manager, had used his position to dishonestly target, befriend, and ‘groom’ wealthy and vulnerable elderly residents in order to profit from them. Following the example set by the United States, since 2013, European countries have been passing legislation to permit the use of negotiated settlements, Deferred Prosecution Agreements (DPAs), to tackle serious corporate economic crime. DPAs are essentially ‘a bargain under which the prosecutor undertakes not to proceed with the prosecution of a corporation for a fixed time in return for the defendant mending its ways and paying a financial penalty for the privilege’.40 England and Wales were the first European countries to adopt DPAs, in the Criminals and Courts Act 2013, Schedule 17, which came into effect in February 2014,41 and none of the six court-­approved DPAs to date has involved healthcare or pharmaceutical corporations.42 In 2016, France enacted the Sapin II law,43 which introduced a version of a DPA scheme, the Convention judiciaire d’interêt public (CJIP).44 In the first two years of the scheme, five CJIPs were approved for offences involving the bribery of public officials and aggravated money laundering.45 It is likely that other European states will follow the example of France and 37  It appears that dishonestly failing to fulfil a contractual obligation, deception for a nonfinancial purpose (e.g., obtaining confidential information), and dishonestly infringing another’s legal right would not be covered by the 2006 Act: Home Office, Fraud Law Reform (2004) para. 38. 38  Scott v Metropolitan Police Commissioner [1975] AC 819, 840. 39  [2020] EWCA Crim 575. 40  House of Lords Select Committee on the Bribery Act 2010, The Bribery Act 2010: Post-­Legislative Scrutiny, HL Paper 303 (2019) para. 233. 41  The Serious Fraud Office (SFO) and Crown Prosecution Service (CPS) have issued a Code of Practice in relation to the scheme: SFO/CPS, Deferred Prosecution Agreements Code of Practice: Crime and Courts Act 2013 (2014), https://www.cps.gov.uk/sites/default/files/documents/publications/ dpa_cop.pdf 42  Serious Fraud Office v Standard Bank PLC [2016] Lloyds Rep. F.C. 102; Serious Fraud Office v XYZ Ltd. [2016] Llyoyds Rep. F.C. 509; Serious Fraud v Rolls Royce PLC [2017] Lloyds Rep. F.C. 249; Serious Fraud Office v Tesco Stores Ltd [2019] Lloyds Rep. F.C. 283; Serious Fraud Office v Serco Geographix Ltd. [2019] Lloyds Rep. F.C. 518; Director of the Serious Fraud Office v Airbus SE [2020] 1 WLUK 435, Southwark Crown Court, Case No.U20200108. See C King and N Lord, ‘Deferred Prosecution Agreements in England & Wales: Castles Made of Sand? (2020) Public Law 307. 43  Loi Sapin II pour la transparence de la vie économique (law no 2016-­1691). 44  In force 1 June 2017. On the 27 June 2019, the Agence Francaise Anticorruption and financial crime enforcement agency Parquet National Financier (PNF) published guidelines in relation to CJIPs: Lignes directrices sur la mise en œuvre de la convention judiciaire d’interet public (2019) https://www. agence-­francaise-­anticorruption.gouv.fr/files/files/Lignes%20directrices%20PNF%20CJIP.pdf 45  Clifford Chance, First Guidelines on CJIP (2019), https://www.cliffordchance.com/content/dam/ cliffordchance/briefings/2019/07/first-­guidelines-­on-­cjip.pdf, p. 5; A Kirry, FT Davis, and A Bisch, Financial Crime in France: Overview (2020) Thomson Reuters Practical Law, https://uk.practicallaw. thomsonreuters.com/2-­519-­9872?transitionType=Default&contextData=(sc.Default)&firstPage=true&b hcp=1. None of these involved pharmaceutical or healthcare companies.

Healthcare Fraud and Abuse in Europe   399 the UK and establish DPA schemes to resolve corporate criminal investigations: in 2018, the Law Reform Commission of Ireland recommended the adoption of a DPA system based on the UK model,46 and the Office of the Attorney General of Switzerland (OAG) proposed the inclusion of a new Article 318bis to introduce a DPA scheme in the Criminal Procedure Code.47 Pan-­ European measures to tackle fraud and abuse include the MEDICRIME Convention (the Council of Europe Convention on the Counterfeiting of Medical Products),48 which provides a binding criminal law treaty aimed at tackling fraud in relation to medicines and medicinal devices. The Convention requires States Parties to adopt domestic legislation which criminalises the counterfeiting of medical products or devices or documents relating to these products and supplying, offering to supply, and trafficking such products or devices. It also provides a framework for national and international cooperation, national coordination, preventive measures to be used by public and private sectors and the protection of victims and witnesses.49 In addition, EU legislation50 regulates online pharmacies by requiring them to be authorised by a ‘competent authority’ in the relevant Member State51 and to display on every page of their website a European common logo containing a hyperlink to their entry in the list of registered online sellers so that customers are able to check that they are buying from a legitimate, registered seller.52 Civil law actions may also be brought to recover financial compensation in respect of fraud and abuse. For example, in 2003, the UK Department of Health brought civil actions against seven pharmaceutical companies53 for alleged anti-­competitive price 46  Law Reform Commission of Ireland, Report on Regulatory Powers and Corporate Offences, LRC 119–2018 (2018) 264–279; King and Lord, supra n. 42. 47  AM Garbarski and AV Julen Berthod, Chambers & Parties: White Collar Crime 2019: Switzerland: Trends and Practices (2019), https://practiceguides.chambers.com/practice-­guides/white-­collar-­ crime-­2019/switzerland/trends-­and-­developments; D Igra, ‘Global Spread of DPA Regimes: What They Mean for Multinationals’ (2020) 13(50) IICJ 6419. 48  Council of Europe Convention on the counterfeiting of medical products and similar crimes involving threats to public health, CETS No.211, available at https://www.coe.int/en/web/conventions/ full-­list/-­/conventions/treaty/211. The Convention was signed on 28 October 2011 and came into force on 1 January 2016. To date it has been ratified by Albania, Armenia, Belgium, Croatia, France, Hungary, Moldova, Portugal, Russian Federation, Spain, Switzerland, Turkey, and Ukraine. 49  CM Romeo Casabona, AU Mora, PN Jiménez, and A Alarcón-­Jiménez, ‘International Strategies in Fighting Against Medicaments Fraud and Other Similar Offences. The MEDICRIME Convention ’ (2017) 68 Crime, Law and Social Change 95. 50  Falsified Medicines Directive 2011/62/EU; Regulation EU/699/2014. In the UK, these provisions have been brought into effect via the Human Medicines (Amendment) Regulations 2013 (SI 2013/1855). 51  Directive 2011/62/EU (6). The relevant ‘competent authority’ in the UK is the Medicines and Healthcare products Regulatory Authority (MHRA): Human Medicines Regulations 2012, Regulations 256A–­256N. 52  Directive 2011/62/EU (25). 53  Goldshield Group, Kent Pharmaceuticals, Norton Healthcare, Generics UK, Ranbaxy UK, Regent-­GM Laboratories, and DDSA Pharmaceuticals. A criminal prosecution of Goldshield Group, Kent Pharmaceuticals, Norton Healthcare, Generics UK, Ranbaxy UK, and nine employees or directors of these companies alleging conspiracy to defraud failed: R v Goldshield Group Plc [2008] UKHL 17, [2009] 1 WLR 453.

400   Tracey A. Elliott fixing in connection with the supply to the NHS in England of generic drugs. The claims were settled without liability being admitted.54 In Continental Europe, healthcare funds,55 or supervisory authorities,56 commonly have the power to take administrative measures to recover damages and, in appropriate cases, to take other necessary measures, such as terminating contracts and reporting healthcare practitioners to regulatory bodies.57

3.2  Definitional Issues In the Introduction, it was noted that the absence of a single ‘European approach’ towards fraud and abuse creates potential for these terms to be interpreted in­con­sist­ently, causing possible confusion.58 The European Healthcare Fraud & Corruption Network (EHFCN) created the EHFCN Waste Typology Matrix in an attempt to promote a consistent approach when types of healthcare fraud are being described or reported.59 Whilst this is laudable, and the Waste Typology Matrix has been adopted by the Medical Evaluation and Inspection Department (MEID) of the National Institute for Health and Disability, which plays the leading role in counter-­healthcare fraud activities in Belgium,60 it has not universally been adopted, and different national approaches to the categorization of types of fraud and abuse still exist.61 For, example, the French universal sickness insurance scheme, Assurance Maladie, has its own scheme of categorization, distinguishing between ‘fraud’, ‘abusive activities’, and ‘negligent activities’,62 whilst the NHS in the UK uses more than 130 classifications to identify and define instances of fraud.63 Even where states have clear legal norms which may be used to criminalise fraudulent conduct, definitional problems arise in ‘grey areas’ where disputes arise as to whether 54  See, e.g., BBC News, ‘Norton Settles Price Fixing Claim’ (2006) April 4; C Dyer, ‘Prosecution over Drug Price Fixing Collapses’ (2008) 337(7662) BMJ 131; Guardian Online, ‘Drugs Firm Agrees Payout over Price-­Fixing Claim’ (2007) June 22, https://www.theguardian.com/society/2007/jun/22/health.uknews. Similar actions brought by the health authorities in Scotland and Northern Ireland were also settled; see, e.g., K Attwood, ‘Goldshield Settles £1m “Price-­Fixing” claim’ (2008) The Independent March 6. 55  E.g., the Zavod za zdravstveno zavarovanje Slovenije (ZZZS) in Slovenia: B Kramberger, ‘Slovenia’, in Mikkers, Sauter, Vincke, and Boertjens, eds., supra n. 24, 237, at 247. 56  E.g., the Dutch Healthcare Authority (NZa) in the Netherlands and L’Assurance Maladie in France; see J Boertjens, ‘The Netherlands’; and DACCRF, ‘France’ in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 191, at 199 and 155, at 161, respectively. 57  Boertjens, Wikkers, Sauter, and Vincke, supra n. 24, 13. 58  Elliott and Krause, ‘Introduction: Fraud and Abuse Law’, supra n. 58, fn 11–17. 59  Elliott and Krause, supra n. 58; EHFCN, The EHFCN Waste Typology Matrix, http://www.ehfcn. org/what-­is-­fraud/ehfcn-­waste-­typology-­matrix/; Boertjens, Wikkers, Sauter, and Vincke, supra n. 24, 9. 60  P Vincke, ‘Belgium’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 173, 177–179. 61  Boertjens, Wikkers, Sauter, and Vincke, supra n. 24, 9. 62  Elliott and Krause, supra n. 58, fn. 17; L’Assurance Maladie, CIR:1/2012, www.mediam.ext.cnamts. fr/ameli/cons/CIRCC/2012/CIR-­1-­2012.PDF 63  G Brooks, M Button, M Tunley, and J Gee, ‘United Kingdom’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 137, 144.

Healthcare Fraud and Abuse in Europe   401 conduct may be considered to be abusive or fraudulent.64 For example, in what may be regarded as overtreatment, evidence suggests that there are wide variations in the use of surgical procedures,65 testing, and treatment66 between and even within states. Medical evidence may be contradictory, contested, limited, or simply not understood by practitioners.67 Patients may pressurize doctors to prescribe drugs or request interventions that are not strictly necessary.68 In the light of all of these factors, the point at which the provision of ineffective or excessively expensive medical treatment should be considered fraudulent may not be clear. Whether overtreatment is regarded as fraud is likely to vary depending on the medical specialty involved, domestic professional codes, and social perceptions of what conduct should be criminalised.69 Generally, treatment will be regarded as overtreatment based on a failure to follow relevant medical standards and, in the absence of factors which may be regarded as demonstrating culpable intention, such as deception, or repetition,70 may fall into a ‘grey zone’ whereby, in certain circumstances, it may be regarded as abuse but in others, it may be regarded merely as error or waste.71 For example, in France, treatment beyond a patient’s medical needs is generally regarded as abuse,72 whereas in Poland, the prescribing of drugs in amounts which exceed the therapeutic requirements of the patient is regarded as error.73

4  Ethical and Operational Issues 4.1  Ethical Issues Unsurprisingly, healthcare fraud is commonly viewed as a crime against property because the most evident harm caused by the crime will be the interference with the 64  P Vincke and J Cylus, ‘Health Care Fraud and Corruption in Europe: An Overview’ (2011) 17(4) Eurohealth 14, 15. 65  See, e.g., MC Tayade and SD Dalvi, ‘Fundamental Ethical Issues in Unnecessary Surgical Procedures’ (2016) 10(4) J Clin Diagn Res JE01-­JE04; JD Birkmeyer, BN Reames, P McCulloch, AJ Carr, WB Campbell, and JE Wennberg, ‘Understanding regional variation in the use of surgery’ (2013) 382(9898) Lancet 1121–1129. 66  See, e.g., K Størdal, C Wyder, A Trobisch, Z Grossman, and A Hadjipanayis, ‘Overtesting and Overtreatment: Statement from the European Academy of Paediatrics (EAP)’ (2019) 178(12) Eur J Pediatr 1923. 67  P Ryan, ‘Transcending Fraud: Clinical Uncertainty, Cognitive Limitations and Medical Overuse’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 59. 68  Id., 72–74. 69  Id., 62, 66–69. 70  E.g., in the Netherlands, the consistent overhospitalization of patients following minor procedures (such as varicose vein treatment) may be regarded as fraud: Boertjens, supra n. 56, 197. 71 Vincke, supra n. 60, 179. 72 DACCRF, supra n. 56, 159. Cf. the position in Italy, where the provision of unnecessary health services (e.g., medically unnecessary caesarean sections) is likely to be regarded as abuse or waste: F Fantoni and F Macchia, ‘Italy’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 255, 270. 73  The Ministry of Health of Poland, ‘Poland’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 207, 216.

402   Tracey A. Elliott property rights of the owner.74 There is a tendency to regard the harms caused by nonviolent property offences as being less serious and therefore less morally blameworthy than offences against the person,75 but this is not always the case. In particular, the impact of healthcare fraud extends beyond financial loss. In systems with a constrained budget, the diversion of funds allocated for healthcare into the hands of fraudsters will decrease the money available to treat patients and is likely to increase healthcare rationing. In extreme cases, it may even threaten the provision of universal health coverage.76 The poor governance which tends to go hand in hand with higher levels of corruption in state healthcare systems may have a wide variety of implications for patient health, including, it has been suggested, higher rates of antibiotic resistance77 and child mortality.78 The provision of unnecessary treatment79 and sale of counterfeit medicines also have the potential directly to harm the health of patients.80 Whilst ethical considerations in relation to the culpability of offenders may shape the priorities of investigators, influence prosecutors when charging decisions are made,81 74  More generally, property crimes also threaten the system of property rights and the security of property rights: J Feinberg, Harm to Others: The Moral Limits of the Criminal Law (Oxford University Press 1987), 63. Cf. SP Green, ‘Property Offences’, in MD Dubber and T Hörnle eds., The Oxford Handbook of Criminal Law (Oxford University Press 2014), 787; AP Simester and GR Sullivan, ‘On the Nature and Rationale of Property Offences’, in RA Duff and SP Green eds., Defining Crimes: Essays on the Special Part of the Criminal Law (Oxford University Press 2005), 170, 186. 75 Green, supra n. 74, 787. 76  Transparency International, The Ignored Pandemic: How Corruption in Healthcare Service Delivery Threatens Universal Health Coverage (2019), http://ti-­health.org/content/the-­ignored-­pandemic/ 77  P Collignon, P Athukorala, S Senanayake, and F Khan, ‘Antimicrobial Resistance: The Major Contribution of Poor Governance and Corruption to This Growing Problem’ (2015) 10(3) PLoS ONE:e0116746; http://dx.doi.org/10.1371/journal.pone.0116746 78  M Hanf, A Van-­Melle, F Fraisse, A Roger, B Carme and M Nacher, ‘Corruption Kills: Estimating the Global Impact of Corruption on Children Deaths’ (2011) 6(11) PLoS ONE e26990. 79  See, e.g., R v Paterson [2017] EWCA Crim 1625, [2018] 4 WLR 109 (Surgeon conducted unnecessary biopsies and mastectomies on patients and was convicted of 17 counts of wounding with intent, contrary to s.18 Offences Against the Person Act 1861, and 3 counts of unlawful wounding (s.20)); Appleton v Garrett [1996] PIQR P1 (Dentist damaged healthy teeth through unnecessary treatment and was found liable to pay aggravated damages for civil trespass to the person). 80  See, e.g., A O’Hagan and A Garlington, ‘Counterfeit Drugs and the Online Pharmaceutical Trade: A Threat to Public Safety’ (2018) 6(3) Forensic Res Criminol Intl J 151; E Taylor, AC Banyard, H Bourhy, F Cliquet, DL Horton, RS Mani, T Müller, CE Rupprecht, MJ Schnell, V Del Rio Vilas, and AR Fooks, ‘Avoiding Preventable Deaths: The Scourge of Counterfeit Rabies Vaccines’ (2019) 37(17) Vaccine 2285. 81  Although the extent of prosecutorial discretion in relation to charging decisions varies between European states. Whereas in some (e.g., England and Wales) prosecutors enjoy a good deal of discretion in deciding whether an accused person is to be prosecuted and what charges are to be brought (see Crown Prosecution Service (CPS) Code for Crown Prosecutors 2018 (2018) https://www. cps.gov.uk/publication/code-­crown-­prosecutors-­2018-­downloadable-­version-­and-­translations), in some prosecutors have only limited discretion not to prosecute where there is evidence to support a charge (e.g., Italy, Germany, and the Netherlands), whilst some countries retain the principle of mandatory prosecution (e.g., Russia and Bulgaria). See, e.g., the International Criminal Court (ICC) table (ICC-­02/17-­134-­AnxI 06-­12-­2019 1/18 NM PT OA OA2 OA3 OA4) at https://www.icc-­cpi.int/ RelatedRecords/CR2019_07328.PDF; E Luna and M Wade, The Prosecutor in Transnational Perspective (Oxford University Press 2012).

Healthcare Fraud and Abuse in Europe   403 and will be considered post-­conviction at the sentencing stage,82 where they conflict with pragmatic considerations, the latter may prevail. In Eastern European countries where the health system is underfunded and there is a shortage of doctors, prosecutors may choose not to crack down on doctors who commit offences of dishonesty such as charging for more hours than are actually worked83 or stealing medicines to sell overseas.84 Even when doctors are prosecuted, courts may be disposed to treat them leniently. For example, in Hungary in 2017, an oncologist who failed to inform patients that they were going to die soon in order to continue to take informal payments from them was convicted, but, on appeal, his sentence was reduced on the basis that he was an important specialist whose expertise was needed.85 The use of DPAs in England and Wales and of CJIPs in France86 provides an illustration of how the ethical/pragmatic nature of prosecutorial decisions may be contested. On one hand, DPAs may be regarded as a practical means of avoiding the difficulties involved in proving corporate criminal liability, securing the cooperation of errant companies, and compelling them to ‘behave ethically, lawfully and transparently’.87 However, they may also be regarded as a means by which rich corporations can buy themselves out of prosecutions, and the Executive Director of Transparency International has suggested that they are ‘a soft option for companies that should be prosecuted for serious crimes’.88

4.2  Operational Issues The effectiveness of the operation of healthcare systems and the organisations involved in anti-­fraud and abuse activities plays a key role in the identification, policing, and prevention of fraud and abuse. Savedoff and Hussman have suggested that three factors are likely to impede transparency and accountability and make a health system vulnerable to corrupt practices: uncertainty, asymmetry of information, and large numbers of actors within the system.89 Uncertainty as a factor will always be present within health systems. For example, health predictions as to when and how people will become ill are frequently inaccurate, opinion about what treatment may be effective may be divided, and patients may not be aware that they require treatment or what treatments are

82  E.g., in England and Wales, the Sentencing Council’s sentencing guidelines for sentencing fraud offences requires judges to take into account matters such as breach of a position of trust when assessing culpability and both financial and the detrimental effect upon the victim ‘whether financial or otherwise’, when assessing the level of harm: Sentencing Council, Fraud (2014), https://www. sentencingcouncil.org.uk/offences/crown-­court/item/fraud/ 83 Ecorys, Updated Study on Corruption in the Healthcare Sector, European Commission B-­1049, pp. 34–35 (re Greece). 84 Id., supra n. 83, 82 (re Poland). 85 Id., supra n. 83, 71. 86  Elliott and Krause, supra n. 58, fn. 24. 87  O Cooke and D Hyde, ‘DPAs: Still Dividing Opinion?’ (2018) 168 New L J June 7799. 88 Id. 89  Savedoff and Hussman, supra n. 9.

404   Tracey A. Elliott appropriate or necessary.90 Asymmetry of information will also commonly be a factor in transactions within health systems. In particular, doctors will usually know more about health conditions and appropriate treatments than will patients, and more about the health conditions and appropriate treatment of a particular patient than will insurers.91 This may make a system vulnerable to fraud and abuse because it may make it difficult to assess whether treatment has been necessary or appropriately billed. Health systems in modern European states are highly complex systems involving interactions between patients and providers (e.g., healthcare professionals, hospitals), payers (e.g., insurers or government departments), suppliers of medical equipment and medicines, and government and professional regulators.92 The involvement of so many parties, with responsibilities being delegated from one actor to another or being split between actors, increases the opportunities for advantage to be taken of weaknesses in the system and makes it more difficult to obtain and analyse information, maintain transparency, and identify when fraud occurs within the system and who is responsible.93 Whilst these factors cannot be removed from healthcare systems, operational measures play a key role in reducing the amount of fraud and abuse within a system. For example, it is useful to have clear and transparent charging and payment systems, reliable IT systems, and trained personnel so that digital data may be audited, strong internal control systems may be established, and there can be the nurturing of an anti-­fraud culture.94 Some European countries have enjoyed notable successes in tackling fraud. For example, the National Health Service Counter Fraud Authority (NHSCFA) in England,95 an intelligence-­led organisation which employs counter-­fraud specialists and data-­mining tools to identify fraudulent behaviour, achieved its first successful prosecution in March 2018, of Andrew Taylor, a locksmith, who had defrauded the Guy’s and St. Thomas’ NHS Foundation Trust of nearly £600,000 when purchasing security supplies.96 Since then, NHSCFA has successfully applied for more than 80 orders in Crown Courts to recover the proceeds of criminal activity against the NHS.97 In France in 2017, L’Assurance Maladie detected fraud to the value of €270 million, 10% more than 90  Id., p.5. Cf. KJ Arrow, ‘Uncertainty and the Welfare Economics of Medical Care’ (1963) Am Econ Rev 53. 91  Savedoff and Hussman, supra n. 9, 5–6; A Coffinhal and A Frankowski, ‘Wasting with Intention: Fraud, Abuse, Corruption and Other Integrity Violations in the Health Sector’, in OECD ed., Tackling Wasteful Spending on Health (2017), ch.7, 267, https://dx.doi.org/10.1787/9789264266414-­10-­en. 92  Savedoff and Hussman, supra n. 9, 5. 93  Id., 6–7; Coffinhal and Frankowski, supra n. 91. 94  Boertjens et al., supra, n. 24, 17. 95  The NHSCFA is a health authority responsible for identifying, investigating, and preventing fraud within the NHS. It replaced the former counter-­fraud organisation, NHS Protect, from November 2017: NHSCFA, NEWS RELEASE: NHS Counter Fraud Authority Launched (2017) https://cfa.nhs.uk/ about-­nhscfa/latest-­news/nhscfa-­launched-­november-­2017. The equivalent body in Wales is the NHS Counter Fraud Service Wales. 96  NHSCFA, ‘Guy's Hospital Locksmith Jailed for Defrauding NHS of Nearly £600,000’ (2018) March 26, https://cfa.nhs.uk/about-­nhscfa/latest-­news/guys-­hospital-­locksmith-­ convicted#:~:text=The%20former%20Locksmith%20employed%20by,Health%20Authority%20in%20 November%202017 97 NHSCFA, Annual Report and Accounts 2018–2019 (2020), https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/859362/nhscfa-­annual-­report-­and-­ accounts-­2018-­2019.pdf

Healthcare Fraud and Abuse in Europe   405 the previous year.98 The Zavod za zdravstveno zavarovanje Slovenije (ZZZS) in Slovenia is regarded as having incorporated some of the best counter-­fraud practice in the modernisation of its healthcare system, introducing a systemic approach with improved IT infrastructure, including microchipped patient healthcare cards, which identify patient eligibility, who is paying, and for what.99 However, there is still much work to be done. In 2004, the European Healthcare Fraud and Corruption Declaration called for the ‘development of a European common stand­ ard of risk measurement, with annual statistically valid follow up exercises to measure progress in reducing losses to fraud and corruption throughout the EU’,100 but to date this has not been followed up with EU or domestic legislation,101 and all European countries still need to focus on preventing fraud (rather than merely identifying and policing it) by removing operational weaknesses that provide opportunities for fraudsters.102 Reports from states indicate that insufficient funding plays a major role in inhibiting counter-­fraud activities. Whilst Eastern European states have had particularly acute problems because of the underfunding of their healthcare systems,103 insufficient funding has been a problem in other European countries (e.g., in the United Kingdom),104 and underinvestment in the human resources needed for counter-­fraud measures to be effective has been reported in a number of states. In particular, a shortage of suitably trained investigators (such as medically trained inspectors, accredited counter-­fraud specialists, or specialists in forensic digital data analysis) has been identified as being a systemic weakness in reports on the United Kingdom, Belgium, France, Poland, Slovenia, and Lithuania.105 Finally, the successful litigation of both civil and criminal fraud cases may be compromised by corruption and a lack of independence in the justice system. Whilst research by Transparency International suggests that judges in the EU and Western Europe are

98  Espaceinfirmier.fr, ‘270 millions d'euros de fraude à l'Assurance maladie en 2017’ (2019) January 10, https://www.espaceinfirmier.fr/breves/270-­millions-­d-­euros-­de-­fraude-­a-­l-­assurance-­maladie-­ en-­2017.html 99 Eurosmart, Healthcare Fraud (2012), http://www.securitydocumentworld.com/creo_files/upload/ client_files/eurosmart-­healtcare-­fraud1.pdf, pp. 18, 30-­31; Kramberger, ‘Slovenia’, supra n. 55, 251–252. 100  European Healthcare Fraud and Corruption Conference (EHFCC), Countering Healthcare Fraud and Corruption in Europe: The European Healthcare Fraud and Corruption Declaration (2004), http:// www.ehfcn.org/wp-­content/uploads/2016/10/Declaration-­Final.pdf 101  J Gee and M Button, The Financial Cost of Healthcare Fraud 2019: The Latest Data from Around the World (2019), http://www.crowe.ie/wp-­content/uploads/2019/08/The-­Financial-­Cost-­of-­ Fraud-­2019.pdf, p. 13. 102  B Jones and A Jing, ‘Prevention Is Better Than Cure’ (2011) 89(12) Bull WHO 853. Cf. NHSCFA, NHS Fraud Prevention, https://cfa.nhs.uk/fraud-­prevention. 103  See Section 6. 104  Brooks et al., supra n. 63, 151; National Audit Office (NAO), NHS Financial Sustainability, HC 1867, Session 2017–2019 (2019); The King’s Fund, The NHS in a Nutshell (2020), https://www.kingsfund. org.uk/projects/nhs-­in-­a-­nutshell 105  Brooks et al., supra n. 63; Vincke, supra n. 60, 187; DACCRF, ‘France’, supra n. 56, 172; Ministry of Health of Poland, supra n. 73, 224; Kramberger, ‘Slovenia’, supra n. 55, 247–248, 252; JV Šveikauskas and V Srogė, ‘Lithuania’, in Mikkers, Sauter, Vincke, and Boertjens, supra n. 24, 305.

406   Tracey A. Elliott much less likely to accept bribes than their counterparts in other parts of the world,106 there have been recent well-­publicised arrests of judges in Lithuania107 and Albania108 for corrupt practices, and EU reports have raised particular concerns about the in­de­pend­ ence and probity of the judiciary in the Czech Republic, Greece, Hungary, Romania, and Bulgaria.109

5  Typologies of Healthcare Fraud Whilst every sector of the healthcare system of all European countries is affected by fraud in some way, opportunistic fraudsters inevitably seek to exploit weaknesses within the system and public concerns about disease and illness. A stark example of this may be found in the manner in which, during the Covid-­19 pandemic in the spring of 2020, internet websites swiftly emerged to take advantage of a shortage of personal protective equipment (PPE) and public concerns about the disease by selling counterfeit face masks and gowns, sanitisers, medicines, and test kits. Many of these items displayed a fake CE marking, a mandatory declaration that goods comply with EU health and safety standards.110 An investigation by the European Anti-­Fraud Office (OLAF) found that more than 340 companies were involved in the unlawful trade in counterfeit products linked to the Covid-­19 health emergency.111 The sale of counterfeit medicines via online pharmacies has proliferated in recent years, with it being estimated that Europeans spend more than US$14 billion a year on

106  Transparency International, Global Corruption Report 2007: Corruption in Judicial Systems (Cambridge University Press 2007), 11, table 1. Cf. European Commission, Eurobarometer: Corruption Report (2012), https://ec.europa.eu/commfrontoffice/publicopinion/archives/ebs/ebs_374_en.pdf 107  G Henry, ‘Top Lithuanian Judges Arrested in Anti-­Corruption Crackdown’ (2019) Politico, 20 February, https://www.politico.eu/article/top-­lithuanian-­judges-­arrested-­in-­anti-­corruption-­crackdown/ 108  L Semini, ‘20 Former Albanian Judges, Prosecutors Accused of Corruption’ (2020) ABC News, 8 May, https://abcnews.go.com/International/wireStory/20-­albanian-­judges-­prosecutors-­accused-­ corruption-­70587105. Regarding concerns about lack of independence and systemic corruption in the Judiciary in Albania, see European Commission, Albania 2015 Report (2015) https://ec.europa.eu/ neighbourhood-­enlargement/sites/near/files/pdf/key_documents/2015/20151110_report_albania.pdf 109  See, e.g., European Commission, supra n. 3, 219, 245, 250; A Mungiu-­Pippidi, The Good, the Bad and the Ugly: Controlling Corruption in the European Union (2013) ERCAS Working Paper No. 35, 45–46. In 2018, the Consultative Council of European Judges (CCJE) adopted an opinion on ensuring judicial independence and fighting corruption: Preventing Corruption Among Judges (2018) CCJE Opinion No. 21, https://rm.coe.int/ccje-­2018-­3e-­avis-­21-­ccje-­2018-­prevent-­corruption-­amongst-­judges/16808fd8dd 110  OLAF, ‘Inquiry into Fake COVID-­19 Products Progresses’, Press Release No.16/2020 (2020) 13 May; European Safety Federation, ‘Covid 19: Suspicious Certificates for PPE’ (2020) 9 July, https://www. eu-­esf.org/covid-­19/4513-­covid-­19-­suspicious-­certificates-­for-­ppe; Interpol, ‘Global Operation Sees a Rise in Fake Medical Products Related to COVID-­19’ (2020) 19 March, https://www.interpol.int/News-­ and-­Events/News/2020/Global-­operation-­sees-­a-­rise-­in-­fake-­medical-­products-­related-­to-­COVID-­19 111 OLAF, supra n. 110.

Healthcare Fraud and Abuse in Europe   407 illicitly sourced (and commonly fake) drugs.112 Operation Pangea, an international effort coordinated by Interpol since 2008 to disrupt such sales, has made more than 3,000 arrests and shut down 82,000 websites,113 but this type of fraud continues to pose particularly difficult challenges for counter-­fraud authorities, with new illegal websites springing up to replace those which have been closed down, and organised criminal groups (frequently from outside European jurisdictions) increasingly taking precautions, such as sending products in small, anonymous parcels that are less likely to be detected and stopped.114 Although fraud occurs in relation to all interactions within healthcare systems, with concerns being raised about procurement fraud115 and fraudulent entitlement claims by patients,116 much concern has been focused on fraud in the charging for medical services and products. ‘Upcoding’ involves the charging for a service which is more expensive than that actually provided to the patient and is widespread in European healthcare systems, with those that separate the provision of care and the financing of that care being particularly vulnerable to the manipulation of charging codes to increase renumeration.117 For example, in Germany between 2003 and 2012, hospitals upcoded at least 12,000 premature infants, with a resulting gain in reimbursement of approximately €100 million,118 and, in 2014, a major investigation by the Dutch Care Authority (NZa) found that a single hospital had issued incorrect invoices for €26.4 million of medical 112  WHO, ‘Growing Threat from Counterfeit Medicines’ (2010) 88 Bull WHO 241. Analysis of the results of seizures made during Operation Pangea indicated that at least 11% of medical products sole online are counterfeit: Interpol, ‘Operation Pangea: Shining a Light on Pharmaceutical Crime’ (2019) 21 November, https://www.interpol.int/en/News-­and-­Events/News/2019/Operation-­Pangea-­shining-­a-­ light-­on-­pharmaceutical-­crime 113 Interpol, supra n. 112. 114 OLAF, supra n. 110; Interpol, supra n. 112. Cf. A Jack, ‘Can Anyone Stop the Illegal Sale of Medicines Online?’ (2016) https://doi.org/10.1136/bmj.i1317; KS Lee, SM, Yee, STR Zaidi, RP Patel, Q Yang, YM Al-­Worafi, and LC Ming, ‘Combating Sale of Counterfeit and Falsified Medicines Online: A Losing Battle’ (2017) 8 Front Pharmacol 268, O’Hagan and Garlington, supra n. 80. 115  European Commission, Updated Study on Corruption in the Healthcare Sector: Final Report (2017) https://ec.europa.eu/home-­affairs/sites/homeaffairs/files/20170928_study_on_healthcare_ corruption_en.pdf; NHSCFA, Pre-­Contract Procurement Fraud and Corruption: Guide for Prevention and Detection (2018), https://nhsprocurement.org.uk/wp-­content/uploads/2018/08/NHSCFA-­Pre-­ contract-­procurement-­fraud-­guidance-­v1.0-­July-­2018.pdf; OECD, Health and Public Procurement, https://www.oecd.org/gov/public-­procurement/health/. It has been estimated that, in the UK, £266 million per year are lost to procurement and commissioning fraud: NHS England, Tackling Fraud, Bribery and Corruption: Economic Crime Strategy 2018–2021 (2019), https://www.england.nhs.uk/ wp-­content/uploads/2013/06/tackling-­fraud-­bribery-­and-­corruption-­economic-­crime-­ strategy-­2018-­2021.pdf, p. 7. It has been suggested that the Covid-­19 pandemic is likely to have led to an increase in procurement fraud as national procurement laws have been relaxed in order to obtain supplies of PPE and ventilators quickly: Transparency International, ‘First Response: Procure Medical Supplies at Any Cost (and Risk)’ (2020) 29 April, https://www.transparency.org/en/news/ first-­response-­procure-­medical-­supplies-­at-­any-­cost-­and-­risk 116  NHS England, supra n. 115, 11, 22–24. 117  Transparency International, supra n. 76, 14. 118  H Jürges and J Köberlein, ‘What Explains DRG Upcoding in Neonatology? The Role of Incentive and Infant Health’ (2015) 45 J Health Econ13.

408   Tracey A. Elliott c­ are.119 Healthcare professionals may also charge for unnecessary services120 or services that are not provided. An illustration of the latter type of fraud may be found in the Greek ‘healing dressing scandal’, where an investigation revealed that photographs of ulcers, which had to be uploaded to the E-­prescription service in order to obtain reimbursement for dressings, had been used multiple times, in some cases by different physicians.121

6  The Problem of Informal Payments Informal payments provide an important and illustrative example of the problem of fraud and abuse in Europe. These payments, also known as ‘unofficial’, ‘under-­the-­table’, ‘under-­the-­counter’, or ‘envelope’ payments, are generally defined as direct unofficial payments made by patients for services that are meant to be covered by the existing healthcare system.122 They may potentially occur in any healthcare system, even in high-­income countries such as the United Kingdom, France, and Germany,123 but studies indicate that, whilst in most European countries they are uncommon,124 they are particularly prevalent in the former Eastern Bloc125 states of Lithuania, Latvia, Hungary, Poland, the Slovak Republic, Bulgaria, Romania, and Albania. Although the reasons for the prevalence of these payments are contested,126 they may be seen as arising out of a combination of economic, social, and cultural circumstances. Following the collapse of the Eastern Bloc, formerly socialist Central and Eastern European states reformed their healthcare systems, moving from the centralised, state-­controlled Soviet Semashko 119  Boertjens, ‘The Netherlands’, supra n. 56, 201. 120  See, e.g., Boertjens et al., supra n. 24, 11–12. 121 Ecorys, supra n. 83, 40. 122  M Lewis, ‘Informal Payments and The Financing of Health Care in Developing and Transition Countries’ (2007) 26 Health Affairs 984, 985; S Tomini, Informal Payments for Healthcare Services in Albania (Boekenplan 2011), 2; S Allin, K Davaki, and E Mossialos, ‘Paying for “Free” Health Care: The Conundrum of Informal Payments in Post-­Communist Europe’, in Transparency International ed., supra n 9. Cf. P Gaál, P Belli, M McKee, and M Szócska, ‘Informal Payments for Health Care: Definitions, Distinctions, and Dilemmas’ (2006) 31(2) J Health Politics, Policy Law 251. 123  T Vian, ‘Cash Registers Inject Transparency—and revenue—into Kenya’s Coast Provincial General Hospital’, in Transparency International ed., Global Corruption Report 2006 (2006), 53, 54. 124  See, e.g., European Commission, Special Eurobarometer 470 Report: Corruption (2017), http://ec. europa.eu/commfrontoffice/publicopinion, pp. 84–85. Overall, only a small percentage of respondents to this survey (4%) indicated that they had to give an extra payment or valuable gift or make a donation to the hospital. 125  The “Eastern Bloc” is a term used to describe the former Warsaw Pact countries (the Soviet Union, East Germany, Poland, Hungary, Czechoslovakia, Yugoslavia, Bulgaria, Romania, and Albania). See North Atlantic Treaty Organisation (NATO) website: https://www.nato.int/cps/en/natohq/ declassified_138294.htm 126  R Gaál and M McKee, ‘Fee-­for-­Service or Donation? Hungarian Perspectives on Informal Payment for Health Care’ (2005) 60 Soc Sci Med 1445, 1446–1448. Cf. Lewis, supra n. 122; T Stepurko, M Pavlova, I Gryga, L Murauskiene, and W Groot, ‘Informal Payments for Health Care Services: The Case of Lithuania, Poland and Ukraine’ (2015) 6 J Eurasian Studies 46, 48–52.

Healthcare Fraud and Abuse in Europe   409 system127 to more decentralised models. During the transition period, difficulties faced by these healthcare systems included a decline in public funding caused by the rapid transition from centrally planned to market economies,128 inherited inefficient systems, and an oversupply of doctors who were not well-­paid and potentially likely to seek to boost their income by seeking payments from patients.129 In addition, in the health sector, there was a legacy of corruption130 and blat,131 with tolerance of ‘gift-­giving’ and informal practices and inadequate regulatory control to prevent or curb such practices.132 Problems with informal payments are not, however, confined to former Eastern Bloc states. They are also widespread in Greece, where they have been described as an ‘ingrained social institution’.133 In common with the former Eastern Bloc countries, Greece has suffered an economic crisis, underfunding of its health system, and weak control mechanisms to check abuses.134 The economic crisis has had a severe impact on Greece since 2010, with the implementation of austerity measures leading to substantial reductions in the public health sector and health expenditure falling by a fifth between 127  Named after Nikolai Semashko, the first Soviet Peoples’ Commissar of Public Health (1918–1930). For discussion regarding the development of the system, see, e.g., I Sheiman, S Shishkin, and V Shevsky, ‘The Evolving Semashko Model of Primary Health Care: The Case of the Russian Federation’ (2018) 11 Risk Mgmt Health Care Policy 209; World Health Organisation (WHO), ‘Rocky Road from the Semashko to a New Health Model’ (2013) 91 Bull WHO 320, http://dx.doi.org/10.2471/ BLT.13.030513. Cf. R Mandel and C Humphrey eds., Markets and Moralities: Ethnographies of Postsocialism (Bloomsbury 2002). 128  Commonly termed ‘shock therapy’: Tomini, supra note 122, 2; AS Preker and OO Adeyi, ‘Health Care’, in N Barr ed., Labor Markets and Social Policy in Central and Eastern Europe: The Accession and Beyond (World Bank 2005), 176–177; G Chelleraj, OAdeyi, AS Preker, and E Goldstein, Trends in Health Status, Service and Finance, Statistical Analysis, Vol. II, World Bank Technical Paper No. 348 (World Bank 1996). The reduction in funding was more marked in the Baltic States, Romania, and Bulgaria than in Hungary, Poland, and the Czech and Slovak Republics. 129  M McKee, ‘Health Services in Central and Eastern Europe: Past Problems and Future Prospects’ (1991) 45 J Epidemiol Comm Health 260. 130  R Rose, ‘Corruption Is Bad for Your Health: Findings from Central and Eastern Europe’, in Transparency International ed., Global Corruption Report: Special Focus: Corruption and Health (2006), https://www.transparency.org/research/gcr/gcr_health, p. 39. 131  Blat is described by Ledeneva as ‘the use of personal networks and informal contacts to obtain goods and services in short supply and to find a way around formal procedures’. As a term, however, it is not easy to define. Although the pre-­revolutionary definition of Blat refers to criminal activity, and similarities may be found between Blat and criminal conduct such as bribery and corruption, distinctions may be made as Blat is embedded in personal networks and may be seen as a compensatory mechanism allowing individuals to maintain reasonable living conditions in the face of rigid state control: AV Ledeneva, Russia’s Economy of Favours: Blat, Networking and Informal Exchange (Cambridge University Press 1998), 2, 12, 33, 39–47. According to Ledeneva, ‘The word is virtually impossible to translate directly into English’, 2. C.f. A Ledeneva, ‘Open Secrets and Knowing Smiles’ (2011) 25 East Eur Pol Soc 720. 132  R Thompson and S Witter, ‘Informal Payments in Transitional Economies: Implications for Health Sector Reform’ (2000) 15 Intl J Health Planning 169, 171–180; Stepurko et al., supra, n. 126, 52; Allin et al., supra n. 122, 63. 133  L Liaropoulos, O Siskou, D Kaitelidou, M Theodorou, and T Katostaras, ‘Informal Payments in Public Hospitals in Greece’ (2008) 87 Health Policy 72. 134 Lewis, supra n. 122, 993; C Economou, D Kaitelidou, M Karanikolos, and A Maresso, Greece: Health Systems Review 2017 (2017) 19(5) Health Systems in Transition 1, xvii.

410   Tracey A. Elliott 2010 and 2015. Healthcare professionals are also relatively poorly paid. Even before the crisis, the salaries of healthcare professionals were amongst the lowest in the European Union, but in 2010, salaries were cut by 20% in the public sector, with planned performance-­based productivity bonuses being cancelled. Public sector underfunding has led to substantial reliance on private financing, with relatively high levels of out-­of-­ pocket payments, more than a quarter of these being informal payments. In addition, there are difficulties with recruiting and retaining doctors in rural locations, which may lead to supplier-­induced demand in these areas, encouraging informal payments. There also is a broad undersupply of general practitioners and nurses, which has led to problems with understaffing in hospitals and to long waiting lists for some services,135 which may tempt patients to pay informal payments to obtain more speedy treatment. An examination of the prevalence of informal payments in some European countries provides an interesting case study, highlighting the definitional, organisational, and ethical difficulties involved in combatting what is frequently regarded as a corrupt and legally and ethically problematic practice. First, whilst attempts have been made to define ‘informal payments’ as a term,136 there is no generally accepted definition.137 Part of this definitional difficulty arises because the barrier between ‘gifts’ and informal payments is obscure138 and because it may not be easy to distinguish between official payments and ‘under the counter’ payments, either because the latter may become formalised,139 because what appear to be informal payments may in reality be authorised charges collected in an informal (and possibly dishonest) manner,140 or because patients (or those interpreting patient surveys) are unclear as to the status of the payment made.141 It also arises because of the wide variety of forms which informal payments may take, from small gifts and monetary gratuities to the payment of large sums.142 Whilst informal payments may be regarded as ‘acts of corruption’143 or unlawful 135  Economou et al., supra, n. 134, xvi–xvii, 58, 62, 67, 69, 75–79. In 2015, out-­of-­pocket payments amounted to 35% of health expenditure. 136  T Ensor and L Savelyeva, ‘Informal Payments for Health Care in the Former Soviet Union: Some Evidence from Kazakhstan’ (1998) 13 Health Policy and Planning 41. 137  P Gaál, P Belli, M McKee, and M Szócska, ‘Informal Payments for Health Care: Definitions, Distinctions, and Dilemmas’ (2006) 31(2) J Health Politics Policy Law 251, 252; D Balabanova and M McKee, ‘Understanding Informal Payments for Health Care: The Example of Bulgaria’ (2002) 62 Health Policy 243. 138  T Ensor, ‘Informal Payments for Health Care in Transition Economies’ (2004) 58 Soc Sci Med 237, 238; Stepurko et al., supra n. 126, 48. The literal translation of the Hungarian term for an informal payment, hálapénz, is ‘gratitude payment’: P Gaál, ‘Gift, Fee or Bribe? Informal Payments in Hungary’, in Transparency International ed., supra n. 9, 71, 74, fn. 2. 139  Balabanova and McKee, supra n. 137, 246. 140  T Vian, K Brybosk, Z Sinoimeri, and R Hall, ‘Informal Payments in Government Health Facilities in Albania: Results of a Qualitative Study’ (2006) 62 Soc Sci Med 877, 885. 141  See, e.g., Balabanova and McKee, supra n. 137, 252, where the authors refer to the example of a Bulgarian patient who reported paid 100,000 Levs (approximately $65) for an abortion. This had been interpreted as being a user fee, but the authors suggest that the context in which it was made indicates that it was an under-­the-­counter payment to the physician. 142  Gaál et al., supra n. 137, 252; Ensor, supra n. 138, 240. 143  T Vian, ‘Review of Corruption in the Health Sector: Theory, Methods and Interventions (2008) 23 Health Policy and Planning 83; TK Mackey and BA Liang, ‘Combating Healthcare Corruption and

Healthcare Fraud and Abuse in Europe   411 ‘institutionalized bribery’,144 the actual or perceived normative status of these payments may vary,145 and, in societies in which such payments are widespread and accepted both by patients and healthcare providers, people may see the labelling of informal payment as corrupt as being unduly harsh and inappropriate.146 Even if patients ostensibly regard informal payments as being ‘gifts’ or ‘gratuities,’ they may nevertheless be regarded as abusive if the inequality of power between doctor and patient with regard to the provision of medical treatment is exploited to require or place pressure on patients to pay for treatment which should be provided free of charge.147 The payments may also be regarded as being fraudulent if money or goods are obtained dishonestly for payment that is not lawfully due. Second, the tackling of informal payments raises considerable organisational problems for countries in which they are prevalent. Since informal payments are usually prevalent in states in which the healthcare system is underfunded and healthcare professionals poorly paid, they may be regarded as a ‘coping mechanism’ to compensate for inadequate public sector salaries.148 A government seeking to crack down on informal payments in such circumstances faces the dilemma of finding funds to increase health care funding and staff pay to acceptable levels, or potentially risk losing public health workers to the private sector or to wealthier countries where public sector workers are more generously remunerated. In the circumstances, governmental tolerance of informal payments is understandable. Given this, it is unsurprising that the governments of those states in which they are common tend to treat them in an equivocal manner. For example, although informal payments were strictly banned in Bulgaria until 1989, this was not enforced, and subsequent government inaction has led them to acquire semi-­legal status.149 In Hungary, although informal payments are unlawful in principle, in practice they are permitted, Fraud with Improved Global Health Governance’ (2012) 12 BMC Intl Health Human Rights 23. Both WHO and Transparency International regard informal payments as corruption: TK Mackey, T Vian, and J Kohler, ‘The Sustainable Development Goals as a Framework to Combat Health-­Sector Corruption’(2018) 96 Bull WHO 634, http://dx.doi.org/10.2471/BLT.18.209502; Transparency International, supra n. 9. 144 Lewis, supra note 122, 985. 145  N Cohen, ‘Informal Payments for Health Care: The Phenomenon and Its Context’ (2012) 7 Health Economics Policy Law 285, 287–288. 146  T Stepurko, M Pavlova, I Gryga, and W Groot, ‘Informal Payments for Health Care Services: Corruption or Gratitude? A Study on Public Attitudes, Perceptions and Opinions in Six Central and Eastern European Countries’ (2012) 46 Communist and Post-­Communist Studies 419; T Vian, FG Feeley, S Domente, A, Neguta, A Matei, and J Habicht, ‘Barriers to Universal Health Coverage in the Republic of Moldova: A Policy Analysis of Formal and Informal Out-­of-­Pocket Payments’ (2015) 15 BMC Health Services Res 319, 326–327; M Schaaf and SM Topp, ‘A Critical Interpretive Synthesis of Informal Payments in Maternal Health Care’ (2019) 34 Health Policy and Planning 216, 223. 147  Cf. A Szende and A Culyer, ‘The Inequity of Informal Payment for Health Care: The Case of Hungary’ (2006) 75 Health Policy 262. 148  T Vian, ‘Corruption in Hospital Administration’, in Transparency International ed., supra n. 9, 49, 54; Gaál et al., supra n. 137, 256–257. 149  Balabanova and McKee, supra n. 137, 244.

412   Tracey A. Elliott must be reported in tax returns, and are taxed by the government, which may be seen as tacit acceptance of the practice. Condonation of the practice may in turn lead to problems with enforcing legal prohibitions. For example, the Hungarian Criminal Code precludes prosecution in cases where a person ‘commits an act on the mistaken assumption that it is not harmful to society and who has reasonable grounds for this presumption’,150 and it has been suggested that this may apply to some doctors accepting informal payments.151 Third, whilst informal payments may be regarded as being ethically problematic in that they reinforce health inequalities, may force lower-­income patients into health poverty or to delay or avoid medical care, may undermine attempts by government to improve accountability,152 may contribute to the growth of corruption in public services,153 and may even lead to a public loss of faith in the healthcare system,154 positive attitudes to them may be found in the countries in which they occur. Patients may see them as being a means of showing appreciation of and respect to those who treat them and their families. In addition, the payments may foster long-­lasting, continuous relationships between patient and physician, improve staff morale, and keep health workers from leaving the public health system or moving to work in wealthier European states.155

7 Conclusion An examination of the law and regulation in relation to fraud and abuse in Europe confirms that, in common with the United States and other countries, fraud is a universal problem. Fraudsters are always looking to explore and exploit weaknesses in the system for personal financial gain, and the sheer amount of money poured into healthcare systems, even in less affluent countries, makes them an attractive target. Healthcare fraud has also become increasing internationalised, with organised lawbreakers being ever ready to use change to their own ends. For example, the ubiquity of the internet has provided fraudsters (both within and outside Europe) with new worldwide opportunities to make money by selling (frequently counterfeit) medical equipment and prescription-­ only drugs directly to the public. However, although fraud is universal, it is also historically, politically, and socially situated. Although it may be tempting to regard Europe as an entity, the scope and magnitude of the problem varies widely, even between EU Member States. Whilst some states, particularly the more affluent Western European and Scandinavian countries, 150  Hungarian Criminal Code (2012), Section 20(2). An English translation may be accessed at https://www.legislationline.org/download/id/5619/file/HUngary_Criminal_Code_of_2012_en.pdf 151  See, e.g., G Adam, ‘Gratuity for Doctors and Medical Ethics’ (1989) 14 J Med Philosophy 315, who suggests that this would apply to gratuities; Gaál et al., supra n. 137, 267. 152  Vian et al., supra n. 140, 887; Lewis, supra n. 122, 990; Cohen, supra n. 145. 153  Allin et al., supra n. 122, 67. 154  Vian et al., supra n. 140, 878; Gaál, supra n. 138, 88. 155  Vian et al., supra n. 140, 877–878; Balabanova and McKee, supra n. 137, 249, 259, 261.

Healthcare Fraud and Abuse in Europe   413 may be regarded as having strong measures in place to prevent, detect, and punish fraud and abuse, less affluent countries (particularly former Eastern Bloc and eastern Mediterranean states) suffer from what may be regarded as a double burden in the fight against fraud. First, for historical, political, and social reasons, their health systems may be regarded as being more vulnerable to fraud, corrupt practices, and abuse because of underfunding and a widespread toleration of informal payments and, second, because the organisational controls in place to combat unlawful activity are currently not equal to the task that they face.

Bibliography Alexa, J., Rečka, L., Votápková, van Ginneken, E., Spranger, A. and Wittenbecher F., ‘Czech Republic: Health System Review’ (2015) 17(1) Health Systems in Transition 1. Atanasova, E., Pavlova, M., Moutafova, E., Rechel, B. and Groot, W., ‘Informal payments for Health Services: The Experience of Bulgaria After 10 Years of Formal Co-Payments’ (2013) 24 European Journal of Public Health 733. Delcheva, E., Balabanova, D. and McKee, M,. ‘Under-the-Counter Payments for Health Care: Evidence from Bulgaria’ (1997) 42 Health Policy 89. Dimova, A., Rohova, M., Koeva, S., Atanasova, E., Koeva-Dimitrova, L., Kostadinova, T. and Spranger, A., ‘Bulgaria: Health System Review’ (2018) 20(4) Health Systems in Transition 1. Ensor, T., and Witter, S., ‘Health Economics in Low Income Countries: Adapting to the Reality of the Unofficial Economy’ (2001) 57 Health Policy 1. Federal Ministry of Labour and Social Affairs, Social Security at a Glance 2019 (2019), http:// www.bmas.de Federal Ministry Labour, Social Affairs, Health and Consumer Protection, The Austrian Health Care System Key Facts, Updated edition (2019), https://www.sozialministerium.at/ cms/siteEN/attachments/6/3/9/CH4129/CMS1417371984289/bmasgk_the-austrian_healthcare-system__keyfacts__web.pdf Gaál, P., Szigeti, S., Csere, M., Gaskins, M., and Panteli, D., ‘Hungary: Health System Review’ (2011) 13(5) Health Systems in Transition 1. Mitenbergs, U., Taube, M., Misins, J., Mikitis, E., Martinsons, A., Rurane, A., and Quentin, W., ‘Latvia: Health System Review’ (2012) 14(8) Health Systems in Transition 1. Murauskiene, L., Janoniene, R., Veniute, M., van Ginneken, E., and Karanikolos, M., ‘Lithuania: Health System Review’ (2013) 15(2) Health Systems in Transition 1. Smatana, M., Pažitný, P., Kandilaki, D., Laktišová, M., Sedláková, D., Palušková, M., van Ginneken, E., and Spranger, A., ‘Slovakia: Health System Review’ (2016) 18(6) Health Systems in Transition 1. Sowada, C., Sagan, A., Kowalska-Bobko, I., Badora-Musiał, K., Bochenek, T., Domagała, A., Dubas-Jakóbczyk, K., Kocot, E., Mrożek-Gąsiorowska, M., Sitko, S., Szetela, A., Szetela, P., Tambor, M., Więckowska, B., Zabdyr-Jamróz, M., and van Ginneken, E., ‘Poland: Health System Review’ (2019) 21(1) Health Systems in Transition 1. Tomini, S., Informal Payments for Healthcare Services in Albania (Maastricht: Boekenplan. 2011). Vlãdescu, C., Scîntee, S. G., Olsavszky, V., Hernández-Quevedo, C., and Sagan, A., ‘Romania: Health System Review’ (2016) 18(4) Health Systems in Transition 1.

T H E T R E AT M E N T R E L AT IONSH I P: C ON F I DE N T I A L I T Y, C ONSE N T, A N D C ON F L IC T S OF INTER EST

chapter 22

Pr i vacy a n d I n tegr it y of M edica l I n for m ation Sharona Hoffman and Jean Herveg

1 Introduction Medical privacy is cherished throughout the world. The value of privacy was ­recognized as early as the 5th century bce and is included in the Hippocratic Oath. Physicians reciting the oath have traditionally said: “whatsoever I shall see or hear in the course of my profession . . . if it be what should not be published abroad, I will never divulge, holding such things to be holy secrets.”1 Worldwide, the rule is that healthcare practitioners are bound by the duty not to disclose any information related to their patients, with exceptions applying only in some specific situations. This chapter explores contemporary regulation of medical privacy in the United States and Europe2 and its challenges. It discusses regulatory strengths and shortcomings and highlight gaps in the law. The authors also suggest further safeguards that policy-makers should implement in order to protect patients and data subjects. The need for privacy is a fundamental human necessity. Privacy relates to human beings’ ability to maintain their dignity and avoid disclosure of information that would be embarrassing, demeaning, or otherwise make others think less of them. Privacy is also associated with personal autonomy and informational self-determination. 1 Ian E. Thompson, The Nature of Confidentiality, 5 J. Med. Ethics 57, 57 (1979). 2  The notion of professional privilege in healthcare has never been harmonized in Europe. The notion is specific to each European state. Its material and personal scope may vary from state to state, as well as its legal effects. In general, breach of professional secrecy may lead to criminal prosecution of immediate contract termination for the practitioner concerned. Undue use or even possession of information protected by professional privilege may also be a criminal offence.

418   Sharona Hoffman and Jean Herveg Individuals seek privacy in order to control how others perceive them and to avoid abuses that might ensue if their sensitive data were easily available and could be used in harmful ways.3 At the same time, some degree of data sharing is essential to the appropriate treatment of patients as well as to the proper functioning of society in general and the healthcare system in particular. Thus, privacy cannot be limitless. At times, medical team members must share patient data in order to provide effective treatment. Health data about individuals are also essential to medical research and to many public health initiatives, such as responses to infectious disease outbreaks. Likewise, institutions may need to review patient data in order to assess and improve the quality of their services. Consequently, government officials must carefully weigh the benefits and risks of privacy versus information exchange in formulating regulatory policies.4 In the era of omnipresent social media, one might ask whether individuals still value privacy. Many people, especially the young, routinely post intimate details, including medical information, on Facebook and other platforms for large-scale public consumption. When surveyed, however, overwhelming majorities of respondents indicate that privacy is a priority value for them. Social media users trust in their ability to manage privacy settings and to control access to their data.5 Privacy, therefore, remains a vital matter in contemporary society. It is important to understand that three separate terms relate to protecting patients’ interests: privacy, confidentiality, and security. Privacy focuses on the questions of whether information can be acquired and used, by whom, and under what circumstances, and thus on the patient’s rights regarding the use of medical information. Confidentiality is the principle that healthcare providers generally must not disclose patient information to third parties without patient authorization. Thus, while privacy is a patient right, confidentiality is a professional obligation. Data security refers to mechanisms that prevent unauthorized individuals from accessing patient medical records. These can include passwords, encryption, and other technologies.6 In the 21st century, safeguarding privacy has become more challenging than ever before. Medical records no longer take the form of paper files that can be locked away in cabinets. Their replacement, electronic health records (EHRs), can be hacked, easily accessed by unauthorized people through workplace computers, and lost or stolen if they are stored on laptops or USB devices. Medical big data repositories are being 3  Bruce Schneier, Data and Goliath: The Hidden Battles to Collect Your Data and Control Your World 126–127 (W.W. Norton 2015). 4  Sharona Hoffman, Electronic Health Records and Medical Big Data: Law and Policy 116–124 (Cambridge University Press 2016). 5  Id. at 138; David J. Kaufman et al., Public Opinion About the Importance of Privacy in Biobank Research, 5 Am. J. Hum. Genet. 85 (2009), 645; Bernard Debatin et al., Facebook and Online Privacy: Attitudes, Behaviors, and Unintended Consequences, 15 J. Computer-Mediated Comm. 1 (2009) 86, 100. 6  Comm. on Health Research & the Privacy of Health Info.: The HIPAA Privacy Rule, IOM, Beyond the HIPAA Privacy Rule: Enhancing Privacy, Improving Health Through Research 76 (Sharyl J. Nass et al. eds., 2009).

Privacy and Integrity of Medical Information   419 developed by numerous government and private entities for purposes of research, public health, quality improvement, and more.7 These, too, can be hacked or accessed by unapproved personnel. Wearable devices such as Fitbits, Apple Watches, and attachable baby monitors also collect vast amounts of information and raise privacy concerns.8 Both the United States and the European Union (EU) have tackled the problem of privacy with major regulatory initiatives. In the United States, safeguards come in the form of the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule, which became effective in 2003.9 At the level of EU law, medical information confidentiality is ensured by the notion of data protection. EU data protection law aims to protect citizens’ rights over “personal data” in the context of information and communication technologies. The EU has enacted the General Data Protection Regulation (GDPR),10 recognizing new rights for data subjects, which became applicable on May 25, 2018. HIPAA and GDPR have much in common. Both provide meaningful privacy protections for data subjects and require implementation of sound data security measures. Both carve out reasonable exceptions, such as permitting data use without patient authorization for purposes of judicial proceedings, legal requirements, public health measures, and medical treatment. Both provide patients with access to their own medical records and require covered entities to notify authorities of privacy breaches. However, the United States and EU regulations diverge in significant ways. For example, the HIPAA Privacy Rule addresses only protected health information (PHI). By contrast, the GDPR covers all information concerning identified or identifiable persons and is not limited to healthcare. While HIPAA allows patients to place restrictions on use of their medical information, the GDPR sets up a legal framework in which personal data concerning health may be processed. HIPAA allows healthcare providers to disclose PHI without patient consent to third parties such as insurers and billers for purposes of payment and healthcare administrative functions, while the GDPR might require a patient’s consent (e.g., for transmission to the patient’s insurance company) and the mandatory disclosure of some information about data processing to the patient. Another important difference is that the GDPR applies to organizations beyond the borders of the EU so long as they process data pertaining to EU citizens and the processing activities relate to the offering of goods or services or the monitoring of behavior that takes place within the EU. Conversely, the HIPAA Privacy Rule’s reach outside of the United States is limited to business associates of US covered entities. The chapter begins with an analysis of privacy and confidentiality. Section 2 addresses both together because they are interrelated principles that are essentially two sides of the same coin. The first section describes and critiques the US HIPAA Privacy Rule, other 7  Sharona Hoffman & Andy Podgurski, The Use and Misuse of Biomedical Data: Is Bigger Really Better? 39 Am. J. L. Med. 497, 503–515 (2013). 8  Janice Phaik Lin Goh, Privacy, Security, and Wearable Technology, 8 Landslide (2015). 9  45 C.F.R. §§ 160.101–534 (2018). 10  Regulation (EU) 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), OJ L 119 4 May 2016 p. 1.

420   Sharona Hoffman and Jean Herveg federal privacy laws, and state confidentiality mandates. The relevant statutory and regulatory provisions establish physicians’ duty to maintain patient confidentiality but also outline circumstances in which that duty must be breached and information must be revealed to third parties. The section then addresses how European law (both from the Council of Europe and the EU) approaches medical information privacy and confidentiality and how the main legal rules (on data protection in general) interface with statelevel laws about medical professional privilege and secrecy. Section 3 assesses the ways in which health data may lawfully be gathered to ensure data security. In the United States, this is covered by health data security regulations. In the EU context, health data security rules are part of the general rules governing “data processing.” The analysis then shifts in Section 4 to the question of data de-identification, another tool that is used to protect privacy. Section 5 examines remedies for unauthorized privacy breaches in the United States and EU and the means by which the EU’s GDPR is enforced. Section 6 focuses on anti-discrimination laws that prohibit certain uses of health information. In order to avoid privacy harms, regulators must strive both to control access to health data and to manage the uses to which they are put once obtained. Thus, anti-discrimination statutes are a useful adjunct to privacy initiatives. The chapter concludes with a discussion of contemporary challenges to regulating privacy, especially in the United States. In particular, it considers the following: (1) growing and ever-evolving data security threats, (2) the proliferation of health data stemming from sources such as social media, and (3) the emergence of predictive health data.

2  Privacy and Confidentiality 2.1  American Law The word “privacy” does not appear in the US Constitution. However, a variety of celebrated Supreme Court decisions have established that the Constitution encompasses privacy rights.11 Furthermore, the American Medical Association’s Principles of Medical Ethics states that “A physician shall respect the rights of patients, colleagues, and other health professionals, and shall safeguard patient confidences and privacy within the constraints of the law.”12 This principle is reflected in numerous federal and state laws. Nevertheless, American medical privacy laws and regulations have significant gaps and limitations. Moreover, the Supreme Court has declined to determine whether the 11  See Griswold v. Connecticut, 381 U.S. 479, 483, 485–486 (1965) (asserting that “the First Amendment has a penumbra where privacy is protected from governmental intrusion”); Roe v. Wade, 410 U.S. 113, 153 (1973) (determining that the right to privacy encompasses a woman’s decision to terminate her pregnancy, subject to appropriate state regulation). 12  American Medical Association, AMA Code of Medical Ethics: AMA Principles of Medical Ethics (revised 2001), https://www.ama-assn.org/sites/default/files/media-browser/principles-of-medicalethics.pdf.

Privacy and Integrity of Medical Information   421 Constitution establishes a right to informational privacy.13 Many state laws preceded the enactment of the federal HIPAA regulations, but the latter superseded any state statutes that provided weaker protections.14 This section examines the substance and scope of US federal and state privacy laws.

2.1.1  The HIPAA Privacy Rule The HIPAA regulations were promulgated pursuant to the Health Insurance Portability and Accountability Act of 1996 and were amended in accordance with the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009.15 The HIPAA Privacy Rule governs the disclosure of electronic and hard-copy medical information and allows patients to control their medical records to a degree. The Privacy Rule establishes that, with some exceptions, hospitals, physicians, health insurers, and other covered entities must obtain patients’ permission before disclosing their medical data to others.16 This is a key confidentiality safeguard for American patients, though it is far from absolute and its limits are discussed in the next sub-section. Under the HIPAA Privacy Rule, covered entities must also give patients notice of their privacy practices17 and must allow patients to view their health records and request that they be modified or used restrictively.18 In addition, covered entities that experience privacy breaches of unsecured data, such as incidents of hacking, must notify affected individuals and the Department of Health and Human Services (DHHS), and, in the case of large breaches, must notify the media as well.19 2.1.1.1  The Boundaries of the HIPAA Privacy Rule While the HIPAA regulations are the most detailed and comprehensive medical privacy regulations in the United States, many critics argue that they fall far short of adequately protecting American patients. The federal regulations’ limitations are rooted primarily in three factors: (1) the definition of “covered entity,” (2) the numerous Privacy Rule exceptions, and (3) the definition of “protected health information.” The Definition of “Covered Entities.” The HIPAA Privacy Rule does not govern many entities and individuals who handle private health information. The regulations define “covered entities” as including health plans, healthcare clearinghouses, and healthcare 13  National Aeronatutics and Space Administration v. Nelson, 131 S.Ct. 746, 756–757 (2011). 14  U.S. Department of Health and Human Services, Does the HIPAA Privacy Rule Preempt State Laws?, https://www.hhs.gov/hipaa/for-professionals/faq/399/does-hipaa-preempt-state-laws/index. html (last reviewed July 26, 2013). 15  42 U.S.C. §§ 1320d–1320d-9 (2010); Health Information Technology for Economic and Clinical Health (HITECH) Act, Title XIII of Division A and Title IV of Division B of the American Recovery and Reinvestment Act of 2009 (ARRA), Pub. L. No. 111–115 (Feb. 17, 2009). 16  45 C.F.R. §§ 164.508–.510 (2018). 17  45 C.F.R. § 164.520(a) (2018). 18  45 C.F.R. §§ 164.520, 164.522 (2018). 19  45 C.F.R. §§ 164.400–.408 (2018). The media must be notified if a breach involves “more than 500 residents of a State or jurisdiction.” Id. at § 164.408. Unsecured protected health information means information “that is not rendered unusable, unreadable, or indecipherable to unauthorized persons through the use of a [specified] technology or methodology,” such as encryption.

422   Sharona Hoffman and Jean Herveg providers who transmit health information electronically for purposes of HIPAArelevant transactions and their business associates.20 Thus, they do not cover employers; marketers; website operators; insurers issuing life, disability, or long-term care policies; and numerous others who may handle large volumes of health information.21 There is no doubt that health information is routinely processed and stored by those outside the healthcare industry. Employers, for example, frequently subject applicants and employees to medical inquiries and examinations.22 They also can obtain records for purposes of workers’ compensation claims, wellness programs, reasonable accommodation requests by individuals with disabilities, or Family Medical Leave Act (FMLA) requests.23 Patients themselves often willingly disclose data to non-covered entities, such as Fitbit vendors or website operators. For example, WebMD’s Symptom Checker asks users to provide details about their age, sex, zip code, e-mail, and symptoms in order to identify their potential health conditions and suggest treatments.24 Entities that are not covered by HIPAA remain free of its regulatory obligations. For example, no matter how much health information employers, websites, or marketers possess, they are not required to implement HIPAA’s data security measures. An additional concern is that many US healthcare providers outsource work such as medical transcription, billing, and reading radiological tests (x-rays, computed tomography [CT] scans, magnetic resonance imaging [MRIs]) to workers in developing countries as a cost-saving measure.25 While the regulations technically cover those engaged in such offshore work as “business associates,” in reality, it is extremely unlikely that the US government would be able to reach such individuals for purposes of any enforcement action in case of noncompliance. Therefore, the practice of outsourcing to offshore professionals creates another regulatory gap and exacerbates patients’ privacy vulnerabilities. Regulatory Exceptions. The HIPAA regulations’ scope is further limited by numerous exceptions that allow covered entities to disclose health information without patient authorization. First, covered entities may divulge patients’ medical information without their permission for purposes of treatment, payment, and healthcare operations.26 Thus, physicians can consult colleagues or speak to nurses about a patient and can have administrators review records for billing or other office-related purposes without the patients’ knowledge. Reportedly, during the course of a typical hospitalization, up to 20  45 C.F.R. §§ 160.102–160.103 (2018); 42 U.S.C. §17934 (2010). 21  Sharona Hoffman & Andy Podgurski, In Sickness, Health, and Cyberspace: Protecting the Security of Electronic Private Health Information, 48 B.C. L. Rev. 331, 337 (2007). 22  42 U.S.C. §12112(d) (2010). 23  Sharona Hoffman, Employing E-Health: The Impact of Electronic Health Records on the Workplace, 19 SPG Kan. J.L. & Pub. Pol’y 409, 409–410 (2010). 24 WebMD, WebMDsymptomchecker, http://symptoms.webmd.com/#introView. 25  Nir Kshetri & Dholakia Nikhilesh Offshoring of Healthcare Services: The Case of the Indian Medical Transcription Offshoring Industry, 25 J. Health Org. & Management 94, 94 (2011). 26  45 C.F.R. §164.506 (2018).

Privacy and Integrity of Medical Information   423 150 individuals may see the patient’s records, including doctors, nurses, medical technicians, and billing clerks.27 In addition, healthcare providers do not need to obtain patient authorization for disclosures that are (1) required by law; (2) necessary for public health activities; (3) related to victims of abuse, neglect, or domestic violence; (4) required for purposes of health oversight activities; (5) necessary for judicial and administrative proceedings; (6) required for law enforcement purposes; (7) made in order to facilitate cadaveric organ, eye, or tissue donation; (8) provided for medical research purposes (with certain privacy safeguards in place); (9) necessary to avert a serious threat to health or safety; (10) needed for specialized government functions, such as national security and  intelligence activities; and (11) authorized by law in order to provide workers’ compensation.28 In general, the HIPAA exceptions are reasonable and logical. However, Americans must understand that they may have little awareness of who is viewing their medical information and for what purposes it is being used. The Definition of “Protected Health Information” and Data De-identification. A third important limitation relates to the type of information that the privacy regulations protect. PHI is defined as “individually identifiable health information” that is electronically or otherwise transmitted or maintained.29 However, great volumes of information are stored in de-identified form in databases used for nontreatment purposes such as research, quality assessment, and public health initiatives.30 De-identified data are entirely exempt from HIPAA coverage and can be disclosed without patient authorization. Data de-identification is explored further in Section 4.

2.1.2  Additional Federal Laws Relevant to Medical Privacy The HIPAA Privacy and Security Rules are not the only federal laws or regulations to address medical privacy. Several important federal laws include privacy safeguards. For example, the Americans with Disabilities Act (ADA) allows employers to subject applicants and employees to medical inquiries and examinations but mandates that they maintain the confidentiality of medical information and store it separately from other employee records.31 Likewise, the Family Educational Rights and Privacy Act (FERPA) governs records held by educational institutions, including those containing health data.32 The Genetic Information Nondiscrimination Act (GINA) prohibits employers and insurers from seeking genetic information in order to ensure the privacy of such data.33 An additional privacy safeguard is furnished by the federal research regulations, 27  Judy Foreman, At Risk of Exposure: In the Push for Electronic Medical Records, Concern Is Growing About How Well Privacy Can Be Safeguarded, L.A. Times, June 26, 2006, http://articles.latimes. com/2006/jun/26/health/he-privacy26. 28  45 C.F.R.§§ 164.502–512 (2018). 29  45 C.F.R. § 160.103 (2018). 30  Hoffman & Podgurski, supra note 7, at 506–515. 31  42 U.S.C. § 12112(d)(4)(B) (2010). 32  20 U.S.C. § 1232g (2010). 33  Genetic Information Nondiscrimination Act of 2008, Pub. L. No. 110–223, 122 Stat. 881 (2008) (codified as amended in scattered sections of 29 and 42 U.S.C.).

424   Sharona Hoffman and Jean Herveg known as the “Common Rule,” which mandate that investigators ask research participants for permission to use their identifiable health information.34 None of these laws includes detailed guidelines as to how data security should be maintained. Nevertheless, professionals handling medical information of any type should be familiar with all relevant privacy regulations.

2.1.3  State Statutory and Common Law Privacy Rights Medical privacy rights have been codified in the statutes of all of the states and the District of Columbia. To varying degrees, the state law provisions (1) allow patients access to their medical records; (2) restrict data use and disclosure by providers, employers, government agencies, and others; (3) establish legal privileges, such as the psychotherapist–patient privilege; (4) address specific conditions, such as alcohol or substance abuse, cancer, genetic testing, sexually transmitted disease, HIV/AIDS, and mental health; and (5) require breach notification in particular circumstances.35 Many states have laws that provide stronger privacy protections than HIPAA. For example, they may cover a broader range of entities that handle health information or provide a private cause of action for privacy breaches.36 Some experts consider the HIPAA Privacy Rule to constitute the floor of privacy protections, with states furnishing additional safeguards.37 Like federal law, state laws carve out significant exceptions to their privacy mandates. The states have all established reporting requirements. Generally, healthcare providers must report to state government agencies incidents of particular conditions, such as infectious disease, such as COVID-19 and HIV/AIDS, cancer, and congenital defects, and their reports must include personally identifying details.38 The government, therefore, has significant collections of patient information. The states have also adopted “duty to warn” statutes that either permit or require healthcare providers to disclose patient information in particular circumstances. Generally, disclosure to law enforcement authorities and potential victims is required or permitted if a patient appears intent on harming himself or others. Thus, a mental healthcare provider with whom a patient discussed a well-formed plan to engage in vio-

34  45 C.F.R. §§ 46.102(f) & 46.116 (2018). 35  Americans Health Lawyers Association, State Healthcare Privacy Law Survey (2013); LawAtlas, Public Health Departments and State Patient Confidentiality Laws Map, http://lawatlas.org/ preview?dataset=public-health-departments-and-state-patient-confidentiality-laws. 36  Cal. Civ. Code §§56-56.06 and 1798.84(b) (2018); Tex. Health & Safety Code Ann. § 181.001(b)(2) (2015). 37  Mary Butler, Is HIPAA Outdated? While Coverage Gaps and Growing Breaches Raise Industry Concern, Others Argue HIPAA Is Still Effective, 88 j. ahima 14–17 (April 2017), https://bok.ahima.org/ doc?oid=302073#.XPLIP4hKg2w. 38 Joy L. Pritts, Altered States: State Health Privacy Laws and the Impact of the Federal Health Privacy Rule, 2 Yale J. Health Pol’y, L. & Ethics 325, 335 (2002); LawAtlas, Public Health Departments and State Patient Confidentiality Laws Map, http://lawatlas.org/preview?dataset=public-health-departmentsand-state-patient-confidentiality-laws.

Privacy and Integrity of Medical Information   425 lence could not maintain confidentiality with respect to that discussion.39 Physicians may also be required to disclose private medical information to third parties in order to warn individuals that they are at risk of having been exposed to a disease.40 The enactment of many of these statutes followed the well-known ruling in Tarasoff v. The Regents of the University of California.41 The case involved a patient who murdered a woman after disclosing to his psychologist that he intended to do so because she had rejected him as a romantic partner. The Supreme Court of California held that therapists who determine or should recognize that their patients pose a serious risk of violence to others have an obligation to “use reasonable care to protect the intended victim[s] against such dangers,” even at the cost of breaching patient confidentiality.42

2.2  European Law: The Organization of the Legal Framework In Europe, medical information privacy and confidentiality are protected by fundamental rights: the right to respect for private life and the right to data protection. This statement requires some further elaboration. Formally, European law43 recognizes the “right to private and family life” in the European Convention on Human Rights, a Council of Europe instrument to which 49 European states, including all Member States of the EU, are signatories. The recognition of a right to data protection has been part of the evolution and development of the data protection rules, both in their adoption in legislation and executive implementation and their judicial interpretation, at the level of Member States, including national courts, and through the case law of the Council of Europe’s European Court of Human Rights (in Strasbourg, France) and the EU’s Court of Justice (Luxembourg). The right to data protection has thus emerged and been developed as a phenomenon of reciprocal influences between different levels of legislative, executive, and judicial powers in Europe. The result is a relatively complex set of laws, few of which are specifically designed to regulate medical privacy and confidentiality. Overall, therefore, medical information confidentiality and privacy is ensured by data protection rules adopted at the European level and implemented at the Member State level and by medical professional privilege and secrecy rules adopted at the level of each European

39  Ahmad Adi & Mohammad Mathbout, The Duty to Protect: Four Decades After Tarasoff, Am. J. Psych. Residents’ J. 1–8 (Apr. 2018), https://ajp.psychiatryonline.org/doi/pdf/10.1176/appi. ajp-rj.2018.130402. 40  See Bradshaw v. Daniel, 854 S.W.2d 865 (Tenn. 1993). 41  Tarasoff v. Regents of University of California, 17 Cal. 3d 425 (Cal. 1976). 42  Id. at 431. See also American Law Institute, Restatement Third of the Law, Torts: Liability for Physical and Emotional Harm, § 41 (2010). 43  In the strict sense, European law refers to the law originating from the EU institutions and bodies. In a broader sense, it also refers to the law emanating from the Council of Europe. In this section, we refer to the broad notion of European law.

426   Sharona Hoffman and Jean Herveg state. Neither set of rules derogate from the other: each must be applied alongside the other.44

2.2.1  Council of Europe Law The first work on data protection at the European level began at the Council of Europe in the late 1960s, with the adoption of two recommendations on automatic processing of personal data which shaped the first outline of the legal framework for ensuring data protection in Europe. These recommendations concerned databases in the private sector45 and the public sector.46 The continuation and development of the Council of Europe’s activities in data protection resulted in the adoption of the 1981 Convention for the Protection of Individuals with Regard to Automatic Processing of Personal Data (Treaty no. 108), updated in 2018,47 as well as numerous sectoral or thematic recommendations, among which are included recommendations on medical data.48 Relatively early in time several cases related to data protection were brought before the European Court of Human Rights. Since the Z v. Finland judgment in 1997, the European Court of Human Rights Court has repeatedly and consistently proclaimed that (1) The protection of personal data (and health information are not the least) plays a fundamental role in the exercise of the right to respect for private and family life; (2) Respecting the confidentiality of health information is an essential principle of the legal system of all Contracting Parties to the Convention; it is essential not only to protect patients’ privacy but also to preserve their confidence in the medical profession and health services in general. Without such protection, persons requiring medical care could be discouraged from providing the personal and intimate information necessary to get the appropriate treatment and even to consult a doctor. That could end up jeopardizing their health or, in case of communicable diseases, that of the community. (3) Domestic legislation should therefore provide appropriate safeguards to prevent the use of personal data and in particular any communication or disclosure of personal data relating to health, which does not comply with the guarantees provided by the Article 8 right to private and family life of the European Convention on Human Rights.

44  This means, notably, that, in this extent, professional secrecy rules may not oppose the data subject’s right of access under data protection law. 45  Council of Europe, Resolution (73) 22. 46  Council of Europe, Resolution (74) 29. 47  This Convention has been revised, and the Convention 108+ was adopted by the Committee of Ministers on May 18, 2018, at its 128th meeting. 48  Recommendation 97(5) on the protection of medical data is also under revision (see doc. T-PD(2018)06).

Privacy and Integrity of Medical Information   427 In addition to this assertion of the importance and need to protect personal data for the exercise of the right to respect for private and family life,49 the European Court of Human Rights has developed a substantial case-law in many areas relevant to data protection. Those most relevant to medical privacy and confidentiality include protection of medical data, medical records, medical records security, access rights (including the right to get a copy), data security, and genetic testing. For example, in the case of I. v. Finland (2008), the Court judged that Finland failed to provide a practical and effective protection excluding any possibility of unauthorized access to a medical file.

2.2.2  European Union Law At the level of the European Community (now the EU), the issue of data protection was formally embraced by the European Parliament on April 8, 1976. It resulted in the ­adoption, in 1979, of a Resolution on the protection of human rights in the face of the development of technical progress in the field of informatics.50 Then, after the adoption of the Organisation for Economic Cooperation and Development (OECD) Guidelines for the Protection of Privacy and Transborder Data Flows in 1980, the European Community adopted in 1995 the Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data.51 This was the first binding EU legal instrument on data protection. The right to data protection was explicitly and formally recognized at the “constitutional” level in EU law in the Charter of Fundamental Rights of the European Union on December 7k 2000, Article 8.52 The Treaty on the Functioning of the European Union also recognizes, under its provisions of general application, the right to data protection.53 From May 25, 2018, the GDPR ensures data protection in (and in some circumstances beyond) the EU.54 The GDPR is applicable in 27 countries and concerns directly more than 430,000,000 people within the EU. It also has significant indirect effects, notably in the matter of transfers of personal data to third countries or international organizations.55 49  On the basis of which it could already be argued that each state has a positive obligation to protect personal data. 50  OJ 5 June 1979 no. C 140/34. 51  OJ L 281 23 November 1995 p. 31 (take into account the consolidated text). 52  Charter of fundamental rights of the EU, 2016/C 202/02. See Working Party on the Protection of Individuals with Regard to the Processing of Personal Data Recommendation 4/99 on the inclusion of the fundamental right to data protection in the European catalogue of fundamental rights WP 26 7 September 1999. 53  See Article 16 TFEU. 54  Regulation (EU) 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), OJ L 119 4 May 2016 p. 1. 55  On the Regulation, see: S GUTWIRTH, R LEENES, & P. DE HERT (ed.), Reforming European Data Protection Law, Law, Governance and Technology Series, Issues in Privacy and Data Protection, volume 20 (Springer, 2015).

428   Sharona Hoffman and Jean Herveg It is to this extent that any person who comes under the jurisdiction of a Member State56 has the right to claim the protection of his or her personal data. It is not only an obligation on the part of the healthcare professional or the Member State but also, and above all, a fundamental right which the data subject can claim.57 2.2.2.1  The Boundaries of the GDPR: Material Scope At the level of EU law, the scope of protection for medical information confidentiality and privacy is defined by the scope of the GDPR. The GDPR protects individuals with regard to the “processing” of “personal data,” including personal data related to health. In order to be protected by the Regulation, personal data must be automatically processed, in whole or in part, or at least be included in a paper file.58 Any use of personal data is a kind of data processing. So is the mere fact of looking at personal data. But to be subject to the GDPR, the data processing must be, in addition, automated or be part of a paper filing system. “Personal data” means any information relating to an identified or identifiable natural person (data subject). An identifiable natural person is a human being who can be identified, directly or indirectly, in particular by reference to an identifier. The data subject does not have to be identified. It only has to be possible to identify the data subject.59 The definition of personal data remains substantially unchanged from the earlier Directive, except for the description of the elements likely to help the identification of the data subject.60 According to the GDPR and the case-law of the Court of Justice of the EU, the concept of personal data must be interpreted as widely as possible. For instance, IP addresses or codes for medical nomenclature (this refers to the classification of medical acts through specific codes for social security purposes) are personal data. We could question the relevance of the notion of “personal data,” especially in the context of development of big data and data mining or of artificial intelligence, and argue that the focus should be put on the impact of technologies on the citizen for

56  In the meaning of the first Article of the European Convention on Human Rights to which Article 52 of the Charter of Fundamental Rights of the EU refers. 57  On the right to data protection, see: G. GONZALEZ FUSTER, The Emergence of Personal Data Protection as a Fundamental Right of the EU, Law, Governance and Technology Series, Issues in Privacy and Data Protection, volume 16 (Springer, 2014); B. van der Sloot, “Legal Fundamentalism: Is Data Protection Really a Fundamental Right ?” in R. LEENES, R. van BRAKEL, S. GUTWIRTH, & P. DE HERT, Data Protection and Privacy: (In)visibilities and Infrastructures, Law, Governance and Technology Series, Issues in Privacy and Data Protection, volume 36 (Springer, 2017), 3. 58  The filing system means any structured set of personal data which are accessible according to specific criteria, whether centralized, decentralized, or dispersed on a functional or geographical basis (Article 4.6 of the Regulation). 59  Article 4.1 of the Regulation. 60  Such as a name, an identification number, location data, an online identifier, or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural, or social identity of that natural person.

Privacy and Integrity of Medical Information   429 justifying and developing their regulation, rather than on the personal nature of the data per se. 2.2.2.2  Health Data in the GDPR Personal data concerning health are a special category of personal data. They are defined in the GDPR as “personal data related to the physical or mental health of a natural person, including the provision of healthcare services, which reveal information about his or her health status.”61 Personal data concerning health should include all data pertaining to the health status of a data subject which reveal information relating to the past, current, or future physical or mental health status of the data subject. This includes information about the natural person collected in the course of the registration for or the provision of healthcare services, as referred to in Directive 2011/24/EU on crossborder healthcare. Personal data concerning health also include a number, symbol, or particular assigned to a natural person to uniquely identify the natural person for health purposes. It covers information derived from the testing or examination of a body part or bodily substance, including from genetic data and biological samples. It also covers any information on, for example, a disease, disability, disease risk, medical history, clinical treatment, or the physiological or biomedical state of the data subject. The notion of personal data concerning health is regardless of its source (e.g., whether from a physician or other health professional, a hospital, a medical device, or an in vitro diagnostic test).62 This definition is very broad to the extent that it covers data which, in themselves, do not relate to the data subject’s health, but from which we could extract information concerning the data subject’s health (such as from data collected when a patient is registered in a hospital: at first sight, these are not related to the patient’s health but are merely administrative information). It also covers data that do not relate to the data subject’s health, but which could help in finding information concerning the data subject’s health (such as a patient’s identifier in a national healthcare system). 2.2.2.3  The Link Between Personal Health Data and Genetic Data In its 2004 working document on genetic data,63 the Article 29 Data Protection Working Party considered that, in the majority of cases, genetic data were personal data. It added that genetic data may provide to an extent a detailed picture of a person’s physical disposition and health condition and therefore could be considered as “data concerning health.” However, it added that genetic data may also describe specific forms of a wide range of physical characteristics, and, in this case, genetic data that determine the color of someone’s hair, for example, may not be regarded as data directly concerning health. Today, the GDPR defines genetic data as personal data relating to the inherited or acquired genetic characteristics of a natural person and providing unique information 61  Article 4.15 of the Regulation.

62  Recital 35 of the Regulation.

63  Article 29 Data Protection Working Party Working, Document on Genetic data, 17 March 2004, WP 91.

430   Sharona Hoffman and Jean Herveg about the physiology or the health of that natural person. In addition, the data must result from the analysis of a biological sample from the natural person in question, in particular chromosomal, deoxyribonucleic acid (DNA), or ribonucleic acid (RNA) analysis, or from the analysis of another element enabling equivalent information to be obtained.64 It can be deduced that, although they should not be confused, genetic data may qualify as personal data concerning health, depending upon the circumstances. 2.2.2.5  Definition of Personal Data Concerning Health in the GDPR We should try to keep a strict and objective definition of the notion of personal data concerning health. It should be limited to data containing an element of knowledge about an individual’s health status, excluding by that any attempt to cover data that do not contain any information on an individual’s health, even if it is possible to deduce personal data concerning health from the information (mainly because of the purpose pursued by the data processing or the context). Furthermore, any data deduced or extracted will automatically be protected under the GDPR by reason of its informational content. In doing so, the data from which information on an individual’s health has been deduced or extracted will not be governed by a status which is not appropriate for these data and which does not offer, in reality, any real protection for the data subject. Moreover, it is not sensible to include within the GDPR’s definition of health data, those data that are not related to health but from which one can deduce information on the data subject’s health. That is for a very simple reason: it is now possible to deduce information relating to an individual’s health from a multitude of completely unspecified data. It is therefore better in terms of legal precision to circumscribe the notion to only data which contain information about the data subject’s health. It should be remembered that data concerning health do not need to come from a health professional, result from an act reserved to health professionals, or arise in the context of health institutions or systems. In addition, data may concern health even when those data are not processed for therapeutic purposes. This is particularly the case with regard to insurance or credit card payments. Moreover, the mere information relating to a physical or psychological aspect of an individual does not necessarily constitute, as such, data concerning health. To gain this last legal qualification, the physical or psychic aspect must teach us something about the health of the data subject. In this sense, data concerning health are all pieces of information relating to the physical or mental health, past, present or future, of a natural person, living or dead. In practice, all patient data collected in European healthcare settings will fall within the protections of the GDPR either as plain personal data or as a special category of personal data.

64  See Recital 34 of the Regulation.

Privacy and Integrity of Medical Information   431 2.2.2.6  The Boundaries of the GDPR: Territorial Scope To be protected by the GDPR, the data processing situation has to fall within its territorial scope. The GDPR applies to the processing of personal data in the context of the activities of an establishment of a data controller65 or a processor66 in the EU, regardless of whether the processing takes place in the EU or not.67 It is thus beyond doubt that the processing of a patient’s data carried out by a healthcare professional established in the EU, providing healthcare to a patient in Europe, falls under the scope of the Regulation. If the data controller or processor is not established in the EU, the Regulation applies to the processing of personal data of data subjects who are in the EU where the processing activities are related to the offering of goods or services to such data subjects in the EU or to the monitoring of data subjects’ behavior as far as their behavior takes place within the EU.68 But the Regulation does not specify what is meant by a person who is on the ­territory of the EU. This concept may cover accidental or tourist presence, transit, mere residence, domicile, or principal or secondary establishment in the territory of the EU (i.e., within the territory of a Member State of the EU). Moreover, these notions do not have necessarily the same meaning in all Member States. Finally the Regulation applies to the processing of personal data by a controller not established in the EU but in a place where Member State law applies by virtue of public international law. The territorial reach of the GDPR is thus wide. It therefore applies in a range of ­cross-border healthcare contexts involving states outside of the EU. 2.2.2.7  The Boundaries of the GDPR: Personal Scope Like the earlier Convention of 1981 or the Data Protection Directive of 1995, the GDPR does not explicitly determine its personal scope. However, the Regulation identifies the main actors in data processing. As in the Directive, the data controller is the person who, alone or jointly with others, determines the purposes and means of the data processing69 and the processor is the one who processes personal data on behalf of the data controller.70 Doctors, other health professionals, hospitals, and other providers of care 65  The [data] controller means the natural or legal person, public authority, agency, or other body which, alone or jointly with others, determines the purposes and means of the processing of personal data; where the purposes and means of such processing are determined by Union or Member State law, the controller, or the specific criteria for its nomination may be provided for by Union or Member State law (Article 4.7 of the Regulation). See Article 29 Data Protection Working Party Opinion 1/2010 on the concepts of “controller” and “processor” WP 169 16 February 2010. 66  The processor means a natural or legal person, public authority, agency, or other body which processes personal data on behalf of the controller (Article 4.8 of the Regulation). 67  Article 3.1 of the Regulation. 68  Article 3.2 of the Regulation. 69  Article 4.7 of the Regulation. See Article 29 Data Protection Working Party Opinion 1/2010 on the concepts of “controller” and “processor” WP 169 16 February 2010. 70  Article 4.8 of the Regulation.

432   Sharona Hoffman and Jean Herveg all fall into the category of data controller or processor. The Regulation also identifies the recipient,71 the third party,72 the representative,73 the enterprise,74 and the group of undertakings.75 There are many controversies about who qualifies as data controller, joint data controller, data processor in hospitals and in the healthcare sector in general. As with the earlier Directive, the GDPR still does not provide a formal definition of the data subject, even though the latter is supposed to be at the heart of the regulatory system. In any case, the Regulation insists on the point that the protection applies irrespective of the nationality or residence of the data subject.76 Protection under the GDPR extends to persons who are not nationals of any Member State and who do not reside in the territory of any Member State but whose data are processed by a data controller subject to the Regulation. This approach means that all patients whose personal data are processed in the context of healthcare systems or situations within Member States of the EU are covered by the GDPR’s protections. In any case, all these actors must be properly identified when processing personal data. This can lead to some problems, in particular in the context of Internet platforms for patient’s data communication, cloud computing services,77 or mobile applications (mHealth).78 Who assumes the role of data controller, joint data controller, or data processor? 71  The “recipient” means a natural or legal person, public authority, agency, or another body to which the personal data are disclosed, whether a third party or not. However, public authorities which may receive personal data in the framework of a particular inquiry in accordance with Union or Member State law shall not be regarded as recipients; the processing of those data by those public authorities shall be in compliance with the applicable data protection rules according to the purposes of the processing (Article 4.9 of the Regulation). 72  The “third party” means a natural or legal person, public authority, agency, or body other than the data subject, controller, processor, and persons who, under the direct authority of the controller or processor, are authorized to process personal data (Article 4.8 of the Regulation). 73  The “representative” means a natural or legal person established in the Union who, designated by the controller or processor, represents the controller or processor with regard to their respective obligations (Article 4.8 of the Regulation). 74  The “enterprise” means a natural or legal person engaged in an economic activity, irrespective of its legal form, including partnerships or associations regularly engaged in an economic activity (Article 4.8 of the Regulation). 75  The “group of undertakings” means a controlling undertaking and its controlled undertakings (Article 4.8 of the Regulation). 76  See Recital no. 14. In any case, this protection is expressly excluded for legal persons (see Recital no. 14). The Regulation is, however, once again ambiguous. Indeed, it states that “This Regulation does not cover the processing of personal data which concerns legal persons and in particular undertakings established as legal persons, including the name and the form of the legal person and the contact details of the legal person.” This last sentence seems to imply a form of derogation, which would mean that there could be some form of protection for other data related to enterprises. In theory, this would be wrong, but this recital brings unnecessary doubt. 77  See Article 29 Data Protection Working Party Opinion 05/2012 on Cloud Computing WP 196 1 July 2012. 78  See Article 29 Data Protection Working Party Opinion 02/2013 on apps on smart devices WP 202 27 February 2013.

Privacy and Integrity of Medical Information   433 2.2.2.8  Substantive/Material Rules of the GDPR on Medical Information Privacy and Confidentiality The GDPR enumerates common uniform substantive rules and details the principles applicable to all data processing within its scope, including to the processing of personal data concerning health. The principles are not that substantially different from the rules previously laid down by the earlier Data Protection Directive. Seven principles apply to all processing of personal data. Principles of personal data processing. Personal data must be processed lawfully, fairly, and in a transparent manner in relation to the data subject (principles of lawfulness, fairness, and transparency).79 Personal data must be collected for specified, explicit, and legitimate purposes and not further processed in a manner that is incompatible with those purposes (principle of purpose limitation).80 Further processing for archiving purposes in the public interest, scientific or historical research purposes, or statistical purposes should not be considered as incompatible with the initial purposes, provided that it is subject to appropriate safeguards for the rights and freedoms of the data subject. These guarantees must ensure that technical and organizational measures are set in place to ensure compliance with the data minimization principle.81 Whenever possible, further processing should not, or no more than previously, allow for the identification of the data subject. Personal data must be adequate, relevant, and limited to what is necessary in relation to the purposes for which they are processed (principle of data minimization). Personal data must be accurate and, where necessary, kept up to date. Every reasonable step must be taken to ensure that personal data that are inaccurate in regard to the purposes for which they are processed are erased or rectified without delay (principle of accuracy). Personal data must be kept in a form that permits identification of data subjects for no longer than is necessary for the purposes for which the personal data are processed (principle of storage limitation). Personal data may be stored for longer periods insofar as the personal data will be processed solely for archiving purposes in the public interest, scientific or historical research purposes, or statistical purposes provided that it is subject to appropriate safeguards for the rights and freedoms of the data subject. These guarantees must ensure that technical and organizational measures are set in place to

79  See Article 29 Data Protection Working Party Guidelines on Transparency under Regulation 2016/679 WP 260 rev.01 11 April 2018. 80  See Article 29 Data Protection Working Party Opinion 03/2013 on purpose limitation WP 203 2 April 2013. 81  These measures may include pseudonymization, to the extent that these purposes can be achieved in this way. Pseudonymization means the processing of personal data in such a manner that the personal data can no longer be attributed to a specific data subject without the use of additional information, provided that such additional information is kept separately and is subject to technical and organizational measures to ensure that the personal data are not attributed to an identified or identifiable natural person (Article 4.5 of the Regulation).

434   Sharona Hoffman and Jean Herveg ensure compliance with the data minimization principle.82 Whenever possible, further processing should not or no more allow for the identification of the data subject. Personal data must be processed in a manner that ensures an appropriate security of the personal data, including protection against unauthorized or unlawful processing and against accidental loss, destruction, or damage using appropriate technical or organizational measures (principle of integrity and confidentiality). Finally, the controller is responsible for compliance with the principles applicable to the processing of personal data. The controller must also, and that is formally new, be able to demonstrate that the data processing is compliant with these principles (principle of accountability).83 2.2.2.9  Data Processing Lawfulness The GDPR, Article 6, lists the categories of situations in which it is a priori lawful (i.e., permitted by law) to process personal data.84 These include where the data subject has given consent for specific purposes, or to protect the vital interests of the data subject or another natural person, or is necessary for a task carried out in the public interest. At first glance, it would seem that processing of health data is barred by the GDPR. As with personal data revealing racial or ethnic origin, sexual orientation, political opinions, religious or philosophical beliefs, or trade union membership, the processing of data concerning health,85 genetic data, or biometric data for the purpose of uniquely identifying a natural person is in principle prohibited.86 But this general prohibition does not apply to many situations in which health data are processed in the healthcare setting. Article 9.2 of the GDPR lists the categories of situations in which it is lawful to process personal data. This includes where explicit consent is given, to protect the vital interests of the data subject, or processing is necessary for reasons of substantial public interest. To put it another way, each of these categories is supposed to represent a situation in which the interests involved are in an acceptable balance. The interests to be taken into consideration are those of the data controller, the data subject, and the community. In line with the legitimation mechanisms set up by the earlier Directive, it is of course necessary to verify in each individual case, for each data processing taken and considered separately and individually, whether there is a fair balance among these three kinds of interests. This balance must be done in concreto and not only a priori and in abstracto. In this respect, changing the balance of interests over time (e.g., because of changing understandings of the interests of communities, data subjects, or data controllers arising 82 Id. 83  See Article 29 Data Protection Working Party Opinion 3/2010 on the principle of accountability WP 173 13 July 2010 84  See Article 6 of the Regulation and the possibility of special arrangements for processing imposed by law or carried out in the public interest or in the exercise of official authority by the controller and the flexibility of the criterion for the compatibility of further data processing. See Article 29 Data Protection Working Party Guidelines on Consent under Regulation 2016/679 WP 259 rev.01 10 April 2018. 85  The definition of health data is discussed supra, at 2.2.2.5. 86  Article 9.1 of the Regulation.

Privacy and Integrity of Medical Information   435 from new technologies) will have the effect of removing the legitimacy of the data processing for the future. The data processing will have to be stopped unless a solution is found to satisfactorily rebalance the interests involved. Surprisingly, while one of the objectives of the reform of the legal framework for data protection was to eliminate inconsistencies between Member States regarding the processing of personal data relating to health, the GDPR provides that, in respect of the subsidiarity principle, matters should be dealt with at the most local level at which they can be resolved, and Member States may maintain or introduce further conditions, including limitations, with regard to the processing of genetic data, biometric data, or data concerning health.87 It follows that the differences between Member States, which have been strongly condemned, are likely to increase in the matter of personal data related to health. Furthermore, the GDPR does not lay down criteria for delimiting the territorial scope of the national provisions that Member States might adopt regarding the processing of genetic data, biometric data, or health data.88 It remains, of course, that, in any case, without prejudice to the Council of Europe’s Treaty No. 108, Member States are bound by the common legal framework that emerges from the case-law of the European Court of Human Rights in the field of data protection and by the rights therefore granted to individuals in terms of data control (situations in which the Court considers that the person is entitled to expect that data will not be disclosed without his or her consent), data access (including access to medical records), or medical records security, for example. In addition to these Council of Europe and EU-level rules, therefore, medical information privacy and confidentiality is subject to national substantive rules which differ across European states.

2.2.3  State Laws The protection of health data is integrated in the general framework which has been set up for protecting personal data in Europe. Hence, unlike in the US, the protection of health data is not restricted to a specific sector of activities (healthcare), to certain professionals (health practitioners) or institutions (health service providers), or to particular categories of citizens (patients). The European legal context is thus quite distinctive as compared to the usual features of the notion of medical professional privilege or professional secrecy. This raises the issue of the distinction between data protection rules (promulgated at European level) and medical professional privilege or professional secrecy rules (promulgated at national levels). As EU law applies only within the scope of EU competence, and Council of Europe law leaves a significant “margin of appreciation” to states parties, despite the European-level norms there remains room for distinctive approaches to medical privacy and confidentiality through the vector of professional privilege and secrecy law.

87  Article 9.4 of the Regulation.

88  Article 9.4 in fine of the Regulation.

436   Sharona Hoffman and Jean Herveg The transposition of the 1995 Data Protection Directive89 into Belgian Law is a good example. To what extent did Belgium’s implementation of data protection rules have an impact on professional secrecy rules? More specifically, did the data subject’s consent, used as a basis of legitimization for the processing of personal data under data protection rules, also discharge healthcare practitioners of their duty not to disclose any information about their patients under provision 458 of the Belgian Criminal Code (the latter protecting professional secrecy)? For instance, in this context, might the patient allow the health practitioner to send medical information to the insurance company? In a classical criminal point of view, it was not permissible for a patient to discharge healthcare practitioners from their duty not to disclose information about them except in some situations vested in the law, the case-law, and the legal literature. The issue was even more sensitive in that it was part of a larger discussion about patients’ rights at national, European, and international levels and especially about empowering patients in the therapeutic relationship (which could concern the power to master the fate of information protected by professional secrecy). Nevertheless, the answer was obvious: it had never been the intent of European data protection rules to modify professional secrecy rules. Indeed, the objective of data protection rules was and still is to protect individuals against the development of information and communication technologies and, in particular, to protect them against paper files and automated processing of personal data and to entrust them with new subjective rights. There has never been any intent to interfere with the regulation of healthcare professions or with the (legal and ethical) duties of healthcare practitioners. Two arguments supported this interpretation. First, the Data Protection Directive referred to professional secrecy rules90 without any hint that it intended to modify them. Therefore, it was already possible to infer from this that professional secrecy rules were separate from data protection rules. Second, the duty to lawfully process personal data91 was interpreted as meaning that it included the duty to comply with the special rules applicable to the processed data, which refers in this case to professional secrecy rules.92 The question is now whether this interpretation is still valid since the adoption of the GDPR. Three reasons suggest that it is. First, the GDPR pursues the same goal as the Directive, though in a more elaborate way. In addition, the GDPR also refers to professional secrecy rules without saying it would impact or modify them.93 Rather, as in the case of the preceding Directive, the GDPR provides that the ban on processing 89  Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data. 90  See Recital 33 of the Directive 95/46/EC and Article 8.3. 91  Article 6.1. a) of Directive 95/46/EC. 92  See: J. Herveg, M.-N. Verhaegen, & Y. Poullet, Les droits du patient face au traitement informatisé de ses données dans une finalité thérapeutique: les conditions d’une alliance entre informatique, vie privée et santé, 2 Revue du Droit de la Santé, 56–84 (2002–2003); C. De Terwangne, «Les cabinets d’avocats et la loi sur la protection des données à caractère personnel », in Cabinets d’avocats et technologies de l’information: balises et enjeux (Bruxelles, AcademiaBruylant, 2005), Cahiers du CRID, no. 26, p. 156. 93  See Article 9.3 of the Regulation on the processing of special categories of personal data.

Privacy and Integrity of Medical Information   437 personal data does not apply to the special categories of personal data (among them those related to health) where those are processed for therapeutic purposes by a health professional subject to the obligation of professional secrecy (under Union or Member State law or rules established by national competent bodies) or, under responsibility of the latter, by another person also subject to an obligation of secrecy (under Union or Member State law or rules established by national competent bodies). Second, Member States are permitted under the GDPR to maintain or introduce further conditions (including limitations) with regard to the processing of genetic data, biometric data, or data concerning health.94 This national discretion covers the question of maintaining, changing, or adopting new professional secrecy rules. Third, the GDPR explicitly provides that Member States may adopt specific rules to set out the powers of the supervisory authorities in relation to controllers or processors that are subject, under Union or Member State law or rules established by national competent bodies, to an obligation of professional secrecy or other equivalent obligations of secrecy where this is necessary and proportionate to reconcile the right of the protection of personal data with the obligation of secrecy.95 Thus, the GDPR has neither the objective nor the effect of modifying existing state-level rules regarding professional secrecy: it refers to them as external rules, and it recognizes the possible application of these external rules, notably when processing personal data related to health for therapeutic purposes. Hence, to summarize, medical data confidentiality and privacy in Europe is regulated at Council of Europe, EU, and national levels, too.

3  Data Security Personal health data can be a sought-after commodity that brings financial and other benefits to numerous parties. For example, employers are interested in health information because they want workers who will not have absenteeism or productivity problems or generate high health insurance costs. Marketers hope to tailor their advertising efforts to consumers’ individualized needs and thus could find certain health details to be a great asset. Even criminals may seek data in order to commit identity theft, credit card fraud, or medical insurance fraud.96 Consequently, without proper security defenses, medical data may be very vulnerable to hacking, theft, and other abuses. This section examines the legal mechanisms that the United States and EU have implemented to promote data security.

94  Article 9.4 of the Regulation. 95  See Article 90 of the Regulation. These rules are only applicable to personal data received or collected by the data controller or data processor in an activity subjected to this obligation of secrecy. 96  Hoffman, supra note 4, at 59–60.

438   Sharona Hoffman and Jean Herveg

3.1  The HIPAA Security Rule The HIPAA Security Rule is less well-known than the Privacy Rule in the United States but is equally important. It delineates administrative, physical, and technical safeguards to protect the confidentiality, integrity, and availability of electronic health information (EHI).97 The administrative safeguard standards include security management processes and workforce security, such as clearance procedures, information access management, security awareness and training, security incident procedures, and contingency plans.98 Physical safeguards focus on facility access controls, workstation security, and device and media controls.99 Technical safeguards are procedures to control access to EHI (e.g., encryption), to audit activity related to processing EHI, to protect EHI from improper modification or elimination, and to obtain authentication from those seeking access to EHI.100 The HIPAA Security Rule suffers from the same limitations that restrict the scope of the HIPAA Privacy Rule. It covers only health plans, healthcare clearinghouses, healthcare providers who transmit health information electronically for purposes of HIPAArelevant transactions, and their business associates.101 Thus, employers, life insurers, educational institutions, marketers, and many others who possess health information do not have to comply with its important data security requirements. Furthermore, because de-identified data do not constitute protected health information,102 they, too, are not governed by the Security Rule and can be stored in ways that do not comply with its standards.

3.2  Data Security in the GDPR Security of personal data in the GDPR forms part of a series of new uniform substantive rules to which data controllers and processors are subject. These include, for instance, record keeping, cooperation with supervisory authorities, and privacy impact assessment. As far as data security is concerned, the data controller and processor must implement appropriate technical and organizational measures to ensure a level of security appropriate to the risk. They must take into account the state of the art; the costs of implementation; and the nature, scope, context, and purposes of processing as well as the risk of varying likelihood and severity for the rights and freedoms of natural persons. In assessing the appropriate level of security, they must take into account in particular the risks presented by the data processing, in particular from accidental or unlawful destruction, loss, alteration, unauthorized disclosure of, or access to personal 97  45 C.F.R. §§ 164.302 -.318 (2018). 98  45 C.F.R. § 164.308 (2018).    99  45 C.F.R. § 164.310 (2018). 100  45 C.F.R. § 164.312 (2018).    101  45 C.F.R. §§ 160.102–160.103 (2018); 42 U.S.C. §17934 (2010). 102  45 C.F.R. § 160.103 (2018).

Privacy and Integrity of Medical Information   439 data transmitted, stored, or otherwise processed.103 Those measures must be reviewed and updated where necessary. Where proportionate in relation to processing activities, these measures must include the implementation of appropriate data protection policies by the data controller.104 Privacy by design imposes obligations on the data controller (and processor) to implement, both at the time of the determination of the means for processing and at the time of the processing itself, appropriate technical and organizational measures (such as pseudonymization) which are designed to implement data-protection principles (such as data minimization) in an effective manner and to integrate the necessary safeguards into the processing in order to meet the requirements of the GDPR and protect the rights of data subjects. In doing so, the data controller has to take into account the state of the art; the cost of implementation; and the nature, scope, context, and purposes of processing as well as the risks of varying likelihood and severity for rights and freedoms of natural persons posed by the processing.105 Privacy by default obliges the data controller (and processor) to implement appropriate technical and organizational measures for ensuring that, by default, only personal data that are necessary for each specific purpose of the processing are processed. That obligation applies to the amount of personal data collected, the extent of their processing, the period of their storage, and their accessibility. In particular, such measures must ensure that, by default, personal data are not made accessible without the individual’s intervention to an indefinite number of natural persons.106 In any case, the data controller and processor must take steps to ensure that any natural person acting under the authority of the controller or the processor who has access to personal data does not process them except on instructions from the controller unless required to do so by Union or Member State law. All this means that hospitals and health practitioners must implement and adapt the general requirements imposed by the GDPR to all data processing. For example, the use of cloud computing by a hospital is not necessarily prohibited by the GDPR. But the implementation of data protection rules could require the use of a public cloud for administrative data and of a private cloud for more sensitive data.

103  See Article 32 of the Regulation. 104  See Article 24 of the Regulation. The application of an approved code of conduct or approved certification mechanisms may serve as a means of demonstrating compliance with the obligations of the data controller. 105  On this, see Article 25.1 of the Regulation. An approved certification mechanism may serve as an element to demonstrate compliance with these requirements. 106  See Article 25.2 of the Regulation. Again, an approved certification mechanism can serve as an element to demonstrate compliance with these requirements.

440   Sharona Hoffman and Jean Herveg

4  Data De-identification, Anonymous Data, and Pseudonymized Data As noted earlier, bountiful health information is stored in de-identified form in databases that are available to researchers, government officials, and sometimes the public at large.107 What is de-identified information? The HIPAA Privacy Rule provides detailed guidance. GDPR distinguishes between anonymous data and pseudonymized data.

4.1  De-Identification Under HIPAA HIPAA states that health information is de-identified if (1) a qualified expert determines that there is only a “very small” risk that the data can be re-identified, and (2) the expert documents his or her analysis.108 The DHHS issued guidance that recognized several effective de-identification techniques: • Suppression: Redacting particular identifiers before data is disclosed (e.g., zip codes, birthdates, income) • Generalization: Transforming particular information into less specific representations (e.g., indicating a 10-year age range instead of exact age) • Perturbation: Exchanging particular data values for equally specific but different values (e.g., changing patients’ ages).109 In the alternative, the HIPAA Privacy Rule lists 18 items that should be removed in order to exempt information from HIPAA coverage. These are names; all geographic subdivisions smaller than a state, including street address, city, county, precinct, zip code, and their equivalent geocodes, except for the initial three digits of a zip code (if, according to the current publicly available data from the Bureau of the Census the geographic unit formed by combining all zip codes with the same three initial digits contains more than 20,000 people, and the initial three digits of a zip code for all such geographic units containing 20,000 or fewer people is changed to 000); all elements of dates (except year) for dates directly related to an individual, including birth date, admission date, discharge date, date of death, and all ages over 89 and all elements of dates (including year) indicative of such age, except that such ages and elements may be 107  Hoffman, supra note 4, at 168–175; supra Part I.A.1.a. 108  45 C.F.R. § 164.514(b)(1) (2018). 109  Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule, U.S. Dep’t of Health & Human Servs. (Nov. 26, 2012), http://www​.hhs​.gov/‌ocr/‌privacy/‌hipaa/‌understanding/‌cover edentities/‌De-identification/‌guidance​.html​#guidance​determination (noting that techniques such as suppression and generalization are often used in combination).

Privacy and Integrity of Medical Information   441 aggregated into a single category of age 90 or older; telephone numbers; fax numbers; email addresses; social security numbers; medical record numbers; health plan beneficiary numbers; account numbers; certificate/license numbers; vehicle identifiers and serial numbers, including license plate numbers; device identifiers and serial numbers; web universal resource locators (URLs); Internet Protocol (IP) address numbers; biometric identifiers, including finger and voice prints; full face photographic images and any comparable images; and any other unique identifying number, characteristic, or code.110 In theory, removal of these 18 elements makes it impossible to connect medical records with patients’ names. However, experts have determined that even with redaction of these identifiers, there is a small chance (perhaps 0.01–0.25%), that skilled attackers could re-identify records. They could do so by matching de-identified data to publicly available information, such as voter registration records or news stories about individuals with illnesses or injuries.111 In some cases, data analysts may need some of the listed data elements to conduct their analysis. Obtaining consent from thousands or even millions of patients in a database may be prohibitively costly and burdensome. Consequently, the HIPAA Privacy Rule creates a further exemption for three purposes: research, public health, and healthcare operations. It allows covered entities to disclose “limited datasets” for these three uses without patient consent if data recipients sign data use agreements containing specified restrictions and privacy protections. Limited datasets remove most of the 18 listed elements, but they retain dates and geographic locales, though not patients’ exact addresses.112 As valuable as these details are for analysts, they may also make it considerably easier for attackers to re-identify records. By some estimates, the risk of reidentification of limited datasets may be as high as 10–60% depending on what other data are publicly available.113

110  45 C.F.R. § 164.514(b)(2)(i) (2018). In addition, information will not be considered de-identified if an entity has “actual knowledge that the information could be used alone or in combination with other information to identify an individual who is a subject of the information.” Id. at §164.514(b) (2)(ii). 111  Nat’l Comm. on vital & Health Statistics, Report to the Secretary of Health and Human Services on enhanced Protections for Uses of Health Data: A Stewardship Framework for “Secondary Uses” of Electronically collected and Transmitted Health Data 36 n.16 (2007), http://bok.ahima.org/PdfView?oid=76869 ; Sharona Hoffman & Andy Podgurski, Balancing Privacy, Autonomy, and Scientific Needs in Electronic Health Records Research, 65 SMU L. Rev. 85, 105–107 (2012). 112  45 C.F.R. §§164.514(e)(1)-(4) (2018). 113  Kathleen Benitez & Bradley Malin, Evaluating Re-identification Risks with Respect to the HIPAA Privacy Rule, 17 J. Am. Med. Informatics Ass’n 169, 169 (2010).

442   Sharona Hoffman and Jean Herveg

4.2  Anonymous Data and Pseudonymized Data Under the GDPR The difference between personal data, anonymous data, and pseudonymized data is important. Personal data and pseudonymized data fall under the scope of the GDPR, and all its rules must be respected. Anonymous data are outside the GDPR’s scope. Anonymous data are data for which it is not reasonably possible to make a link with the data subject. It is always a perilous task to decide whether data are anonymous or not in the healthcare sector and especially in the field of scientific research. Pseudonymized data are data that can no longer be attributed to a specific data subject without the use of additional information, provided that such additional information is kept separately and is subject to technical and organizational measures to ensure that the personal data are not attributed to an identified or identifiable natural person (cf. Article 4 (5) of the GDPR). Pseudonymization is a useful form of data processing that can help the data controller meet data security requirements and comply with privacy by design and privacy by default requirements (e.g., the data minimization principle).

5  Mechanisms for Enforcement of Medical Information Confidentiality, Remedies for Privacy Breaches, and Enforcement of Privacy Rights 5.1  American Law 5.1.1  The HIPAA Privacy and Security Rules The most extensive American privacy regulations, the HIPAA Privacy and Security Rules, do not offer aggrieved individuals a private cause of action. Thus, individuals whose data were disclosed to unauthorized parties cannot sue wrongdoers for damages under federal law. Instead, enforcement is left up to the DHHS Office for Civil Rights (OCR) and state attorneys general offices.114 On its website, OCR indicates that between April 2003 and August 2019 it received more than 216,000 HIPAA complaints, and it initiated 979 compliance reviews on its 114  45 C.F.R. §160.306 (2018); 42 U.S.C.A. § 1320d-5(d) (2010).

Privacy and Integrity of Medical Information   443 own.115 Of these, OCR investigated and resolved approximately 27,225 cases, offering technical assistance and requiring privacy practice changes or other corrective action. In addition, OCR reached settlements or imposed civil money penalties in 65 cases, recovering $102,681,582. In the vast majority of cases, OCR found no HIPAA violation. HIPAA authorizes the government to punish violators with harsh fines and imprisonment.116 However, robust enforcement depends on adequate staffing. In an era in which many are hostile to “big government,” the federal workforce and its resources will likely shrink rather than grow. In the absence of a threat of private litigation, the HIPAA regulations’ efficacy could be compromised. Some covered entities may calculate that privacy violations are unlikely to be detected and severely punished and thus may be more lax about data security and disclosure than full compliance would require.

5.1.2  State Law Unlike HIPAA, state common law enables patients to sue healthcare providers for ­privacy breaches. One tort theory is breach of confidentiality. The elements of this cause of action are (1) the existence of a doctor–patient relationship and (2) a physician’s or medical ­entity’s disclosure to a third party of confidential information that was gained through this relationship.117 For example, in the case of Alberts v. Devine, the court asserted: “We hold today that a duty of confidentiality arises from the physician–patient relationship and that a violation of that duty, resulting in damages, gives rise to an action sounding in tort against the physician.”118 Courts have found that the right of confidentiality is rooted in a variety of sources, including privilege statutes protecting physician– patient communications, licensing statutes prohibiting the disclosure of patient information without authorization, and medical ethics principles articulated in the Hippocratic Oath.119 A second tort theory that is available to plaintiffs in limited circumstances is invasion of privacy. Under the common law, the right to privacy can be invaded by “unreasonable publicity given to the other’s private life.”120 The tort of invasion of privacy consists of four elements: (a) public disclosure (b) of a private fact (c) that would be objectionable and offensive to a reasonable person and (d) that is not of legitimate public concern.121 Courts have also granted relief to those harmed by unauthorized data releases under a number of other theories. These include breach of trust, breach of implied contract, 115  US Department of Health & Human Services, Enforcement Highlights, https://www.hhs.gov/ hipaa/for-professionals/compliance-enforcement/data/enforcement-highlights/index.html (last reviewed on November 9, 2017). 116  42 U.S.C.A. §§ 1320d-5–1320d-6 (2010). 117  Neil M. Richards & Daniel J. Solove, Privacy’s Other Path: Recovering the Law of Confidentiality, 96 Geo. L.J. 123, 157 (2007). 118  Alberts v. Devine, 479 N.E.2d 113, 120 (Mass. 1985) 119 Peter A. Winn, Confidentiality in Cyberspace: The HIPAA Privacy Rules and the Common Law, 33 Rutgers L.J. 617, 654–655 (2002). 120  Newkirk v. GKN Armstrong Wheels, Inc., 168 F.Supp.3d 1174, 1201 (N.D. Iowa 2016). 121  See Diaz v. Oakland Tribune, 188 Cal. Rptr. 762, 766 (Cal. Ct. App. 1983).

444   Sharona Hoffman and Jean Herveg defamation, and negligence.122 Some state laws provide statutory causes of action as well.123

5.1.3  EU Law To ensure data protection effectiveness, EU law requires Member States to have s­ pecific data protection authorities, as well as to provide for specific mechanisms and remedies. 5.1.3.1  Supervisory Authorities At the level of the Member States, each Member State must provide for one or more independent public authorities to be responsible for monitoring the application of the GDPR in order to protect the fundamental rights and freedoms of natural persons in relation to processing and to facilitate the free flow of personal data within the EU.124 Each supervisory authority must act with complete independence in performing its tasks and exercising its powers.125 The data controller, the processor, and, where applicable, their representatives, must cooperate on request with the supervisory authority.126 At the level of the EU, the European Data Protection Board replaces the Working Party on the protection of individuals with regard to the processing of personal data.127 The Board is composed of the head of one supervisory authority of each Member State and of the European Data Protection Supervisor, or their respective representatives. The Board must act independently when performing its tasks or exercising its powers. In the performance of its tasks or the exercise of its powers, the Board will neither seek nor take instructions from anybody. The Board will draw up an annual report regarding the protection of natural persons with regard to processing in the EU and, where relevant, in third countries and international organizations. The European data protection supervisor will provide the secretariat of the Board.128 122  Confidentiality and Privacy of Personal Data, in Institute of Medicine, Health Data in the Information Age: Use, Disclosure, and Privacy 149 (Molla S. Donaldson & Kathleen N. Lohr, Eds. 1994). 123  Id.; Sharona Hoffman & Andy Podgurski, E-Health Hazards: Provider Liability and Electronic Health Record Systems, 24 Berkeley Tech. L. J. 1523, 1558–1560 (2009). 124  See Article 51 of the Regulation on the principle of independence and Article 55 on the issue of the competence of the supervisory authority (cf. Article 4.22 of the Regulation for the definition of the supervisory authority concerned). It is expressly provided that the supervisory authorities are not competent to review the processing operations carried out by the courts in the exercise of their judicial function (Article 55.3 of the Regulation). The duties and powers of the supervisory authorities are detailed in Articles 57 and 58 of the Regulation. See Article 29 Data Protection Working Party Guidelines for identifying a controller or processor’s lead supervisory authority WP 244 rev.01 5 April 2017. 125  See Article 52 of the Regulation. 126  Article 31 of the Regulation. The application of an approved Code of Conduct or an approved certification mechanism may serve as an element to demonstrate compliance with date processing security requirements. 127  See Article 68 of the Regulation. Article 70 lists its missions. 128  The European Data Protection Supervisor is also the supervisory authority for EUROPOL.

Privacy and Integrity of Medical Information   445 5.1.3.2  Data Subject’s Remedies Right to lodge a complaint with a supervisory authority. Every data subject has the right to lodge a complaint with a supervisory authority, in particular in the Member State of his or her habitual residence, place of work, or place of the alleged infringement if the data subject considers that the processing of personal data relating to him or her infringes the GDPR.129 Right to an effective judicial remedy against a supervisory authority. Each natural or legal person shall have the right to an effective judicial remedy against a legally binding decision of a supervisory authority concerning them.130 Without prejudice to any other administrative or nonjudicial remedy, each data subject shall have the right to an effective judicial remedy where the supervisory authority which is competent does not handle a complaint or does not inform the data subject within 3 months on the progress or outcome of the complaint.131 Right to an effective judicial remedy against a controller or processor. Each data subject shall have the right to an effective judicial remedy where he or she considers that his or her rights under the GDPR have been infringed as a result of the processing of his or her personal data in noncompliance with the GDPR.132 Right to compensation and liability. Any person who has suffered material or nonmaterial damage as a result of an infringement of the GDPR has the right to receive compensation from the controller or processor for the damage suffered.133 Any data 129  It is not easy to argue that this right exists in the case of a breach of a rule which would be imposed by a Member State within the scope of the discretion which would be accorded to the state for the implementation of a particular provision of the Regulation. See Article 80 on the question of the representation of data subjects. 130  Directive 95/46/EC already provided that Decisions by the supervisory authority which give rise to complaints may be appealed against through the courts (Article 28.3, in fine). See Article 78.1 of the Regulation. Proceedings against a supervisory authority must be brought before the courts of the Member State where the supervisory authority is established. Where proceedings are brought against a decision of a supervisory authority which was preceded by an opinion or a decision of the Board in the consistency mechanism, the supervisory authority shall forward that opinion or decision to the court (Article 78.4 of the Regulation). 131  See Article 78.2 of the Regulation. Proceedings against a supervisory authority must be brought before the courts of the Member State where the supervisory authority is established. Where proceedings are brought against a decision of a supervisory authority which was preceded by an opinion or a decision of the Board in the consistency mechanism, the supervisory authority shall forward that opinion or decision to the court (Article 78.4 of the Regulation). 132  See Article 79.1 of the Regulation. Proceedings against a controller or a processor must be brought before the courts of the Member State where the controller or processor has an establishment. Alternatively, such proceedings may be brought before the courts of the Member State where the data subject has his or her habitual residence, unless the controller or processor is a public authority of a Member State acting in the exercise of its public powers. 133  Court proceedings for exercising the right to receive compensation must be brought before the courts competent under the law of the Member State where the data controller or processor has an establishment. Alternatively, such proceedings may be brought before the courts of the Member State where the data subject has his or her habitual residence, unless the controller or processor is a public authority of a Member State acting in the exercise of its public powers.

446   Sharona Hoffman and Jean Herveg controller involved in processing is liable for the damage caused by processing which infringes the GDPR. A processor is liable for the damage caused by processing only where it has not complied with obligations of the GDPR specifically directed to processors or where it has acted outside of or contrary to lawful instructions from the data controller. A data controller or processor is exempt from liability if it proves that it is not in any way responsible for the event giving rise to the damage. Where more than one data controller or processor, or both a data controller and a processor, are involved in the same processing and where they are responsible for any damage caused by processing, each data controller or processor is liable for the entire damage in order to ensure effective compensation of the data subject.134 5.1.3.3  Administrative Obligations of Data Controllers Notification of personal data breach to supervisory authorities and data subjects. In the case of a breach of security leading to the accidental or unlawful destruction, loss, alteration, unauthorized disclosure of, or access to personal data transmitted, stored, or otherwise processed (known as “personal data breach”135), the data controller must without undue delay and, where feasible, not later than 72 hours after having become aware of it,136 disclose the personal data breach to the competent supervisory authority.137 The data controller is exempted from this duty when the personal data breach is unlikely to result in a risk to the rights and freedoms of natural persons. But, in any case, the data controller must document any personal data breaches, including the facts relating to the personal data breach, its effects, and the remedial action taken. That documentation must enable the supervisory authority to verify the compliance with the obligations applicable to the data controller.

134  See Article 82 of the Regulation. Where a controller or processor has, in accordance with paragraph 4, paid full compensation for the damage suffered, that controller or processor shall be entitled to claim back from the other controllers or processors involved in the same processing that part of the compensation corresponding to their part of responsibility for the damage. 135  Article 4.12 of the Regulation. See Article 29 Data Protection Working Party Working Document 01/2011 on the current EU personal data breach framework and recommendations for future policy developments WP 184 5 April 2011, Opinion 03/2014 on Personal Data Breach Notification WP 213 25 March 2014, and Guidelines on Personal data breach notification under Regulation 2016/679 WP 250 rev.01 6 February 2018. 136  See Article 33 of the Regulation. Where the notification to the supervisory authority is not made within 72 hours, it has to be accompanied by reasons for the delay. Where, and in so far as it is not possible to provide the information at the same time, the information may be provided in phases without undue further delay. 137  The notification must at least describe the nature of the personal data breach including where possible, the categories and approximate number of data subjects concerned, and the categories and approximate number of personal data records concerned; communicate the name and contact details of the data protection officer or other contact point where more information can be obtained; describe the likely consequences of the personal data breach; and describe the measures taken or proposed to be taken by the controller to address the personal data breach, including, where appropriate, measures to mitigate its possible adverse effects.

Privacy and Integrity of Medical Information   447 Similarly, the processor must notify the data controller without undue delay after becoming aware of a personal data breach. It must be assumed that it is also required to document any data breaches even if this is not expressly foreseen in the Regulation. Asymmetrically in relation to the obligation to notify the supervisory authority, the data controller must only disclose the personal data breach to the data subject if the breach is likely to result in a high risk to the rights and freedoms of natural persons. This could be the case in the situation in which a hacker publishes online data from a hospital and these data contain information about the health condition of patients. The communication must be done without undue delay. The communication to the data subject must describe in clear and plain language the nature of the personal data breach including, where possible, the categories and approximate number of data subjects concerned and the categories and approximate number of personal data records concerned. It must also contain the name and contact details of the data protection officer or any other contact point where more information can be obtained, the likely consequences of the personal data breach, and the measures taken or proposed to be taken by the controller to address the personal data breach including, where appropriate, measures to mitigate its possible adverse effects. However, even in the event of a high risk to rights and freedoms, this communication is not always required (e.g., if the hospital has taken all the appropriate measures to mitigate the risks for the data subject). Furthermore, if the data controller has not already communicated the data breach to the data subject, the supervisory authority may, after examining whether this data breach is likely to result in a high risk, require the data controller to communicate or decide that the controller is in one of the situations in which he is exempted from doing so.138 Privacy impact assessment. Prior to processing, the data controller must carry out an assessment of the impact of the envisaged processing operations on the protection of personal data139 where a type of processing, particularly when using new technologies and taking into account the nature, scope, context, and purposes of the processing, is likely to result in a high risk to the rights and freedoms of natural persons. The controller will seek the advice of the data protection officer, where designated, when carrying out a data protection impact assessment.140 138  Article 34 of the Regulation. 139  See: D. Wright & P. De Het (ed.), Privacy Impact Assessment, Law, Governance and Technology Series, volume 6 (Springer, 2012); and Article 29 Data Protection Working Party Guidelines on Data Protection Impact Assessment (DPIA) and determining whether processing is “likely to result in a high risk” for the purposes of Regulation 2016/679 WP 248 4 April 2017. 140  See Article 35 of the Regulation. A single assessment may address a set of similar processing operations that present similar high risks. The controller shall seek the advice of the data protection officer, where designated, when carrying out a data protection impact assessment (leaving open the question of the obligation to do so when the controller had no obligation [formally or in the framework of technical and organizational measures] to designate one but still did it). The supervisory authority must establish and make public a list of the kind of processing operations which are subject to the requirement for a data protection impact assessment. The supervisory authority must communicate those lists to the European Data Protection Board. The supervisory

448   Sharona Hoffman and Jean Herveg The data controller will consult the supervisory authority prior to processing where a data protection impact assessment indicates that the processing would result in a high risk in the absence of measures taken by the controller to mitigate the risk.141 If the supervisory authority is of the opinion that the processing would violate the GDPR, especially when the data controller has insufficiently identified or mitigated the risk, the supervisory authority must, within period of up to 8 weeks of receipt of the request for consultation, provide written advice to the controller. In addition where applicable to the processor, the supervisory authority may use any of its investigating powers, remedial powers, advisory powers, or any other power conferred by its national law.142

5.1.4  Administrative Fines and Penalties Depending on the circumstances of each individual case, each supervisory authority may impose effective, proportionate, and dissuasive administrative fines143 in addition to or in place of corrective measures.144 Member States must lay down the rules on other penalties applicable to infringements of the GDPR, in particular for infringements that are not subject to administrative fines. They must take all measures necessary to ensure that these penalties are authority may also establish and make public a list of the kind of processing operations for which no data protection impact assessment is required. The supervisory authority shall communicate those lists to the European Data Protection Board. Prior to the adoption of the lists, the competent supervisory authority will apply the consistency mechanism where such lists involve processing activities which are related to the offering of goods or services to data subjects or to the monitoring of their behavior in several Member States or may substantially affect the free movement of personal data within the Union. Compliance with approved codes of conduct by the relevant controllers or processors must be taken into due account in assessing the impact of the processing operations performed by such controllers or processors, in particular for the purposes of a data protection impact assessment. Where appropriate, the data controller must seek the views of data subjects or their representatives on the intended processing, without prejudice to the protection of commercial or public interests or the security of processing operations. 141  When consulting the supervisory authority pursuant to paragraph 1, the controller shall provide the supervisory authority with (Article 36.3 of the Regulation): 1. where applicable, the respective responsibilities of the controller, joint controllers, and processors involved in the processing, in particular for processing within a group of undertakings; 2. the purposes and means of the intended processing;  3. the measures and safeguards provided to protect the rights and freedoms of data subjects pursuant to this Regulation; 4. where applicable, the contact details of the data protection officer; 5. the data protection impact assessment; 6. and any other information requested by the supervisory authority. 142  See Article 58 of the Regulation. 143  On all of this and in particular the factors to be taken into account in each individual case, see Article 83 of the Regulation. 144  See the list of corrective measures in Article 58.2, a) to h), and j) of the Regulation. See also Article 29 Data Protection Working Party Guidelines on the application and setting of administrative fines for the purposes of the Regulation 2016/679 WP 253 3 October 2017. Public bodies enjoy a special regime.

Privacy and Integrity of Medical Information   449 implemented (and enforced). Such penalties must be effective, proportionate, and dissuasive.145 To date, we are witnessing an increase in the number of infringements of the GDPR investigated by supervisory authorities, including in the healthcare sector.

6  Anti-Discrimination Laws Despite the law’s best efforts to protect privacy, individuals’ health information is routinely seen by a multitude of entities and persons. Some parties may want to use medical data to their own advantage in ways that do not serve data subjects’ best interests. Thus, employers, financial institutions, marketers, and others with financial agendas might use medical data to make adverse decisions that deprive individuals of various opportunities or exploit their vulnerabilities.146 Consequently, American law includes point-ofuse legislation that prohibits particular parties from using medical data to discriminate against data subjects. In Europe, although EU law prohibits discrimination in some contexts (notably employment) on a number of forbidden grounds, health status is not one of those grounds. Thus, most relevant anti-discrimination laws are at the state level, with protection at the EU level occurring through the application of more general data protection rules.

6.1  American Law Most notably, the ADA establishes a far-reaching anti-discrimination mandate. The ADA prohibits a large variety of parties from engaging in disability-based discrimination. Title I of the statute applies to employers with 15 or more employees. Title II relates to public services, including any instrumentalities of state and local governments. Title III governs “public accommodations and services provided by private entities,” such as banks, insurance offices, private educational institutions, sales establishments, and more.147 Thus, entities that possess identifiable health information and are aware of individuals’ disabilities may not use the data for discriminatory purposes. For example, an employer may not reject a qualified applicant just because it is aware that the applicant has a history of cancer, diabetes, or epilepsy. The Rehabilitation Act of 1973 was the first federal disability discrimination law, but its reach is more limited than the ADA’s because it focuses on federal entities. It prohibits programs operated by federal agencies, programs receiving federal financial assistance, federal employers, and federal contractors from discriminating against individuals 145  See Article 84 of the Regulation. 146  Sharona Hoffman, Big Data and the Americans with Disabilities Act, 68 Hastings L. J. 777, 778 (2017). 147  42 U.S.C. §§ 12111–12189 (2010).

450   Sharona Hoffman and Jean Herveg because of their disabilities.148 The Rehabilitation Act coexists with the ADA, which does not cover the executive branch of the federal government. GINA prohibits employers and health insurers (but not others) from discriminating based on genetic information.149 Recall that employers and health insurers cannot intentionally seek genetic data, including family health histories, about individuals.150 Moreover, if they happen to possess genetic data, they must not use it to make adverse decisions regarding data subjects.151 Section 1557 of the Affordable Care Act (ACA) is another provision that bans ­disability-based discrimination. It covers any health program or activity that is funded and/or administrated by DHHS and any health insurance marketplace insurers.152 Also relevant is the Fair Housing Act. This law prohibits discrimination in the sale, rental, and financing of housing based on disability and other protected categories.153 Almost all states have also adopted statutory anti-discrimination mandates that protect individuals with disabilities.154 Many are very similar to federal law, but some offer additional protections, such as covering employers with fewer than 15 workers.

6.2  EU Law EU law prohibits discrimination on various protected grounds, including disability, in certain contexts, particularly employment.155 However, in general, when it comes to using health data to discriminate, people in Europe must look to their own country’s law for protection. Some national legislation bans or restricts the use of medical information under certain circumstances (insurance, credit, employment, school, etc.). For instance, “medical condition” is a prohibited ground of discrimination under the French Labor Code. At the EU level, the main protection against discrimination coming from the use of medical data must be found in data protection rules. First, as explained earlier, personal data must be processed lawfully and for a legitimate purpose. The latter means that no one may process data in order to unlawfully 148  29 U.S.C. §§ 791-794(a) (2010). 149  Genetic Information Non-Discrimination Act, Pub. L. No. 110–233, 122 Stat. 881 §§ 201(4) & 202(a) (2008); 29 U.S.C. § 1182 (2010); 42 U.S.C. §2000ff-1(a) (2010). Genetic information is defined as including (i) an individual’s genetic tests, (ii) the genetic tests of an individual’s family members, and (iii) the manifestation of a disease or disorder in an individual’s family members. 42 U.S.C. §§ 2000ff (4)(A) (2010). 150  29 U.S.C. § 1182 (c) & (d) (2010); 42 U.S.C. § 2000ff-1(b) (2010). 151  29 U.S.C. § 1182 (a) & (b) (2010); 42 U.S.C. § 2000ff-1(a) (2010). 152  42 U.S.C. §18116 (2010). 153  42 U.S.C. 3604 (2010). 154  National Conference of State Legislatures, State Laws on Employment-Related Discrimination, July 2015, http://www.ncsl.org/documents/employ/Discrimination-Chart-2015.pdf; Stephen A. Rosenbaum, Disability Rights and Public Accommodations: State-by-State (2011), http:// adasoutheast.org/publications/ada/public_accommodations_disability_rights_state-by-state_Final.pdf. 155  Council Directive 2000/78/EC of 27 November 2000, establishing a general framework for equal treatment in employment and occupation. Official Journal (EC) L 303/16 of 2 Dec. 2000.

Privacy and Integrity of Medical Information   451 ­ iscriminate against data subjects. Moreover, personal data must be adequate, relevant, d and limited to what is necessary in relation to the legitimate purposes for which they are processed (this refers to the principle of data minimization). And personal data must be protected against unauthorized or unlawful processing. The data controller is responsible for compliance with these obligations, which should, at least in principle, prevent discrimination against data subjects. It must be reiterated that the assessment of the legitimacy of data processing is sensitive to other aspects of the implementation of data protection, such as the level of confidentiality and security of the data processing, the level of control exercised by the national supervisory authority, the degree of necessity of the purpose pursued, and so on. It is also important to remember that Member States may maintain or introduce ­further conditions, including limitations, with regard to the processing of genetic data, biometric data, or data concerning health. Therefore, medical information privacy and  confidentiality is subject to national substantive rules, which differ across European states.

7  Contemporary Legal Challenges 7.1  Data Security As hackers and other cybercriminals become increasingly sophisticated, new data security threats arise with dizzying speed. For example, healthcare providers must now worry about “ransomware attacks” by which hackers encrypt data and demand ransom in exchange for decryption keys, thereby denying users access to their own data until they comply with payment demands. The scope of this threat became clear in the wake of the May 2017 “WannaCry” ransomware attack that crippled the British National Health Service and many other providers around the globe.156 The HIPAA Security Rule details many administrative, physical, and technical safeguards. But its recommendations are not exhaustive because its authors could not possibly anticipate every risk that would arise in the future. Moreover, the Security Rule deliberately leaves implementation mechanisms to the discretion of covered entities.157 In fact, a provision entitled “Flexibility of Approach” states: “Covered entities may use any security measures that allow the covered entity to reasonably and appropriately implement the standards and implementation specifications.”158

156  I. Glenn Cohen et al., Your Money or Your Patient’s Life? Ransomware and Electronic Health Records, 167 Annals Intern. Med. 587 (2017). 157  45 C.F.R. § 164.306 (b) (2018). 158  45 C.F.R. § 164.306(b)(1) (2018).

452   Sharona Hoffman and Jean Herveg A flexible approach reduces covered entities’ compliance burdens. Moreover, the approach recognizes the ever-changing nature of technology and its need to continuously evolve to respond to new forms of cybersecurity threats. Yet, alongside its benefits, flexibility can raise significant risks in the regulatory context. It can leave covered entities with anemic and inadequate guidance on how to comply with legal requirements. Resource-poor healthcare providers who do not have sophisticated information technology departments might be at a particular disadvantage. They would be well-advised to consult security experts and adopt best industry practices, but some may have other priorities in light of tight budgets.159 Because attackers are quick to exploit newly discovered vulnerabilities in software systems, it is critical that covered entities diligently assess and respond to these risks. It is also essential that regulators tirelessly follow cybersecurity threats and periodically reevaluate the HIPAA Security rule to determine whether and how regulations should be revised.

7.2  Data Subjects’ Rights to Their Medical Information Under the HIPAA Privacy Rule, patients have a right to inspect and obtain copies of their medical records other than psychotherapy notes. Access requests can be denied under particular circumstances, such as the existence of a reasonable likelihood that seeing the data will result in harm to the patient or another party.160 In addition, patients may ask that their health records be amended or used restrictively, though covered entities may deny the requests for legitimate reasons specified in the regulations (e.g., the records are accurate and should not be changed).161 The question of the patient’s access to medical information in Europe finds a first answer in patient’s rights legislation but also under data protection law. And where the old Data Protection Directive formally recognized three rights (right of access, right to object to data processing, and right not to be subject to individual automated decisions), the GDPR grants data subjects eight rights (right to information, right of access, right to rectification, right to erase, right to limit treatment, right to data portability, right to object to data processing, and right not to be subject to automated individual decisions).162 159  Sharona Hoffman, Electronic Health Records and Medical Big Data: Law and Policy 69–70 (Cambridge University Press 2016); Hoffman & Podgurski, supra note 21, at 350–354. 160  45 C.F.R. § 164.524 (2018) 161  45 C.F.R. §§ 164.524 and 164.526 (2018). 162  See the limits which may be imposed on these rights by Union law or by the law of the Member State to which the controller or processor is subject, by means of legislative measures, in accordance with Article 23 of the Regulation. These limits are permissible only if they respect the essence of fundamental rights and freedoms and are necessary and proportionate measures in a democratic society to guarantee one of the objectives listed in this provision. See Article 29 Data Protection Working Party Guidelines on Transparency under Regulation 2016/679 WP 260 rev.01 11 April 2018 and Guidelines on Automated individual decision-making and Profiling for the purposes of Regulation 2016/679 WP 251 rev.01 6 February 2018.

Privacy and Integrity of Medical Information   453 In particular, the right to data portability163 means that, where the data are processed on the basis of the data subject’s consent or a contract and by automated means, the data subject has the right to request and receive in a structured, commonly used, and machine-readable format the data he or she has provided to the data controller. The data subject is then entitled to forward these data to another data controller. The data subject may also ask the first controller to send them directly to another data controller if technically feasible.164 This right inevitably brings to mind the situation in which the patient’s medical record is shared between healthcare professionals to ensure the continuity of care. The implementation of this newly formalized right may therefore not be a problem in the health sector as long as it is extended to data not provided by the patient.165 That being said, the real challenge is to know how these rights will really and effectively prosper in light of the debates around cloud computing services, big data, and mobile applications and whether this formal increase in the number of rights will improve data protection and benefits for data subjects. Doubt is permitted. There should be more information for the public on the way mobile applications work and more transparency about algorithmic governance in healthcare contexts. For example, public authorities and bodies should disseminate more information about the numerous actors and data flows occurring behind the screens of smartphones that gather health (and other) data.166

7.3  The Proliferation of Health-Related Data It is a mistake to think that personal health information is restricted to patients’ electronic health records. A large number of private and public sector entities are creating data resources for “secondary use” (i.e., use for non-treatment purposes). Such purposes

163  Article 29 Data Protection Working Party Guidelines on the right to data portability WP 242 rev.01 5 April 2017. 164  See Article 20 of the Regulation. This right is without prejudice to the right to erasure or to be forgotten. That right does not apply to processing necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller. In addition, it cannot adversely affect the rights and freedoms of others. 165  See, e.g., Article 4.2 (f) of Directive 2011/24/EC of the European Parliament and of the Council of 9 March 2011 on the application of patients’ rights in cross-border healthcare (see the consolidated text), which provides that in order to ensure continuity of care, patients who have received treatment are entitled to a written or electronic medical record of such treatment and access to at least a copy of this record in conformity with and subject to national measures implementing Union provisions on the protection of personal data, in particular Directives 95/46/EC and 2002/58/EC. 166  Recently, the British Medical Journal has published a very interesting study on the data flows behind mobile health applications (Q. Grundy, K. Chiu, F. Held, A. Continella, L. Bero, & R. Holz. Data sharing practices of medicines related apps and the mobile ecosystem: traffic, content, and network analysis, 364 BMJ 1920 (2019). These kind of studies should be more encouraged, and they should receive more coverage from the news media.

454   Sharona Hoffman and Jean Herveg include research, public health initiatives, healthcare institutions’ quality assessment and improvement efforts, and more. Health information resources are often termed “big data,” which means large collections of data characterized by their high volume, variety, and velocity (the speed with which they are produced and grow).167 National, regional, and local government agencies, as well as private enterprises, have launched big data initiatives. The National Institutes of Health describe their All of Us program as follows: “The All of Us Research Program is a historic effort to gather data from one million or more people living in the United States to accelerate research and improve health. By taking into account individual differences in lifestyle, environment, and biology, researchers will uncover paths toward delivering precision medicine.”168 A federal-state-industry partnership sponsored by the Agency for Healthcare Research and Quality has developed the Healthcare Cost and Utilization Project, which created the State Inpatient Databases (SID). The SID contain hospital discharge records that are available for purchase in some states.169 Another example, this from the private sector, is IBM Watson Health. This project “brings together individual clinical research and social data from a diverse range of health sources creating a secure cloud-based data sharing hub.”170 Big data collections often consist of de-identified EHRs. Under the HIPAA Privacy Rule, covered entities can disclose fully de-identified information without patient consent, and information distributed by noncovered entities is not subject to any disclosure constraints. Thus, many patients will never discover that their records have been incorporated into a database and are being used by third parties. As noted earlier, the risk of re-identification can never be completely eliminated no matter how thoroughly identifiers are expunged. More startlingly, when some identifiers are retained (e.g., by entities not covered by HIPAA), the risk of re-identification increases dramatically. In fact, it is estimated that up to 87% of the US population could be uniquely identified based on three items alone: gender, zip code, and date of birth.171 In the pre-HIPAA era, Latanya Sweeney, now a Harvard professor, became famous for having identified Massachusetts Governor

167  Gil Press, 12 Big Data Definitions: What’s Yours?, Forbes, Sept. 3, 2014, https://www.forbes.com/ sites/gilpress/2014/09/03/12-big-data-definitions-whats-yours/#6301d92913ae. 168  National Institutes of Health, All of Us Research Program (last visited Dec. 21, 2017), https:// allofus.nih.gov/. 169  Healthcare Cost and Utilization Project, Overview of the State Inpatient Databases (SID) (last modified Apr. 18, 2017), http://www.hcup-us.ahrq.gov/sidoverview.jsp. 170  Margaret Bays, IBM Watson Health: How Does it Work? IBM Developer Works (June 15, 2017), https://www.ibm.com/developerworks/community/blogs/e3ec7365-1b09-44f2-906f-19826275860f/ entry/IBM_Watson_Health_How_Does_it_work?lang=en. 171  Hoffman & Podgurski, supra note 111, at 105; Latanya Sweeney, Simple Demographics Often Identify People Uniquely, Carnegie Mellon University, Data Privacy Working Paper 3. Pittsburgh (2000), https://dataprivacylab.org/projects/identifiability/paper1.pdf.

Privacy and Integrity of Medical Information   455 William Weld’s records based on “anonymized” hospital discharge data that she matched to voter registration information when she was a graduate student in 1996.172 Health information databases are not the only source of data that could be available to others without patients’ knowledge. There is an abundance of what one scholar calls “medically inflected data.”173 Data miners seeking individuals’ health data have no shortage of sources available to them. Supermarket loyalty cards, credit card transactions, web browsing histories, social media interactions, phone call records, and other data fall outside the jurisdiction of the HIPAA Privacy Rule, but they can reveal a great deal about individuals’ health. For example, experts analyzed the Facebook “likes” of approximately 60,000 volunteers and were able to discern “sexual orientation, ethnicity, religious and political views, personality traits, intelligence, happiness, use of addictive substances, parental separation, age, and gender.”174 The risk to privacy is also demonstrated by a 2012 Forbes magazine article entitled “How Target Figured Out a Teen Girl Was Pregnant Before Her Father Did.”175 Target was able to determine that the teenager was pregnant based on her purchases and began sending her coupons for baby goods. Who might be interested in information related to individuals’ health? The list includes employers, lenders, marketers, advertisers, and anyone else with a stake in a person’s economic future. To illustrate, employers seek workers who will be healthy, productive, and not generate high health insurance costs. Lenders want borrowers who are healthy enough to work and pay back their loans. Marketers wish to tailor their advertising to fit customers’ needs and desires. Consequently, a growing industry of data miners and data brokers offers a wealth of personal information for sale.176 In the contemporary world virtually no one can escape the disclosure of personal details. Health data are no exception. Legislators must ensure that HIPAA and other privacy laws do not ignore emerging data trends and are updated to provide appropriate privacy protections.177

172  Kathleen Benitez & Bradley Malin, Evaluating Re-identification Risks with Respect to the HIPAA Privacy Rule, 17 J. Am. Med. Informatics Ass’n 169, 169 (2010). 173 Nicolas P. Terry, Big Data Proxies and Health Privacy Exceptionalism, 24 Health Matrix 65, 77 (2014). 174  Michal Kosinskia,1, David Stillwella, & Thore Graepel, Private Traits and Attributes Are Predictable from Digital Records of Human Behavior, 110 PNAS 5733, 5733 (2013), http://www-psych. stanford.edu/~knutson/bad/kosinski13.pdf. 175  Kashmir Hill, How Target Figured Out a Teen Girl Was Pregnant Before Her Father Did, Forbes (Feb. 16, 2012), https://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured-out-a-teengirl-was-pregnant-before-her-father-did/#7c4f90266668 (discussing Target’s practice of data-mining its customers’ purchasing records in order “to figure out what you like, what you need, and which coupons are most likely to make you happy”). 176  Sharona Hoffman, Citizen Science: The Law and Ethics of Public Access to Medical Big Data, 30 Berkeley Tech. L. J. 1741, 1773–1780 (2016). 177  For recommendations see Hoffman, supra note 4, at 73-79 and 148-51.

456   Sharona Hoffman and Jean Herveg

7.4  Predictive Health Information Thus far, the chapter has largely addressed data that reveal something about individuals’ current health status. However, certain types of data may be indicators of future illnesses, even for people who appear to be perfectly healthy at the present time. Researchers are aggressively searching for predictive health information. For example, a study published in 2014 suggested that physicians could use blood tests to predict imminent dementia.178 Investigators found that people who suffered from cognitive impairment when their blood was drawn or within a few years of the test had lower levels of 10 phospholipids. Data algorithms can also forecast certain diseases before their symptoms are apparent. Researchers have used algorithms to analyze EHR and insurance claims data to predict clinical depression, diabetes, and heart failure.179 Data that are not explicitly medical in nature can be illuminating as well. Smoking and childlessness are known to generate health risks.180 Researchers have found that people who shop in bicycle stores are generally in good health, and those who vote in midterm elections are healthier than those who do not.181 By contrast, individuals with low credit scores may not have the financial means to fill prescriptions and obtain good medical care and thus are vulnerable to poor health. One can learn a great deal about people’s health from their smoking status, parental status, purchases, voting patterns, and credit scores. Yet such information remains outside the scope of the HIPAA Privacy and Security Rules if it is not contained in a medical record. Moreover, Americans are not protected against discrimination based on predictions of illness in later years. The ADA covers only individuals who (1) have a physical or mental impairment that substantially limits one or more major life activities, (2) have a record of such an impairment, or (3) are regarded as currently having such an impairment.182 It therefore does not reach healthy individuals who are subject to discrimination because they are suspected of being at risk of future health problems. Consequently, under the ADA, nothing would stop employers, lenders, educational institutions, or 178  Alison Abbott, Biomarkers Could Predict Alzheimer’s before It Starts, Nature, March 9, 2014, http://www.nature.com/news/biomarkers-could-predict-alzheimer-s-before-it-starts-1.14834. 179  Mohana Ravindranath, IBM Used Predictive Analytics to Find Patients at Risk of Heart Failure, Wash. Post, February 20, 2014, https://www.washingtonpost.com/business/on-it/ibm-used-predictiveanalytics-to-find-patients-at-risk-of-heart-failure/2014/02/20/9b0ddb3c-9a47-11e3-b88d-f36c07223d88_ story.html; Arthur Allen, Big Brother Is Watching Your Waist, Politico, July 21, 2014, http://www. politico.com/story/2014/07/data-mining-health-care-109153; Susan H. Babey et al., Prediabetes in California: Nearly Half of California Adults on Path to Diabetes, UCLA Center for Health Policy Research Health Policy Brief (March 2016), http://healthpolicy.ucla.edu/publications/Documents/ PDF/2016/prediabetes-brief-mar2016.pdf. 180  Sharona Hoffman, Big Data and the Americans with Disabilities Act, 68 Hastings L. J. 777, 784–785 (2017). 181  Rachel Emma Silverman, Bosses Tap Outside Firms to Predict Which Workers Might Get Sick, Wall St. J., February 17, 2016, https://thebenefitblog.com/2016/02/19/ bosses-tap-outside-firms-to-predict-which-workers-might-get-sick/. 182  42 U.S.C. § 12102 (2010).

Privacy and Integrity of Medical Information   457 others from rejecting applicants because of anticipated medical conditions. Almost no other anti-discrimination law (with the exception of GINA, which governs only genetic information) is broad enough to prohibit discrimination based solely on predictive data. As prognostic abilities improve, third parties may increasingly seek predictive data from data brokers. They will be valuable for anyone who has a stake in others’ health as an employer, lender, advertiser, etc. Once the information is in such parties’ possession, they are free to use it as they see fit. This includes making adverse decisions about individuals in order to avoid business risks and save costs. Consequently, the privacy laws are not the only legislation that may need to be amended in order to be aligned with current technological and medical capabilities. The ADA and other anti-discrimination laws, which focus only on current disabilities, also require revision. It is inevitable that some personal health-related information will fall into the hands of others without patient consent. Patients would be well-served by protections that limit the uses to which such information can be put. American law should uniformly prohibit discrimination based on predictive health information. By contrast, with sectoral or point-of-use legislation, the GDPR provides a comprehensive framework regulating the processing of all personal data. However, there is room for some national legislation in addition to the GDPR rules, even if the limits of Member States freedom are not always straightforward in the matter of data protection. It is also important to recall that the GDPR rules must be analyzed, implemented, and interpreted in light of Convention no. 108 and all the sectorial recommendations adopted by the Council of Europe in the matter of data protection.

8 Conclusion With the enactment of the HIPAA Privacy and Security Rule early in the 21st century, American law took a leap forward in the area of medical privacy. However, contemporary privacy and anti-discrimination laws still contain significant gaps and limitations. Much regulatory work remains to be done to ensure that Americans enjoy comprehensive legal protections. In Europe, the GDPR has consolidated the previous applicable rules from Directive 95/46/EC. Furthermore, it has expanded and developed certain requirements, and it has also created some new obligations for data controllers and data processors. But the GDPR does not provide a specific set of rules for the processing of personal data concerning health, except the rules for lifting the ban on their processing (Article 9 of the GDPR). The consequence is that processing of health data is mainly subject to the general regime put into place by the GDPR, with some special rules about when it is lawful to process health data. In fact, the real challenge lies more in making the GDPR an effective tool for protection of citizens’ rights when it comes to the processing of personal data related to them. Today, this challenge should focus on how to inform data subjects about their rights and on the way new information and communication technologies

458   Sharona Hoffman and Jean Herveg work, including new actors in data flows in eHealth and mHealth. This is all the more true when considering mobile health applications and big data in the context of scientific and medical research. As in the United States, there is more regulatory work to be done.

Author Note This work has been done with financial support from the EU’s Horizon 2020 research and innovation program under Grant Agreement no. 730953 (Inspex) and in part by the Swiss Secretariat for Education, Research, and Innovation (SERI) under Grant no. 16.0136 730953. This contribution only reflects the author’s view and does not engage the Commission. It is partially based on a previous publication: J. Herveg, “Data protection and patient mobility in Europe,” in A. Den Exter (ed.), Cross-border healthcare and European Union law, Erasmus University Press, 2017, pp. 209–231.

chapter 23

Patien t Au tonom y, Ca pacit y, a n d Consen t Children Jessica Berg and Emma Cave

1 Introduction Not all children have the necessary capacities to consent to medical treatment, and even those who do may be protected by laws designed to protect their best interests. Parents may also claim a right to information and decision-making powers with respect to their children, and these, too, are limited to varying degrees by children’s interests and the state interest in sustaining life. Ideally, decisions are made for and with children on the basis of a partnership model between doctors and families taking into account the capacities of the child, the interests of the parents, and the interests of the state. Sometimes the partnership model fails to reach a mutually acceptable solution, and mediation or judicial intervention is required. A case in point is the UK 2017 case of Charlie Gard.1 Charlie was born in 2016 with a rare mitochondrial disease that has no proven treatment and is usually lethal in infancy. He was admitted to Great Ormond Street Hospital (GOSH). His parents, Chris Gard and Connie Yates, wanted to take Charlie to the United States to access nucleoside therapy, which they argued had a theoretical, though unproven, chance of improving Charlie’s health. They raised £1.3 million through a crowdfunding campaign and located a doctor in the United States willing to provide the therapy. However, in light of evidence

1  E Cave, E Nottingham ‘Who Knows Best (Interests)? The Case of Charlie Gard’ (2017) 26(3) Med Law R 500.

460   Jessica Berg and Emma Cave that Charlie’s condition had declined significantly,2 GOSH decided that it was contrary to Charlie’s best interests to make the referral. The case came before the High Court, where it was held to be in Charlie’s best interests for artificial ventilation to be withdrawn and palliative care to be administered.3 The decision was upheld on appeal by the Court of Appeal, Supreme Court, and European Court of Human Rights (ECtHR),4 notwithstanding interventions from commentators across the globe, including the Pope, US President Donald Trump, and the US Congress. The Charlie Gard case is an extreme example of the potential conflict between parents and the state—here with respect to who gets to determine an infant’s best interests. The issues do not resolve simply because the child is older, although cases involving children with greater levels of capacity raise the potential for additional conflicts between the child’s wishes and the parents and/or the state. And even the concept of capacity (and developing autonomy) can be complicated when referring to children. A narrow view of autonomy that focuses on rationality and self-determination seems the very antithesis of ill children, who may be characterised as vulnerable and de­pend­ ent. And yet the effect of illness on children’s rationality and self-determination is also apparent in adults, and child welfare, like that of adults’, is enhanced by participation and control. Moreover, autonomy develops as children develop: encouraging (or even requiring) participation in decision-making can facilitate the development of fully autonomous individuals. The law steers a careful and often uncertain course between protecting the child’s future autonomy by escalating their clinical interests and respecting their current views and wishes. Internationally, the age of full majority varies from 14 to 21. But many countries confer upon children the right to make certain treatment decisions before that point. The following sections will first provide a general overview of children’s rights with respect to making medical decisions in both the United States and Europe. We then discuss the best interests standard (which is usually applied in cases of minors) and then how to consider capacity in the context of children. In our discussions of European approaches, we cover relevant international and regional human rights law. As space does not allow a comprehensive account of the laws in each European country, we focus on the jurisdiction of England and Wales in some cases, by way of example. Similarly, we provide a general overview of US state approaches and federal law, but we do not provide a detailed multistate analysis. We conclude by noting some new areas of medical decision-making which challenge the traditional models.

2 See Great Ormond Street Hospital’s Position Statement Issued at High Court on 13 July 2017, FD17P00103, http://www.gosh.nhs.uk/search?s=charlie%20gard; accessed 1 June 2019). 3  Great Ormond Street Hospital v Yates, Gard and Gard [2017] EWHC 972 (Fam), [27]. 4  In the matter of Charles Gard [2017] EWCA Civ 410; Supreme Court decision declining permission to appeal, available at https://www.supremecourt.uk/news/permission-to-appeal-hearing-in-thematter-of-charlie-gard.html; Gard and Others v the United Kingdom—39793/17; Decision [2017] ECHR 605 (27 June 2017).

Patient Autonomy, Capacity, and Consent: Children   461

2  Overview of Children’s Rights in the United States Children’s rights in the United States are governed both by state laws and the federal Constitution.5 According to Justice Blackmun in a 1976 Supreme Court decision, ‘[c]onstitutional rights do not mature and come into being magically only when one attains the state-defined age of majority. Minors, as well as adults, are protected by the Constitution and possess constitutional rights’.6 Both the Bill of Rights (the first 10 amendments to the Constitution) and the Fourteenth Amendment apply to children,7 albeit in different ways than they might to adults. These differences arise in part because the interests of parents (such as the interest in raising their children in accord with a particular religious tradition) and the interests of the state (such as maintaining life or in having an educated populace) must also be considered alongside the rights of children and in part because children are not considered fully autonomous individuals but rather are developing autonomy. Medical decision-making involving children raises all of these issues concerning developing autonomy and relative weight of parental and state interests. The US Constitutional right to control what happens to one’s body is drawn from the concept of autonomy. To the extent that a child has the capacity to exercise autonomy in a particular context, one might assume that she has a constitutional right to control what happens to her body. But the US Constitution also recognizes the fundamental right of parents to determine how to raise their children, including decisions about education, religion,8 and medical care. In Parham v J.R., the Supreme Court explicitly acknowledged parental rights to make health decisions for their children, stressing that the presumption is that the parents act in their children’s best interest.9 State interference with parental decision-making is only allowed under a narrow set of circumstances, and parents have a ‘fundamental liberty interest’ in determining the ‘care, custody, and control of their children’.10 In the typical medical decision-making context, the physician recommends treatment and parents consent on their child’s behalf, both because the child lacks decision-making capacity and the parents are viewed as the best representatives of the child’s interests. Things become more complicated when parents refuse treatment and therefore may not be promoting their child’s health, or, as in the Gard care, when parents request treatment 5  See Guttmacher Institute, ‘An Overview of Minors’ Consent Law’, https://www.guttmacher.org/ state-policy/explore/overview-minors-consent-law; accessed 1 June 2019. 6  Planned Parenthood of Central Missouri v. Danforth, 428 U.S 52 (1976). 7  In re Gault, 387 U.S. 1, 13 (1967). 8  Meyer v. Nebraska 262 US 390 (1923); Prince v. Massachusetts 321 US 158 (1944), Wisconsin v. Yoder 406 US 205 (1972). 9  Parham v. J.R., 422 US 584 (1979). 10  Troxel v. Granville, 530 US 57 (2000).

462   Jessica Berg and Emma Cave that does not seem to serve the child’s interests. There also is added complexity when mature children object to treatment that seems desirable in terms of their best interests. When parents refuse recommended treatment, their decisions generally are overruled only if the lack of treatment has the potential to seriously compromise their children’s health, especially if not treating would deprive the children of the right to make the decisions themselves at the age of majority (that is, when the result would be the child’s death before the age of majority). The most common of these cases involve parents who refuse blood transfusions for their children on the basis of religious principles. Courts are fairly uniform in overruling these parental refusals, holding that the state’s interest in maintaining a child’s health outweighs the parental interests in making decisions relating to religion.11 More difficult are cases where the decisions in question are not life or death, or which do not involve serious disability, but even in these cases the trend seems to be towards treatment in accord with medical recommendations, at least when the parent(s)’ reason for refusal is religiously based.12 Still, religiously based refusals often are permitted. This doesn’t mean that only religious refusals are tolerated. Parental refusals of care, even those against majority medical recommendations, are allowed in a wide array of situations that do not involve immediately life-threatening situations. For example, in some US states parents are allowed to make decisions to refuse infant and child vaccinations for philosophical reasons.13 The conflict between parental interests and child interests is complex. Parents do not have an absolute right to play a role in all medical decisions for their children. For example, in the context of abortion decisions, the minor child’s right to make reproductive decisions is strong enough to overcome parental rights to participate in the decision, thus requiring at least a judicial bypass mechanism as an alternative to requiring parental consent.14 In other cases, the parental right to make decisions regarding significant medical care has been balanced against the child’s right to live to become a fully autonomous person who can then make his or her own decisions.15 In some extreme cases, a parent’s refusal to consent to needed medical care has been viewed as child abuse or neglect. For example, in 2017, a Pennsylvania couple were charged with involuntary manslaughter and child endangerment when their two-year-old daughter died after they did not get her treated for pneumonia because of their belief in faith healing.16 By contrast, parental decisions 11  A Lederman, ‘Understanding Faith: When Religious Parents Decline Conventional Medical Treatment for Their Children’ (1995) 45 Case Western Res Univ L Rev 891. 12  D Ridgway, ‘Court-Mediated Disputes Between Physicians and Families over the Medical Care of Children’ (2004) 158 Arch Pediatric Adolescent Med 891. 13  See http://www.ncsl.org/research/health/school-immunization-exemption-state-laws.aspx; accessed 1 June 2019. 14  Planned Parenthood of Central Missouri v. Danforth 428 US 52 (1976), Belloitti v. Baird 443 US 622 (1979). 15  J Feinberg, ‘A Child’s Right to an Open Future’ in Whose Child? Children’s Rights, Parental Authority and State Power, W Aiken and H LaFollette (eds) (Rowman and Littlefield 1980). 16  Associated Press (2018) https://www.nbcphiladelphia.com/news/local/Faith-Healing-Trial-GirlDeath-Trial-Pennsylvania-477837243.html; accessed 31 July 2019.

Patient Autonomy, Capacity, and Consent: Children   463 to consent to care in accord with medical recommendations, but over the objections of the minor child, are more likely to be viewed favorably by courts (if the case even makes it into the courtroom—most of these are resolved in the clinical setting). There are nine categories of state laws that specifically give a minor the right to make certain kinds of medical decisions without parental involvement—contraceptive services, treatment for sexually transmitted infections (STIs), mental health/substance abuse treatment, sexual assault or rape treatment, abortion, prenatal care, adoption of the minor’s child, medical care of the minor’s child, and donation of blood or bone ­marrow. Not all state laws address each of these situations the same way: some allow decision-making in a few of these contexts but not others, and some have different rules related to the age at which a minor can consent in a particular context.17 The rationale for allowing some of these kinds of decisions on a broad scale is based less on the idea that minors have capacity than on the concern that requiring parental consent will prevent minors from even seeking treatment (e.g., for contraception or STIs). In other words, these are cases in which parental involvement would interfere with the child’s best interests. Other areas, such as the authority to make decisions related to a minor’s own child may be in part a recognition of parental rights—in this case the minor parent’s rights. All states recognize ‘emancipated minors’ (minors who are no longer being supported by their parents and thus no longer subject to parental authority), although the circumstances which will allow a minor to get this designation may vary by state,18 and, as a practical matter, the designation is rare. Emancipated minors are treated as adults under the law. A handful of states also recognize ‘mature minors’ (who are considered mature enough to make their own decision in a specific situation)19 but are not emancipated and are not viewed as legal adults. The ‘mature minor’ designation remains controversial and is not widely applied, although the concept appears in some judicial decisions. More commonly the question is the extent to which a minor has the capacity to ­participate in decision-making along with the parents. We will return to this idea in more detail later, under the heading of ‘capacity’, where we also discuss the concept of ‘best interests’ of the child.

17  National District Attorneys Association (2013) https://ndaa.org/wp-content/uploads/MinorConsent-to-Medical-Treatment-2.pdf; accessed 31 July 2019. 18  Legal Information Institute (2017) https://www.law.cornell.edu/wex/emancipation_of_minors; Associated Press (2018) https://www.nbcphiladelphia.com/news/local/Faith-Healing-Trial-GirlDeath-Trial-Pennsylvania-477837243.html; accessed 31 July 2019; National District Attorneys Association (2013) https://ndaa.org/wp-content/uploads/MinorConsent-to-Medical-Treatment-2.pdf; accessed 31 July 2019; Legal Information Institute (2017) https://www.law.cornell.edu/wex/emancipation_of_minors; accessed 1 June 2019. 19  AE Driggs, ‘The Mature Minor Doctrine: Do Adolescents Have the Right to Die?’ (2001) 11 Health Matrix 687. https://scholarlycommons.law.case.edu/cgi/viewcontent.cgi?article=1567&context=healthm atrix; accessed 1 June 2019.

464   Jessica Berg and Emma Cave

3  Overview of Children’s Rights in Europe Three international instruments in particular have influenced the development of ­children’s rights in Europe. The European Convention on Human Rights (ECHR) protects human rights and political freedoms in Europe and the Council of Europe (CoE), Convention on Human Rights and Biomedicine, and its additional protocols protects rights in the biomedical sector. Children also gain special and additional protection by virtue of the UN Convention on the Rights of the Child (CRC), which sets out international human rights standards for children and applies in Europe and beyond. The CRC is a core human rights treaty ratified by 196 states (the United States is not one of them). Four principles in particular guide parties to the CRC: Article 3 ensures that actions taken on behalf of the child have respect to their best interests. It relates to both individual children and children as a group. Article 6 protects the child’s right to life and development, Article 2 prohibits discrimination, and Article 12 promotes autonomy through an evolving capacities approach and a right to be heard.20 Though the CRC is widely ratified, the manner of implementation is left to individual signatories.21 The CRC established a Committee on the Rights of the Child which issues guidance and reports on States’ implementation, though the reports are not legally binding. Pressure is mounting to enhance implementation through direct incorporation. The United Kingdom, for example, has not incorporated the Convention into domestic law, and implementation differs by country and sector.22 Wales and Scotland have introduced duties incumbent on public authorities and ministers to protect and promote children’s CRC rights. The CoE is not bound by the CRC, but it has been influential in the interpretation of the ECHR by the ECtHR,23 which has developed a large body of case law developing children’s rights. In particular, Article 2 protects the right to life, and Article 8 protects private and family life. Article 8 has usually been considered from the perspective of the parents’ rights. For example, in Glass v the United Kingdom,24 Article 8 was breached when diamorphine was administered to a disabled child in a non-emergency situation over objections from his mother.

20  UN CRC General Comment no 5 CRC/CG/2003/527 (2003). 21  UNICEF ‘The UN Convention on the Rights of the Child: A Study of Legal Implementation in 12 Countries’ (2012) https://downloads.unicef.org.uk/wp-content/uploads/2012/11/UNICEFUK_2012CRCim plementationreport-FINAL-PDF-version.pdf?_ga=2.66202601.694347071.1552387991-1941889077.1552387991; accessed 1 June 2019. 22  A Lang, ‘House of Commons Library Briefing Paper: UN Convention on the Rights of the Child: A Brief Guide’ no 7721 (2016) 3. 23 See Neulinger and Shuruk v. Switzerland (App no 41615/07, 6 July 2010) [135]. 24  Glass v. UK no 61827/00, 9 March 2004, ECtHR. And see M.A.K. and R.K. v. the UK, nos 45901/05 and 40146/06, 23 March 2010, ECtHR.

Patient Autonomy, Capacity, and Consent: Children   465 The Convention on Human Rights and Biomedicine—known as the Oviedo Convention—has been signed by 34 members of the CoE and ratified by 28. Those who have not signed it, often because it is considered overly restrictive, include Austria, Germany, the United Kingdom, and Russia. Article 6 sets out protections for those unable to consent, requiring that the intervention is for their direct benefit and (except in an emergency) has the authorisation of a representative—usually a parent. It further requires that ‘the opinion of the minor shall be taken into consideration as an increasingly determining factor in proportion to his or her age and degree of maturity’.

4  The Best Interests Standard in Both the United States and Europe According to the Committee on the Rights of the Child, the best interests principle operates in three separate ways: as an interpretive principle whereby legal provisions open to interpretation are guided by the child’s best interests, a procedural rule whereby decision-making processes must evaluate the likely impact on the child and children, and a substantive legal right.25 Our concern in this section is primarily with the latter. Best interests is a flexible concept that evolves with the child’s development—as a child gains autonomy, the child is given more authority to determine what is in his or her best interests, as opposed to simply deferring to parental determinations. In the healthcare arena, the concept of best interests differs between states in its prioritization, definition, and linkage to other legal concepts. Some states require that child welfare is the paramount consideration. Whilst limitable, this reduces the scope to put the interests of others before the child. It is used in particular contexts, such as in the family law setting as adopted in the Children Act 1989 in England and Wales. The Act sets out a welfare checklist which includes the wishes and feelings of the child as well as their needs. The result is to limit parental rights to decide on behalf of the child. Assessments of the best interests by the court are from the child’s perspective: the parental view is a relevant consideration, but not binding.26 Many other European countries prefer to adopt the more flexible position taken in the CRC. Article 3 requires that best interests ‘shall be’ a primary consideration which eliminates discretion in its application but leaves scope for additional considerations. The Committee is clear that ‘an adult’s judgment of a child’s best interests cannot override the obligation to respect all the child’s rights under the Convention’.27 However, consideration of best interests does not necessitate their prioritization. A primary consideration gives leeway to balance the child’s best interests with their other rights and with other 25  UN CRC General Comment no 14 CRC/C/CG/14 (2013) para 6. 26  Re T (Wardship: Medical Treatment) [1997] 1 W.L.R. 242 at 250 per Butler-Sloss L.J.; Wyatt v. Portsmouth NHS Trust [2006] 1FLR 554; NHS Trust v. MB and B [2006] EWHC 507 (Fam), [16]. 27  UN CRC General Comment no 13 CRC/C/CG/13 (2011) para 61.

466   Jessica Berg and Emma Cave people’s rights. For example, some countries allow parents to refuse infant and child vaccinations whereas others mandate it.28 As medical paternalism has given way to individual autonomy as the predominant ethos of medical regulation, the legal definition of best interests has adapted to encompass a patient-centered definition as opposed to one controlled and defined by the medical profession. At one point, best interests was closely aligned to the concept of the reasonable doctor. What was best for the patient was judged by the doctor to be best medical interests, and the doctor was then held to account on the basis of the reasonableness of her decision. Today, the concept is demedicalized, focusing on best interests ‘in the widest sense and to include every kind of consideration capable of impacting on the decision’.29 The Committee has stated that The fact that the child is very young or in a vulnerable situation (e.g. has a disability, belongs to a minority group, is a migrant, etc.) does not deprive him or her of the right to express his or her views, nor reduces the weight given to the child’s views in determining his or her best interests.30

While a child’s own preferences can be taken into account in determining best interests, the CRC comment is rather broad. Lack of capacity should be taken into account in giving weight to a child’s views, in the same way lack of capacity would impact on the weight we would give to an adult expression of preferences (and CRC comments regarding a child’s developing autonomy support this approach). While Alderson has shown that even babies can communicate their feelings about treatment and suggests this should be a relevant consideration when deciding what is appropriate,31 it is highly unlikely that an infant’s expression of displeasure with treatment would be determinative. Instead, the concept of best interests can be interpreted to accommodate the effect of the process of treatment on the child, such that some treatments may be considered not in the child’s overall interests, even if the eventual effect is to extend life. For example, the fact that a treatment causes a great deal of pain or removes the child from the family for a long period of time can be taken into account under a best interests analysis. It is less clear how to think about external factors such as the impact of the treatment on the overall cohesiveness of a family or the financial burdens on the family of particular treatments under a best interests analysis. In the United States, the best interests standard has undergone a variety of changes over time. In general, family (and specifically parents) are thought to be the best judge of what is in a child’s best interests. In part, this is a recognition that the family knows the child best, and, in part, it is a recognition that the child’s interests are linked to those of 28  ASSET Project, ‘Compulsory Vaccination and Rates of Coverage Immunisation in Europe’ http:// www.asset-scienceinsociety.eu/reports/pdf/asset_dataviz_I.pdf; accessed 1 June 2019. And see comparative analysis, E Cave, ‘Voluntary Vaccination: The Pandemic Effect’ (2016) 37(3) Legal Studies 279; And see Vavřička and Others v. The Czech Republic applications 47621/13, 3867/14, 73094/14, 19306/15, 19298/15, and 43883/15 (ECtHR April 8, 2021). 29  An NHS Trust v. A [20087] EWHC 1696, [40]. 30  UN CRC General Comment no 14 CRC/C/CG/14 (2013) para 54. 31  P Alderson, Children’s Consent to Surgery (Open University Press 1993).

Patient Autonomy, Capacity, and Consent: Children   467 the family (who is likely to be caring for the child). Accordingly, as discussed earlier, the law generally defers to the parents’ assessment of a child’s best interests, with legal intervention when the parents are making decisions that clearly seem to compromise the child’s health. For example, when a health professional believes that a parental decision is inconsistent with the child’s best interests, the professional may take issue with the decision and ask a court to determine what is best for the child.

4.1 Capacity The best interests standard does not go away as a child develops capacity. Rather, the focus is on the interplay between the child’s interests in participating in decisions and the role of the parents (and state) in determining best interests. The intersection of best interest considerations with assessment of competence or capacity differs according to the health context. This section focuses on treatment decisions. While there is potential tension between Articles 3 (best interests) and 12 (evolving autonomy) of the CRC, the UN Committee on the Rights of the Child outlines a complementary role of the two principles: [O]ne establishes the objective of achieving the best interests of the child and the other provides the methodology for reaching the goal of hearing either the child or the children. In fact, there can be no correct application of article 3 if the components of article 12 are not respected.32

Eekelaar has argued that the concept of ‘dynamic self-determinism’ resolves potential tension, whereby the objective view of the minor’s best interests are tempered by reference to the values and beliefs of the minor.33 Indeed, the Committee on the Rights of the Child emphasises that best interests requires development of an approach to secure ‘holistic physical, psychological, moral and spiritual integrity of the child and promote his or her human dignity’.34 Best interests therefore includes respect for the child’s right to express opinions and for them to be given due weight. The Article 12 concept of the evolving capacities of the child is also relevant to a best interests determination. The Committee has said that The more the child knows, has experienced and understands, the more the parent, legal guardian or other persons legally responsible for him or her have to transform direction and guidance into reminders and advice, and later to an exchange on an

32  UN CRC General Comment no 12 CRC/C/CG/12 (2009) para 74. 33  J Eekelaar, ‘The Interests of the Child and the Child’s Wishes: The Role of Dynamic Self-determinism’ (1994) 8(1) Int’l J Law Pol’y & Family 42. 34  UN CRC General Comment no 14 CRC/C/CG/14 (2013) para 5.

468   Jessica Berg and Emma Cave equal footing. Similarly, as the child matures, his or her views shall have increasing weight in the assessment of his or her best interests.35

4.1.1  US Law The general rule in the United States is that minors are not legally competent to consent to medical treatment, although the age of majority may vary across states. Most states use age 18, but a couple use 19, and others have statutes which explicitly allow children below the age of majority to make certain decisions, including medical decisions (see earlier discussion).36 But while legal competence may be set at a certain age, capacity to make medical decisions is a case-specific analysis. Capacity develops throughout the child’s life. As a shorthand, some commentators (and physicians) refer to the ‘rule of sevens’, which developed in English Common Law regarding responsibility for torts and crimes. As applied to capacity, it would presume that children under age 7 lack capacity for medical decision-making, presume that children 14 and older have capacity, and assume that children in the middle range would vary, needing a case-by-case close evaluation. These presumptions can be overcome in any particular situation. The rule simply provides a shorthand, allowing age cutoffs to function as a trigger to a more in-depth exploration of a child’s decision-making capacity. So, for example, children age 7 and younger will likely play only a limited role in decisions, while many older adolescents (age 14 and above) may demonstrate the same decision-making capacity as young adults—and should play a significant role in their own medical decisions. This does not mean that very young children are not participants but that, since they are likely to exhibit minimal capacity, their participation will be more limited. An older child, by contrast, might be afforded the legal authority to make his or her own decisions, even over the objections of parents or healthcare providers. Thus there are cases in which a court has upheld the decisions of 16- and 17-year-olds to refuse continued treatment for cancer or to refuse a blood transfusion on the basis of religious beliefs. It is important to stress that participation does not mean the child has the legal authority to make a decision without any judicial or parental involvement. Similarly, saying a child lacks legal authority does not mean that he or she should not play any role in decision-making. Whether or not a child demonstrates full capacity to give a legally valid consent, it is widely believed that children at least should be asked to participate in decision-making in some manner.37 This might entail simply explaining what is going to happen to a young child, rather than just proceeding without any warning, or providing a child a chance to ask questions. One way to obtain more participation from children 35  Id., para 44. 36  See, e.g., https://download.lww.com/wolterskluwer_vitalstream_com/PermaLink/ASA/A/ ASA_2015_05_13_SMITH_1414141414_SDC4.pdf; accessed 1 June 2019. 37  AL Katz, SA Webb, and Committee on Bioethics, ‘Informed Consent in Decision-Making in Pediatric Practice’ (2016), http://pediatrics.aappublications.org/content/pediatrics/early/2016/07/21/ peds.2016-1485.full.pdf; accessed 1 June 2019.

Patient Autonomy, Capacity, and Consent: Children   469 who exhibit developing capacity is to ask for the child’s assent to treatment. The concept of assent appears in the US federal research regulations and refers to obtaining the child’s affirmative agreement to participate in a research trial. The regulations note that the assent should be accompanied by the permission of the child’s parents or guardian to meet the legal requirements of informed consent. While assent is required in a research context, it does not appear in state treatment consent laws so its adoption remains informal.

4.1.2  European Law Legal competence of minors to consent to medical treatment differs across Europe. Some favor legal certainty and adopt a bright line approach which might or might not coincide with the age of majority. Others favor an evolving capacities approach, setting out a statutory or common law test for capacity. Still others align laws on capacity to consent to medical treatment with existing approaches to other rights such as driving, alcohol consumption, or sex.38 In 2015, an EU report outlined considerable variation in approach.39 Nine Member States fix the age of consent to medical treatment by either age or a test for capacity. Others make child consent subject to parental approval until a certain age (e.g., the Netherlands) or allow parents to define when the child is capable of consent (e.g., Germany and Slovak Republic). In many countries, the provisions around child consent differ according to the medical procedure. When the potential for therapeutic value is reduced (as, for example, in clinical research or tissue donation), additional limitations may be placed on children providing consent. There is also variation in the extent of child participation. In some it is limited to information about the decision and in others it extends to participation in the decision itself. In relation to competence-based tests, the 1986 decision of Gillick v West Norfolk and Wisbech Area Health Authority40 from England and Wales has been influential. It establishes that children who satisfy a capacity test that is based on maturity and understanding can provide consent to treatments in their best interests. Lord Scarman said that I would hold that as a matter of law the parental right to determine whether or not their minor child below the age of 16 will have medical treatment terminates if and when the child achieves a sufficient understanding and intelligence to enable him or her to understand fully what is proposed. 38  See European Union Agency for Fundamental Rights, Consenting to Medical Treatment Without Parental Consent (2017), https://fra.europa.eu/en/publication/2017/mapping-minimum-agerequirements/consent-medical-treatments; accessed 1 June 2019. 39  Directorate-General for Justice and Consumers (European Commission) (2015) Evaluation of Legislation, Policy and Practice of Child Participation in the European Union (EU) EUR 309, iii, https:// publications.europa.eu/en/publication-detail/-/publication/3f3c50b2-6a24-465e-b8d1-74dcac7f8c42; accessed 1 June 2019. 40  Gillick v. West Norfolk & Wisbeck Area Health Authority [1986] AC 112. See E Cave ‘Goodbye Gillick? Identifying and Resolving Problems with the Concept of Child Competence’ (2014) 34(1) Legal Studies 103.

470   Jessica Berg and Emma Cave The test for Gillick capacity is functional rather than status-based. That is, capacity is decision-specific and may differ from one context to another. Subsequent cases have confirmed that it is also predicated on the outcome of the decision: Gillick capacity does not confer full authority on the child to make the decision. In Re R Lord Donaldson held that though a child with capacity can provide the doctor with the consent needed to avoid a claim in battery, so, too, can the parent or a court.41 The child’s refusal can therefore be vetoed in their best interests. Harris refers to the right conferred on children as ‘a right to acquiesce in a decision which has already been taken’.42 The distinction results in an asymmetry between consent and refusal. Manson defends the asymmetry by proposing a justification of the normative shared powers between parent and child: the transition of powers is gradual rather than instant.43 Where refusal of consent is based on religious ideals, Article 9 of the ECHR comes into play. Article 9 protects the right to freedom of thought, conscience, and religion. This right is limited by Article 9(2) on the grounds of public safety, public order, health, morals, and ‘for the protection of the rights and freedoms of others’. In Jehovah’s Witnesses of Moscow and Others v. Russia,44 Russia had banned certain practices of Jehovah’s Witnesses in part because the practice of refusing blood transfusions was considered to have a negative impact on life and health.45 Deciding against Russia, the ECtHR referred to the UK case of Re T, where Lord Donaldson said every adult has the right and capacity to decide whether or not he will accept medical treatment, even if a refusal may risk permanent injury to his health or even lead to premature death.46

Where children are concerned, the national courts have proved more reluctant to accept decisions that will result in death or serious adverse consequences to health. In the English case of Re L,47 for example, a 14-year-old Jehovah’s Witness with severe burns was required by the court to have a blood transfusion without her or her parents’ consent. It was held that she lacked Gillick capacity because she would not be able to understand the unpleasant nature of her death. The test for capacity here raised the bar considerably from the usual emphasis on understanding the treatment.

41  Re R (A Minor) (Wardship: Consent to Medical Treatment) [1991] 4 All ER 177, Court of Appeal. 42  J Harris ‘Consent and End of Life Decisions’ (2003) 29 J Med Ethics 10. 43  N Manson, ‘Transitional Paternalism: How Shared Normative Powers Give Rise to the Asymmetry of Adolescent Consent and Refusal? (2014) 29 Bioethics 66 See also S Gilmore, J Herring, ‘ “No” Is the Hardest Word: Consent and Children’s Autonomy’ (2011) 23(1) Child Family L Q 3; E Cave, J Wallbank, ‘Minors’ Capacity to Refuse Treatment: A Reply to Gilmore and Herring’ (2012) 20(3) Med L Rev 423. 44  Jehovah’s Witnesses of Moscow and Others v. Russia App no 302/02, Judgment 10.6.2010, ECtHR. 45  C O’Néill, ‘Jehovah’s Witnesses and Blood Transfusions: An Analysis of the Legal Protections Afforded to Adults and Children in European/English Human Rights Contexts’ (2017) 24(4) EJHL 368. 46  Re T (Adult: Refusal of Treatment) [1992] 3 W.L.R. 782, Court of Appeal. 47  [1998] 2 F.L.R. 810. And see Re E [1993] 1 F.L.R. 386.

Patient Autonomy, Capacity, and Consent: Children   471 The 2015 report also highlighted a ‘need for raised levels of awareness of the right of the child to participate’ in health. Provisions within Member States differ according to the qualification criteria for consent, with some imposing a de facto age limit and others determining eligibility in accordance with the maturity or competence of the child. Children generally have fewer opportunities to participate in developing, planning and reviewing healthcare services.48

Frequently participation is token, with little responsibility or even co-responsibility being conferred upon children. Child-friendly information is often absent or flawed, though there were examples of initiatives designed to address this, such as the CoE’s guidance on Child-Friendly Healthcare49 and the European Charter for Children in Hospital, Article 4 of which promotes the child’s right to information.50 Daly has argued that Article 12 of the CRC, which protects participation on the basis of evolving capacity, is empirically and theoretically insufficient to uphold children’s autonomy.51 Daly argues that, within the context of best interests proceedings, children should have choice in how they are involved (‘process autonomy’), and the decision (‘outcome autonomy’) provided to do so will not cause them significant harm. Though the age of majority is relatively standardized across Europe, there remains considerable variation in children’s rights to participate in medical treatment decisions and to provide or refuse consent. The standard of best interests is a guiding tool that applies across jurisdictions but which accommodates the views of parents and the autonomy of children in different ways. The ECtHR has shown reluctance to demand consensus and leaves Member States a wide margin of appreciation.52

5  New Challenges Regarding Medical Decision-Making for Minors There are a variety of new areas which challenge the current framework for medical decision-making for minors, in particular treatment for gender dysphoria, female genital mutilation (FGM), and reproductive decisions. These three areas are provided as examples here of the conflicts that arise when an adolescent child, parents, and state might all define the 48  See note 38, para 3.0. 49  Council of Europe, Committee of Ministers, Guidelines on Child‑Friendly Health Care (2011), https://www.coe.int/en/web/children/child-friendly-healthcare; accessed 1 June 2019. 50  European Association for Children in Hospital, European Charter for Children in Hospital, Leiden (1988, updated 2016), https://www.each-for-sick-children.org/images/stories/2016/Charter_ AUG2016_oSz.pdf; accessed 1 June 2019. 51  A Daly, Children, Autonomy and the Courts: Beyond the Right to be Heard (Brill 2018). 52  Gard and Others v. UK App no 39793/17, ECtHR.

472   Jessica Berg and Emma Cave child’s best interests differently, and the treatment in question is not a traditional ­life-sustaining therapy. In many of these cases, intervening (or not) will irrevocably change the course of the child’s life, and the window to act closes before the child reaches full capacity or the age of majority. Treatment for gender dysphoria is one of these areas. Many countries have liberalised their laws but most require periods of counselling and treatment before gender change is legally recognised. Malta (2015) and Norway (2016) have legislated to allow adults and children over the age of 6 to formally self-identify as male or female. The core principle in both countries is the child’s best interests. Many countries impose restrictions on access to trans-specific medical services (such as hormone therapy or surgery) on the basis of age. Equally intense controversy surrounds the treatment of those born with disorders of sex development (DSD), where the child is born with anatomy that conforms neither to female or male norms. This occurs in around 1 in 2,000 babies.53 Historically, gendernormalizing surgery was used to assign sex, with parental consent. Alternatively, surgery can be delayed until the child can contribute to the decision-making process or make their own competent decision.54 In the Australian case of Re A, a 14-year-old child assessed as genetically female had a condition that caused overproduction of the male sex hormone (congenital adrenal hyperplasia). A had been given surgery to alter the genitals to female appearance and hormone therapy, but A considered himself male. Could A receive surgical treatment including mastectomies, hysterectomy, closure of the labia, and creation of a penis? The court held that A lacked Gillick capacity and that the decision was not within the scope of parental authority. The court, however, could make the decision, and it held that the surgery was in A’s interests.55 These issues are likewise under debate in the United States. An added complication is that many of the treatments in question will result in sterilization (see also the later discussion). Advances also have been made to restrict parents from imposing cultural identity on  children without their consent and contrary to their best interests in the case of FGM. FGM is usually carried out between infancy and the age of 15. It can cause bleeding, infection, and cysts and lead to complications in childbirth, psychological problems, and death. It is a practice that reflects inequality and is internationally recognised as a violation of the human rights of girls and women. In 2012, the European Parliament resolved to end FGM in Europe through measures of prevention, protection, and ­legislation. The UK has legislated to prohibit FGM, and the Family Court has found that FGM amounts to significant harm and can never constitute reasonable parenting.56 Similarly, many US states have passed laws banning the practice, and the Female Genital

53  L Liao, D Wood, S Creighton, ‘Parental Choice on Normalising Cosmetic Genital Surgery’ (2015) BMJ 351 h5124. 54  Re A (Surgical Treatment for Congenital Adrenal Hyperplasia) [1993] Fam L.R. 715 (Australia). 55  K Parlett, K-M Weston-Scheuber, ‘Consent to Treatment for Transgender and Intersex Children’ (2004) 9(2) Deakin L Rev 375. 56  Re B (Children) (Care Proceedings) [2015] 1 F.L.R. 905 (Family Court).

Patient Autonomy, Capacity, and Consent: Children   473 Mutilation Act of 1996 makes it a felony to perform FGM on anyone under age 18 or send her out of the country to have it performed.57 Article 8 of the CRC also protects reproductive rights, including the right to access lawful abortion.58 Abortion law varies considerably between US states and between different countries, and a full exploration of this area for either minors or adults is beyond the scope of this chapter. Performing an abortion without a minor’s involvement in the decision would be unusual, and performing an abortion directly against the minor’s wishes is extremely rare. The debates usually arise in the situation in which a minor is seeking an abortion, either without parental consent or against the parents’ wishes. In addition, US law limits the rights of parents or guardians to make decisions regarding the sterilization of minors, including decisions regarding treatment that has a side effect of damaging reproductive potential (such as cancer treatments or gender reassignment treatment). Article 12 of the CRC protects the right to marry and found a family, indirectly raising the issue of sterilization. Adults can elect to store gametes and later utilize assisted reproductive technology. For females, the options include transplanting or preserving ovarian tissue, oocyte cryopreservation, or embryo preservation (in which case donor sperm would be needed to fertilize the egg). For males, testicular tissue cryopreservation can be utilized, or sperm storage. The options are more limited, and thus the issue more complex, in the case of a prepubertal child. A high level of understanding is needed to balance any risks inherent in the procedure and in delaying treatment against the potential future benefits of having a genetically related child.59 Most countries agree that, at the very least, information should be provided about the risks of reproductive harm and the options for fertility preservation discussed, so that the family can reach a decision.

6 Conclusion Medical decision-making for children is a complicated balance between many different factors, including the child’s best interests, the developing autonomy and rights of children, parental rights to make decisions about their child’s upbringing, and state interests in protecting life and health and assuring children reach adulthood so they can make their own decisions. Even cases involving infants who cannot participate at all in decisionmaking can be difficult since there may be disagreement between medical professionals and families or within families about what constitutes a child’s best interest (as evidenced by the Charlie Gard case). Additional challenges arise as children develop capacity and are included in decision-making. Finally, as involvement in medical decision-making is 57  18 U.S.C. 116; available at https://www.law.cornell.edu/uscode/text/18/116; accessed 1 June 2019. 58  A., B. and C. v. Ireland App no 25579/05. 16 December 2010 (Grand Chamber). 59  See JM Levine ‘Preserving Fertility in Children and Adolescents with Cancer’ (2014) 1(2) Children 166.

474   Jessica Berg and Emma Cave designed both to recognize and facilitate a child’s developing autonomy, questions about treatments that go to fundamental issues of identity, such as sterilization or treatment for gender dysphoria, highlight the difficulties with our current frameworks and indicate how much more work needs to be done in this area.

Bibliography Alderson P (1993) Children’s Consent to Surgery. Open University Press. ASSET Project (2019) Compulsory Vaccination and Rates of Coverage Immunisation in Europe, http://www.asset-scienceinsociety.eu/reports/pdf/asset_dataviz_I.pdf; accessed 1 June 2019. Cave E (2014) ‘Goodbye Gillick? Identifying and Resolving Problems with the Concept of Child Competence’, Legal Studies 34(1), 103–122. Cave E (2016) ‘Voluntary Vaccination: the Pandemic Effect’, Legal Studies 37(3), 279–304. Cave E, Nottingham E (2017) ‘Who Knows Best (Interests)? The Case of Charlie Gard’, Medical Law Review 26(3), 500–513. Cave E, Wallbank J (2012) ‘Minors’ Capacity to Refuse Treatment: A Reply to Gilmore and Herring’, Medical Law Review 20(3), 423–449. Charter of Fundamental Right of the EU Official Journal of the European Communities Official Journal of the European Communities (2000/C-364/01), https://www.europarl. europa.eu/charter/pdf/text_en.pdf; accessed 1 June 2019. Council of Europe, Committee of Ministers (2011) Guidelines on Child‑Friendly Health Care, 21 September 2011. https://rm.coe.int/168046ccef; accessed 1 June 2019. Council of Europe (1997) Convention for the Protection of Human Rights and Dignity of the Human Being with Regard to the Application of Biology and Medicine (1997) Convention on Human Rights and Biomedicine, CETS No. 164, https://rm.coe.int/168007cf98; accessed 1 June 2019. Directorate-General for Justice and Consumers (European Commission) (2015) Evaluation of  Legislation, Policy and Practice of Child Participation in the European Union (EU) EUR 309. https://www.eurochild.org/fileadmin/public/05_Library/Thematic_priorities/05_ Child_Participation/European_Union/Evaluation_of_legislation__policy_and_practice_ of_child_participation_in_the_European_Union.pdf; accessed 1 June 2019). Driggs AE (2001) ‘The Mature Minor Doctrine: Do Adolescents Have the Right to Die?’, Health Matrix 11, 687. https://scholarlycommons.law.case.edu/cgi/viewcontent.cgi?ar­ticle=1567& context=healthmatrix; accessed 1 June 2019). EACH European Association for Children in Hospital (1988, updated 2016) European Charter for Children in Hospital, Leiden. https://www.each-for-sick-children.org/images/stories/2016/ Charter_AUG2016_oSz.pdf; accessed 6 January 2019. Eekelaar J (1994) ‘The Interests of the Child and the Child’s Wishes: The Role of Dynamic SelfDeterminism’, Inter J Law Policy Fam 8(1), 42–61. Legal Information Institute (updated 2017) Emancipation of Minors. https://www.law.cornell. edu/wex/emancipation_of_minors; accessed 1 June 2019). European Commission (2011) ‘An EU Agenda for the Rights of the Child’ (Communication) COM/2011/0060 final. https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:201 1:0060:FIN:en:PDF; accessed 1 June 2019.

Patient Autonomy, Capacity, and Consent: Children   475 European Parliament Directorate-General for Internal Policies (2012) EU Framework of Law for Children’s Rights. Brussels. http://www.europarl.europa.eu/RegData/etudes/note/join/ 2012/462445/IPOL-LIBE_NT(2012)462445_EN.pdf; accessed 1 June 2019. European Parliament (2014, 27 November) Resolution on the 25th anniversary of the UN Convention on the Rights of the Child, 2014/2919(RSP). http://www.europarl.europa.eu/ doceo/document/TA-8-2014-0070_EN.html?redirect; accessed 1 June 2019. European Parliament (2012, 14 June) Resolution on Ending Female Genital Mutilation 2012/ 2684(RSP).  http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT?TA-20120261??????; accessed 1 June 2019. European Union (2007, 13 December) Consolidated version of the Treaty on European Union, 2008/C 115/01. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ%3AC%3A2008% 3A115%3ATOC; accessed 1 June 2019. European Union Agency for Fundamental Rights and Council of Europe (2015) Handbook on European Law Relating to the Rights of the Child. https://fra.europa.eu/en/publication/2015/ handbook-european-law-child-rights; accessed 1 June 2019. Feinberg J ‘A Child’s Right to an Open Future’, in Whose Child? Children’s Rights, Parental Authority and State Power, W. Aiken and H. LaFollette (eds) (Rowman and Littlefield 1980). Gilmore S, Herring J (2011) ‘ “No” is the Hardest Word: Consent and Children’s Autonomy’, Child and Family Law Quarterly 23(1), 3–25. Great Ormond Street Hospital, Position Statement Issued at High Court on 13 July 2017, FD17P00103, https://www.gosh.nhs.uk/file/23611/download?token=aTPZchww; accessed 1 June 2019. Guttmacher Institute (updated 2019) An Overview of Minors’ Consent Law https://www. guttmacher.org/state-policy/explore/overview-minors-consent-law; accessed 1 June 2019. Harris J (2003) ‘Consent and End of Life Decisions’, Journal of Medical Ethics 29, 10–15. Horowicz EM (2017) ‘Intersex Children: Who Are We Really Treating?’, Med Law Int 17(3), 183–218. Katz AL, Webb SA, and Committee on Bioethics Informed Consent in Decision-Making in Pediatric Practice (2016), http://pediatrics.aappublications.org/content/pediatrics/early/ 2016/07/21/peds.2016-1485.full.pdf; accessed 1 June 2019. Lang A House of Commons Library Briefing Paper: UN Convention on the Rights of the Child: A Brief Guide No. 7721 (UK House of Commons 2016), https://researchbriefings. parliament.uk/ResearchBriefing/Summary/CBP-7721#fullreport; accessed 1 June 2019. Levine JM (2014) ‘Preserving Fertility in Children and Adolescents with Cancer’, Children 1(2), 166–185. Liao L, Wood D, Creighton S (2015) ‘Parental Choice on Normalising Cosmetic Genital Surgery’, BMJ 351 h5124. Lundy L, Kilkelly U, Byrne B (2013) ‘Incorporation of the United Nations Convention on the Rights of the Child in Law: A Comparative Review’, Int J Child Rights 21(3) 442–463. Manson N (2014) ‘Transitional Paternalism: How Shared Normative Powers Give Rise to the Asymmetry of Adolescent Consent and Refusal’, Bioethics 29, 66. Minor Consent to Medical Treatment Laws (2013) https://ndaa.org/wp-content/uploads/ Minor-Consent-to-Medical-Treatment-2.pdf; accessed 31 July 2019. Montanari Vergallo G, Marinelli E, di Luca NM and Zaami S (2018) ‘Gamete Donation: Are Children Entitled to Know Their Genetic Origins? A Comparison of Opposing Views. The Italian State of Affairs’, EJHL. doi: 10.1163/15718093-12530378.

476   Jessica Berg and Emma Cave O’Néill C (2017) ‘Jehovah’s Witnesses and Blood Transfusions: An Analysis of the Legal Protections Afforded to Adults and Children in European/English Human Rights Contexts’, EJHL 24(4) 368–389. Parlett K, Weston-Scheuber K-M (2004) ‘Consent to Treatment for Transgender and Intersex Children’, Deakin Law Rev 9(2) 375–397. Taylor MJ, Dove ES, Laurie G, Townsend D (2018) ‘When Can the Child Speak for Herself? The Limits of Parental Consent in Data Protection Law for Health Research’, Med Law Rev 26(3) 369–391. UN Convention on the Rights of the Child, Committee on the Rights of the Child (2003) General Comment No. 5 on the General Measures of Implementation of the Convention on the Rights of the Child, para. 12 CRC/GC/2003/5. UN Convention on the Rights of the Child, Committee on the Rights of the Child (2009) General Comment No. 12: The Right of the Child to be Heard CRC/C/GC/12. UN Convention on the Rights of the Child, Committee on the Rights of the Children (2013) General comment No. 1 on the Right of the Child to Have His or Her Best Interests Taken as a Primary Consideration CRC/C/GC/14. UN General Assembly (1989, 20 November) Convention on the Rights of the Child. United Nations, Treaty Series, vol. 1577. Wade K (2017) ‘Defining the Threshold of Permissible Risk for Non-therapeutic Clinical Trials with Children in Europe’, EJHL 24(4), 414–431. World Health Organization (2018) Female Genital Mutilation Fact sheet. https://www.who. int/news-room/fact-sheets/detail/female-genital-mutilation; accessed 1 June 2019.

chapter 24

Patien t Au tonom y, Ca pacit y, a n d Consen t Vulnerable Adults Mary Donnelly and Jessica Berg

1 Introduction The legal and ethical principle of autonomy has long been a cornerstone of health law. Whether derived from Millian liberalism1 or the Kantian categorical imperative,2 the idea that ‘every human being of adult years and sound mind has a right to determine what should be done with his own body’3 is an unassailable legal principle, protected by both the common law4 and by human rights instruments/bills of rights.5 From this principle emerged initially a legal requirement for consent to healthcare interventions,6 followed by the requirement that such consent must be informed.7 So, too, did the 1  As expounded in particular in John Stuart Mill, On Liberty (1859) in John Grey (ed), On Liberty and Other Essays (Oxford University Press 1991). 2  Immanuel Kant, Groundwork of the Metaphysics of Morals (1785) in Mary Gregor (ed), Kant: Practical Philosophy (Cambridge University Press 1996). 3  Schloendorff v Society of New York Hospital, 105 N.E. 92 (NY 1914). 4  See e.g., United Pacific Railway Co v Botsford, 114 US 250 (1891); S v McC (orse S) and M (DS Intervener); W v W [1972] AC 24. 5  See, e.g., Planned Parenthood of Southeastern Pennsylvania v Casey, 505 US 833 (1992); Washington v Glucksberg, 521 US 702 (1997); Pretty v UK (2002) 35 EHRR 1 (European Convention on Human Rights). 6  Although the requirement for consent is most closely associated with respect for autonomy, it also has a principled basis in respect for human dignity and bodily integrity: see Michael Rosen, Dignity: Its History and Meanings (Harvard University Press 2012); Mary Neal, ‘Respect for Human Dignity as “Substantive Basic Norm” ’ (2014) 10 Int’l J L Context 26. 7  See Ruth Faden and Tom Beauchamp, A History and Theory of Informed Consent (Oxford University Press 1986).

478   Mary Donnelly and Jessica Berg recognition of the right to refuse treatment, even when that treatment is potentially life-­saving or sustaining.8 Indeed, one prominent senior English judge went so far as to describe the right to choose whether to consent to or refuse treatment as ‘absolute . . . notwithstanding that the reasons for making the choice are rational, irrational, unknown or even non-­existent’.9 While this is clearly an exaggeration, there are relatively few overt exceptions to the right to refuse treatment, with those that do arise generally occurring in legislation around mental health and infectious diseases. The legal effect of the requirement for consent is that a medical intervention without consent constitutes an assault/battery, with (theoretically at least) consequences in both criminal and civil law. In practice, in both Europe and the United States, the likelihood of criminal proceedings is slim, other than in situations of deliberate attempt to harm. In common and civil law jurisdictions, civil proceedings for assault/battery are also rare.10 One reason for this in common law jurisdictions is that, for the past three decades, any action based on a lack of informed consent has been found to lie in the tort of negligence rather than in the tort of battery.11 This shift results from the fact that, in many cases, there exists some form of consent to proceed (making battery an inappropriate charge), although perhaps only ‘consent’ in the simplest sense of the term. In fact, some early cases viewed the fact that the individual sought out the services of the practitioner or remained present for the treatment as sufficient to constitute consent. With the shift to a negligence cause of action, we see the development of the legal requirements that consent be informed and that the patient has capacity. And these requirements—information disclosure and decision-­maker capacity—form the core of our emphasis on autonomy in healthcare decision-­making. Law’s conception of autonomy in healthcare is premised on certain assumptions. One is that respect for the principle of personal autonomy operates alongside professional autonomy and cannot be used to require medical professionals to provide treatment which they do not consider to be appropriate. Thus, as the philosopher Onora O’Neill describes, ‘[w]hat is rather grandly called ‘patient autonomy’ often amounts to a right to choose or refuse treatments on offer, and the corresponding obligation of practitioners not to proceed without patient consent’.12

8  The right to refuse emerged initially in religious-­based blood refusal cases in the United States (e.g., Erickson v Dilgard (1962) 44 Misc 2d 27) before extending first to people who were terminally ill (e.g., Re Quackenbush (1978) 156 NJ Super 282) and then more generally (e.g., Bouvia v Superior Court (1986) 179 Cal App 3d 1127). 9  Re T (Adult: Refusal of Medical Treatment) [1992] 3 WLR 782, 786 per Lord Donaldson MR. 10  See Marc Staunch, ‘Medical Malpractice and Compensation in Germany’ (2011) 86(3) Chicago-­ Kent L Rev 1139; Simon Taylor, Medical Accident Liability and Redress in English and French Law (Cambridge University Press 2015). 11  This follows court decisions on both sides of the Atlantic: see Canterbury v Spence (1972) 464 F 2d 772 (US); Chatterton v Gerson [1981] QB 432 (EW); Walsh v Family Planning Services [1992] 1 IR 496 (Irl). 12  Onora O’Neill, Autonomy and Trust in Bioethics (Cambridge University Press 2002) 37.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   479 A second assumption is that law accepts that there are some people in respect of whom the principle of autonomy does not apply. Law relies on the concept of capacity (or competence)13 ‘to sort persons into two classes: (1) those whose voluntary decisions . . . must be respected by others and accepted as binding, and (2) those whose decisions, even if uncoerced, will be set aside and for whom others will act as surrogate decision-­ makers’.14 Within the traditional legal approach, once a determination was made that a person lacked capacity, he or she lost all decision-­making authority, becoming in effect, a person for whom others acted, typically on the basis of an objective (though largely unmonitored) assessment of his or her best interests or on another person’s assessment of what the patient would have done if he or she had had capacity (typically referred to as substituted judgment). This kind of simplistic distinction on the basis of a determination of capacity alone is becoming less legally acceptable. The shift in the law’s approach derives from several different bases. The first, which began to have a significant impact from the 1980s, initially in the United States, was the recognition of the conception of precedent (or prior) autonomy, through which a person with capacity is facilitated in directing, to varying degrees, the fate of his or her subsequently incapacitated self. This conception has been given wide effect through legislation allowing a person to make an advance directive or to grant a lasting (or enduring) power of attorney authorising a designated person to make decisions on the patient’s behalf in the event of a loss of capacity. The concept of precedent autonomy has not been without philosophical or practical complexity, in particular the difficulty in determining whether to be guided by an individual’s prior competent instructions or current (potentially incompetent) wishes.15 Substituted judgment is sometimes referred to as a ‘legal fiction’ since no one can truly determine what another person would have wanted. By contrast, precedent autonomy looks directly at what the individual decided at an earlier point in time while competent. While the two concepts—precedent autonomy and substituted judgment—often work together, with surrogate decision-­makers often looking at prior statements (written or oral) to decide what to do at the current time, they are distinct ideas. In the first, we recognise that respect for autonomy includes respecting choices about future care. The second concept, by contrast, doesn’t seem to promote autonomy at all since it allows someone else to make the decision (albeit based on what the decision-­maker thinks the patient would have wanted to do). In this sense, substituted judgment may be based less on respect for patient autonomy than on a reflection that other facts should come into 13  In this chapter, we use the terms ‘capacity’ and ‘competence’ interchangeably. This reflects different usages in Europe (where the term ‘capacity’ is more commonly used) and the United States (where the term ‘competence’ is more commonly used to indicate legal authority versus ‘capacity’, which is determined by a medical professional). 14  Allen Buchanan and Dan Brock, Deciding for Others: The Ethics of Surrogate Decision-­Making (Cambridge University Press 1989) 47. 15  See Allen Buchanan, ‘Advance Directives and the Personal Identity Problem’ (1988) 17(4) Philosophy and Public Affairs 277.

480   Mary Donnelly and Jessica Berg play in healthcare decision-­making beyond merely medical ones and that healthcare providers cannot simply decide for patients without input from someone who can bring into consideration these other factors. The extent to which these factors can be taken into consideration in substituted judgment has always been a challenge, and the concept works best when the individual making the substituted decision knows the patient well and thus can best determine what factors he or she would want to be considered. The further away we move from a decision-­maker who is close to the patient, the further away we move from substituted judgment. Perhaps, then, autonomy is best promoted for an incompetent patient when the individual, while competent, indicates both the issues of importance to him or her and the person who he or she would like to take those into consideration when making decisions at a later point in time when the individual is incompetent. In this context, precedent autonomy and substituted judgment are combined. Outside the United States, a second basis for the move away from a simplistic capacity-­based distinction derives from the growing role played by human rights instruments in the construction of legal capacity. At an international level, the United Nations Convention on the Rights of Persons with Disabilities (CRPD)16 has sought to alter radically the traditional consent/capacity paradigm. The CRPD has been ratified by 182 States Parties, although not by the United States. Underpinned by a normative basis in equality, nondiscrimination, autonomy, and dignity,17 the CRPD seeks to protect the equal rights of all persons with disabilities, including people with ‘mental’ or ‘intellectual’ impairments.18 The principles underpinning the CRPD include respect for the inherent dignity and the individual autonomy and freedom of choice of persons with disabilities (Article 3). Article 12 is of particular significance for people with impaired capacity. This Article requires States Parties to recognise that ‘persons with disabilities enjoy a right to legal capacity on an equal basis with others in all aspects of life’. It also requires that all measures relating to the exercise of legal capacity must ‘respect the rights, will and preferences of the person’ and imposes a positive duty on States Parties to ensure that persons with disabilities have access to the support they require in exercising their legal capacity.19 Article 12 represents a significant departure from traditional approaches to capacity. It rejects the binary division between people with/without capacity and operates instead on the basis that, through appropriate supports, most

16  A/RES/61/106 Annex 1. The CRPD entered into force on 2 May 2008, on receipt of its twentieth ratification. 17  The applicable principles are set out in Art. 3. 18  On the definition of persons with disabilities, see Art. 1. 19  Aspects of the wording of Article 12 are (deliberately) opaque, reflecting political compromises in the negotiation of the Convention: see Amita Dhanda, ‘Legal Capacity in the Disability Rights Convention: Stranglehold of the Past or Lodestar for the Future?’ (2006–2007) 34 Syracuse J Int’l L & Com 429, 448.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   481 (arguably all20) people can make their own decisions. For this reason, it has been one of the more controversial aspects of the CRPD,21 and its ultimate impact on legal and policy developments remains unclear. Legal frameworks in Europe and in the United States have developed distinct ways of addressing issues relating to capacity and consent to treatment.22 National-­level capacity legislation and human rights norms play a central role in Europe, which means that healthcare decisions in situations of impaired capacity operate (in theory at least) in accordance with a national standard. In the United States, the legal framework is more state-­based (rather than federal), and the courts have played a significant role, with both common law and legislation varying considerably across jurisdictions. Despite these differences, some similar legal principles have developed. Reflecting this, this chapter identifies a number of key themes (the role of competence and capacity, advance directives, and decisions made for others) and analyses the ways these are treated in the United States and in selected European jurisdictions. This thematic discussion is preceded by a more general consideration of the overarching features of the legal approaches to autonomy, capacity, and consent in the United States and in Europe.

2  The Legal Frameworks: An Overview While legal and ethical frameworks in Europe are driven primarily by legislation, the development of the doctrine of informed consent and the attendant concepts of autonomy and capacity are driven primarily by common law in the United States.

2.1  The United States While one of the first cases to indicate a right to consent was decided at the turn of the 20th century,23 the doctrine of informed consent was developed through a series of judicial decisions in the late 1960s and early 1970s. Today, informed consent has become a 20  This appears to be the interpretation of the Committee on the Rights of Persons with Disabilities (the body charged with enforcing the CRPD) in its interpretation of Article 12 in General Comment no 1 (2014): Article 12: Equal Recognition before the Law, CRPD/C/GC/1. General Comment no 1 states that Article 12 requires the rejection of the functional test for capacity (para 15) and the abolition of substitute decision-­making (para 28). It does, however, concede that sometimes notwithstanding that significant efforts have been made, it will not be possible to determine the will and preferences of the person and states that, in such circumstances, the relevant standard should be ‘best interpretation of will and preferences’ (para 21). 21  See, e.g., MC Freeman et al, ‘Reversing Hard Won Victories in the Name of Human Rights: A Critique of the General Comment on Article 12 of the UN Convention on the Rights of Persons with Disabilities’ (2015) 2(9) Lancet Psychiatry 844. 22  Issues relating to consent to research are discussed in the chapter by Carl Coleman in this volume. 23  Schloendorff v Society of New York Hospital, 105 N.E. 92 (NY 1914).

482   Mary Donnelly and Jessica Berg standard part of medical practice. It is an interesting question whether the judicial doctrine drove the ethical one or vice versa. Prior to development of the judicial doctrine, consent played little or no role in standard medical practice in the United States; it does not show up in ethical codes until late in the 20th century. Informed consent for research developed separately from informed consent for medical practice but also was not developed until the middle of the 20th century in the United States. Unlike treatment consent, research consent is governed by a series of federal administrative regulations. Informed consent causes of action are usually brought under a negligence cause of action (all US states but Louisiana are common law jurisdictions). But Pennsylvania still uses a battery cause of action, and all states may allow a battery suit to be brought where the issue is not the lack of informed consent but the lack of any consent at all. For example, absent an emergency, a battery claim would lie when a surgeon does an additional procedure while the patient is unconscious. The choice between using a battery cause of action versus a negligence one makes a difference both in the types of testimony the plaintiff must bring in to make his or her case and the potential defenses allowed to defendants. Generally, negligence is harder to win than battery, although historically judges and juries were loath to find a physician liable for battery (as a pugilist in a brawl), so it is unclear how the two causes of action might compare in practice if they were both used regularly now. Informed consent is really made up of two requirements: a duty to disclose information and a right to make decisions. To meet the information requirement, physicians must disclose basic information about the patient’s diagnosis and treatment options along with their risks, benefits, and alternatives. There is a great deal of debate about how much must be disclosed about the treatment (costs, for example) and whether factors related to the individual doing a procedure must be disclosed (e.g., experience, success rates, financial interests, substance abuse problems, etc.). These debates indicate the limitations of trying to use informed consent to solve different problems. While information is important, we cannot expect that disclosure alone will solve financial conflicts of interest—indeed, there is evidence that disclosure can exacerbate the problem.24 Nor can we rely on patient choice to take the place of professional self-­regulation for assuring competent care or preventing substance abuse. Once provided information, the patient is asked to either consent to or refuse care,25 meeting the decision-­making requirement. The vast literature and extensive case law on informed consent will not be rehashed here. We should note, however, the specific situations in which either or both requirements (disclosure and consent) are altered or avoided.

24  For discussion of this phenomenon, see chapter by Richard Saver on conflicts of interest in this volume. 25  In this sense, ‘informed choice’ may be a better name for the doctrine than ‘informed consent’ since refusals must also be informed.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   483 The established exceptions include incompetence,26 emergencies, waiver,27 therapeutic privilege, and some public health interventions.28 Incompetence will be discussed further under the heading of capacity. Emergency exceptions seem obvious: delaying emergency treatment to facilitate autonomous decisions makes little sense. Unless we know otherwise, we assume that people want care in emergencies. Waiver is the recognition that exercising autonomy to make decisions should also include the right to choose not to make a choice. While this make sense on a conceptual level, the application of waiver can be difficult. A patient who wants to waive the right to make decisions is easier to reconcile with the doctrine of informed consent than one who wants to waive the right to disclosures. The latter begs the question of whether someone has the right to uninformed consent. Therapeutic privilege remains a highly controversial exception and rarely should apply nowadays. It is based on the idea that some information would be so harmful that disclosing it to the patient would prevent the ability to make decisions (it would basically render them unable to process any information). In the past, this exception was used to justify physicians withholding cancer diagnoses from patients. While providers should certainly consider the timing and setting in which to disclose difficult information, the concept of information itself being so very harmful as to undermine autonomy is difficult to reconcile with what we now know about people’s capacity to deal with bad news. As a result, the therapeutic privilege exception applies rarely, but courts have been unwilling to jettison the concept completely. Finally, public health exceptions allow treatment in two contexts: first permitting consent from minors in certain public health contexts regardless of capacity (e.g., contraception access, sexually transmitted infection [STI] treatment, substance abuse treatment, and mental health treatment) and the second allowing the imposition of some treatments for infectious disease either without consent or even over a competent refusal. The former context varies between states, and not all include the same treatment areas or use the same age of consent. The latter context also varies, usually on a case-­by-­ case basis, and forced treatment almost always requires judicial intervention or at least an opt-­out mechanism. In all of these cases the state interest in maintaining public health by assuring that patients get needed care overrides the concern that some of the consents may be uninformed or given by incapacitated patients. Why require individual informed consent, and what is its relationship to autonomy and capacity? There are a few rationales that have been posited. First, individuals are most likely to know their interests and thus make better choices for their health and well-­being. All treatments have varying degrees of potential benefit and possible risks—the patient 26  See Jessica W Berg, Paul S Appelbaum and Thomas Grisso, ‘Constructing Competence: Formulating Standards of Legal Competence to Make Medical Decisions (1996) 48(2) Rutgers L Rev 345. 27  Jessica W Berg, ‘Understanding Waiver’ (2003) 40 Houston L Rev 281. 28  Jessica W Berg, ‘All for One and One for All: Informed Consent and Public Health’ (2012) 50 Houston L Rev 1.

484   Mary Donnelly and Jessica Berg is best able to determine how that balance resolves for him- or herself. Second, the information requirement may increase the likelihood that the intervention will be beneficial since the individual better understands what to expect, including being prepared to recognise problems that may arise. If a patient is left completely in the dark about the treatment and risks, she may be unable to spot problems that arise or activities/other treatments that should be avoided. Third, even if individuals err in some of their choices, we are better off as a society if we encourage individual decision-­making and thus develop autonomous citizens. Autonomous decision-­making is valued in many societies generally (particularly in the United States), and medical decision-­making is one context in which to develop these skills. Fourth, individuals have a right to control what happens to their bodies. We have made a determination as a society that we generally do not want the government or healthcare professionals making these decisions. The first three of these rationales are straightforwardly utilitarian: more utility overall comes from allowing individual choice even if, in a particular situation, one could argue that the individual is making a poor decision. But a utilitarian analysis may not weigh in favour of encouraging (or even allowing) individual autonomy in decision-­making in all circumstances. While overall we may be a better society if we encourage individual decision-­making, individual choice may be less beneficial in a public health context. This may explain the limitations on informed consent in the area of infectious disease. The final rationale for promoting individual informed consent is rights-­based: that a person has the right to control what happens to her body and is linked closely to determinations of capacity.

2.2 Europe Inevitably, there is discernible variation in the approach to issues of consent and capacity across European jurisdictions. However, some common features may be identified. This commonality in approach derives in part from the shared liberal democratic tradition and in part from supranational frameworks, such as the European Union and the Council of Europe (CoE)/European Court of Human Rights (ECtHR). First, as in the United States, there is widespread recognition across European jurisdictions of the necessity for informed consent to medical treatment. In some jurisdictions, for example France, this derives from legislation29; in some, for example England and Wales, it derives from the common law30; and, in some, for example Ireland, it also has a basis in autonomy rights protected by a written constitution.31 The requirement

29  Article 16-­1 of the Code Civil states that ‘[e]veryone had the right to respect for his or her body’, and Article L. 1111–1114 of the Public Health Code states that ‘no medical act and no treatment can be practiced without the free and informed consent of the person and this consent may be withdrawn at any moment’. 30 See Re T (Adult: Refusal of Medical Treatment) [1992] 3 WLR 783. 31  Article 40.3.1 of the Constitution of Ireland (as interpreted in In re a Ward of Court [1996] 2 IR 79).

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   485 for informed consent has also been recognised on a pan-­European level by the ECtHR.32 The requirement for informed consent gives rise to a duty to disclose necessary information to allow a patient to make an informed choice. In the European common law jurisdictions (the United Kingdom and Ireland), failure to disclose information about material risks gives rise to an action in the tort of negligence.33 In civil law jurisdictions, the scope of the action depends on the relevant parts of the applicable civil code. As in the United States, the requirement for informed consent operates alongside a right to refuse treatment. While some of the early case law tended to recognise the right to refuse in principle rather than actually permit refusal in practice,34 this has now changed.35 In England and Wales, in particular, recent jurisprudence indicates robust endorsement of (and practical delivery on) the right of an adult (with capacity) to refuse treatment.36 In many (though not all) European jurisdictions, this general position is statutorily set aside in relation to treatment for a mental disorder where a person has been involuntarily admitted under mental health legislation.37 It is less common for the requirement for consent to treatment to be set aside where the person has an infectious disease. However, in such cases, it is the norm for legislation to provide for compulsory quarantining or admission to a hospital/secure location to prevent the spread of the diseases.38 The Irish care of S v Health Service Executive is a helpful example.39 The applicant here had a highly infectious, antibiotic-­resistant form of tuberculosis for which she refused treatment. By the time her case came before the High Court, she had been detained for more than 11 months in a negatively pressurised hospital room, which was 32  See Wannes Buelens, Coralie Herijgers and Steffi Illegems, ‘The View of the European Court of Human Rights on the Competent Patient’s Right of Informed Consent: Research in Light of Articles 3 and 8 of the European Convention on Human Rights’ (2016) 23(5) Eur J Health L 481. 33 See Montgomery v Lanarkshire Health Board [2015] UKSC 11 (UK); Walsh v Irish Family Planning Services [1992] 1 IR 496 (Irl). 34  See e.g., Re T (Adult: Refusal of Medical Treatment) [1992] 3 WLR 782, where the England and Wales Court of Appeal enthusiastically endorsed the right to refuse but held that, in the circumstances of the case (the refusal of a potentially life-­saving blood transfusion by a young woman on religious grounds), the refusal should not be upheld because of the possibility of undue influence by the young woman’s mother (who was a Jehovah’s Witness) and because the young woman had been misled by healthcare professionals as regards the consequences of refusal. 35  See, e.g., Clayton O’Néill, ‘Jehovah’s Witnesses and Blood Transfusions: An Analysis of the Legal Protections Afforded to Adults and Children in European/English Human Rights Contexts’ (2017) 24(4) Eur J Health L 368. 36  See e.g., St George’s Healthcare NHS Trust v S [1998] 3 WLR 936 (the England and Wales Court of Appeal awarded damages to a woman whose refusal of a caesarean section had not been respected); Re B (Adult: Refusal of Medical Treatment) [2002] 2 All ER 449 (the England and Wales High Court upheld the right of a 43-­year-­old woman with paralysis to have ventilation removed). 37  See e.g., the England and Wales Mental Health Act 1983, s. 63, which states that ‘the consent of patient shall not be required for any medical treatment given to him for the mental disorder from which he is suffering’. Contrast the Irish Mental Health Act 2001, s. 57(1), which requires the consent of ‘involuntary’ patients (i.e., patients who are formally admitted) for treatment unless the patient is incapable of consenting. 38  See, e.g., the England and Wales Public Health and Control of Diseases Act 1984, as amended by the Health and Social Care Act 2008. 39  [2009] IEHC 106.

486   Mary Donnelly and Jessica Berg guarded at all times. The High Court found that the woman’s ongoing detention was lawful under the relevant public health legislation (the Health Act 1947) for as long as she remained infectious, but there was no basis to treat her without her consent while she had capacity. As described by the judge, this meant that ‘the reality of the position is that the patient is going to die [in hospital] unless she consents to the treatment’.40 Second, human rights norms play an important role in European responses to issues around consent and capacity. All 47 of the CoE member states (as well as the European Union41) have ratified the CRPD. In addition, two CoE measures are instrumental in reifying the human rights approach across Europe. All CoE member states are party to the European Convention on Human Rights (ECHR), and many have formally incorporated the ECHR into domestic law. Articles 2, 3, and 8 of the ECHR have been especially significant in establishing human rights norms in respect of consent to healthcare. Article 2 protects the right to life and has been found relevant to the refusal of (life-­ saving) healthcare treatment. Specifically, the ECtHR has found that Article 2 requires that ‘a clear stance’ be taken to ‘remove the risk that the patient had made his decision without a full understanding of what was involved’. Thus, CoE member states are required to provide appropriate guidance regarding the conditions in which refusal of treatment is valid and binding on medical staff.42 Article 3 prohibits inhuman and degrading treatment. It has been accepted by the ECtHR that this can include medical treatment,43 although the Court has also found that treatment which is a ‘medical’ or ‘therapeutic’ necessity cannot be ‘inhuman or degrading’ under Article 3.44 It is important, however, that therapeutic necessity be convincingly established. Article 8 protects the right to ‘private and family life’. In the healthcare context, the ECtHR has found that this extends to protecting the right to physical integrity45 and the right of autonomy.46 Significantly, in respect of capacity, the ECtHR has recognised that the fact that a person lacks capacity does not, of itself, allow treatment to be imposed against the person’s will.47 Because the right to private and family life protected by Article 8 is not an absolute right, Article 8 does not prohibit the treatment of a person who lacks capacity against his or her will. It does, however, require that the 40 Id. 41  European Union ratification of the CRPD relates only to those aspects of the CRPD in which the EU has competence. 42 See Arskaya v Ukraine [2013] ECHR 1235. The medical professionals had accepted the applicant’s son’s refusal of medical treatment notwithstanding his extremely serious medical condition, his highly erratic behaviour, and the views of the hospital psychiatrist that he was extremely unstable. Although some attempts were made to talk to the patient, no formal attempts were made to address the question of capacity. 43 See Herczegfalvy v Austria (1992) 15 EHRR 437: the treatment here included forced feeding, isolation, and the forcible injection of sedatives involving the applicant being handcuffed and tied to a security bed, which led to broken ribs. 44  Herczegfalvy v Austria (1992) 15 EHRR 437, para 82. 45 See Glass v United Kingdom (2004) 29 EHRR 341. 46 See Pretty v United Kingdom (2002) 35 EHRR 1. 47 See Storck v Germany (2005) 43 EHRR 96.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   487 process whereby the treatment decision is made must be fair, that the person’s opinion should be sought, and that his or her will and preferences must be taken into account.48 The second relevant CoE human rights instrument is the Convention on Human Rights and Biomedicine (the Oviedo Convention),49 which has been ratified by most members states of the CoE.50 Article 5 of the Oviedo Convention states that ‘[a]n intervention in the health field may only be carried out after the person concerned has given free and informed consent to it’. The requirements where persons are not able to consent are set out in Article 6. Such an intervention may only be carried out for the person’s ‘direct benefit’ and only with the authorisation of the person’s representative or an authority, person, or body provided by law. Article 6 also requires that the person concerned must as far as possible take part in the authorisation procedure, while Article 9 requires that the previously expressed wishes relating to a medical intervention expressed by the patient must be taken into account. Finally, in most European jurisdictions, issues in respect of capacity and decision-­ making where a patient lacks capacity are dealt with in legislation. This means that the law in Europe tends to be codified and has a decidedly regulatory aspect. One of the richest sources for engagement with capacity has been the Mental Capacity Act 2005, which came into effect in England and Wales in 2007, and which has provided the basis for an extensive body of case law. Following ratification of the CRPD, several European jurisdictions have engaged in legislative reforms, with the aim of complying (or at least ensuring greater compliance) with the CRPD. In Norway, for example, the Storting (Norwegian Parliament) adopted a new Guardianship Act in 2010, which came into effect in 2013.51 One of the more dramatic efforts at law reform has been in Ireland, where the farcically outdated Lunacy Regulation (Ireland) Act of 1871 still remains the applicable legislation at the time of writing. The Assisted Decision-­ Making (Capacity) Act 2015 (ADMCA) which repeals the 1871 Act is expected to come into force in 2022.

3  Competence and Capacity Outsides of requests for ‘futile’ treatment (a topic beyond the scope of this chapter), competent adults/adults with capacity are given broad deference in their decisions 48 See Stanev v Bulgaria (2012) ECHR 46. 49  The Convention for the Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine (ETS no 164), opened for signature 4 April 1997. 50  Some notable nonratifying jurisdictions are Austria, Belgium, Germany, Ireland, Malta, Russia, and the United Kingdom. 51  UN Convention on the Rights of Persons with Disabilities: Norway’s Initial Report, available at https://www.regjeringen.no/contentassets/26633b70910a44049dc065af217cb201/crpd-­initial-­report-­ norway-­english-­01072015.pdf.

488   Mary Donnelly and Jessica Berg regarding medical treatment. That is not to say that individuals can demand any treatment they want, but if they have the ability to pay, the treatment in question is available, and there is a healthcare provider willing to provide the treatment, patient choice is often controlling. But determining competence/capacity raises a separate set of issues.

3.1  The United States ‘Competence’ is a legal construct, one determined by courts or, sometimes, by statute. Adults are presumed competent to make medical decisions absent evidence to the contrary. The presumption is reversed for those below the age of majority. Many states specify the standards for competence in their informed consent statutes. These standards usually involve one or more components, including the ability to make a choice, ability to understand information, ability to appreciate the abstractly understood information (sometimes referred to a ‘deep understanding’, or ability to apply the information to one’s own circumstance), and ability to reason.52 Although it is clear that a patient who lacks the ability to make a choice or fails to understand any information should not be permitted to make medical decisions, it is less clear what to do in the more common situations of patients who evidence some but not all abilities. Moreover, it is also not clear whether different standards should be applied to different types of decisions depending on, for example, the difficulty of the decision, the risks involved, or the severity of possible outcome (e.g., decisions about life-­sustaining treatment might require a higher level of competence than decisions about non–life-­sustaining treatment). States use one or more of the components but rarely provide guidance on the specific application of the standards. Unlike legal determinations of competence, capacity is determined by medical or mental health professionals. It is a much more flexible determination that can be made differently in different situations and at different times. For many patients, particularly those who are very ill, capacity can fluctuate, and thus it is best to use a less formal mechanism for allocating decision-­making authority than a judicial ruling of competence. Capacity and competence are linked, however, and, in theory, the basis for the capacity determination should be the ability of the individual to demonstrate the components required under the legal standard of competence. That is to say, if the legal standard is that to be competent a patient must both understand and appreciate the medical decision in question, the professional determining capacity should evaluate whether the patient understands the information provided as part of an informed consent disclosure and is able to apply that information to the particular situation (i.e., understand what it means for her). In fact, a variety of tests are used to determine capacity, including simple screening mechanisms like whether someone is oriented to time, place, and person. Furthermore, even when a state statute specifies the elements of competence, health professionals will 52  See Jessica W Berg, Paul S Appelbaum and Thomas Grisso, ‘Constructing Competence: Formulating Standards of Legal Competence to Make Medical Decisions’ (1996) 48(2) Rutgers L Rev 345.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   489 still have to consider how to apply the tests in question. Consider a state that requires understanding (which is very common). How much information must be understood in order for a person to be considered capable of making the medical decision in question—100%? 75%? 50%? In addition to such quantitative questions, there are qualitative issues. For example, what particular pieces of information must be understood? Does a patient making a decision about surgery, who understands all information but fails to comprehend that she will be unconscious during the administration of general anaesthesia, have capacity to make the decision? These are areas in which institutional policy can be extremely helpful, especially in the absence of legal guidelines. In theory, capacity determinations should be made for all patients making decisions; in practice a number of heuristics are employed. Thus patients who consent to treatment rarely have capacity challenged. This could be because the patient is making a decision in accord with the physician’s recommendation and thus, by definition, in his medical interests. Although patients who refuse treatment should not be considered automatically incompetent, one may want to further scrutinize refusals of medically recommended care. In addition, patients who are developmentally disabled, under the influence of drugs (legal or illegal), or suffer from severe mental illness are also more likely to be subjected to additional capacity evaluations. Moreover, healthcare providers are also likely to be more questioning of the capacity of patients who are making particularly weighty decisions such as end-­of-­life choices. Here again institutional policy could be helpful in bringing consistency to an area which is driven largely by the preferences of different healthcare providers. Individuals who are legally determined to be incompetent usually have a guardian or conservator appointed to make decisions. In some cases, the competence determination is global, and the guardian is authorised to make all decisions for the ward. In other situations, the individual is considered incompetent only with respect to a particular aspect of decision-­making. For example, someone may be incompetent to make financial decisions but not incompetent to make medical decisions, or vice versa. In each case, the guardian’s decision-­making authority will be clearly delineated. Moreover, once a guardian is appointed, only a court can remove the guardian’s decision-­making authority. Guardians can usually make all the same decisions that the individual could make, with a few exceptions in some states. These limitations usually fall into one of three categories: reproductive decisions, psychiatric treatment, or end-­of-­life decisions. In those cases, additional judicial determinations are needed (either to make the decision or to grant the guardian specific authority).

3.2 Europe The distinction between competence (legal) and capacity (medical/mental health) common in United States discourse is not a feature of European law. Instead, the two concepts tend to be elided, and English-­speaking jurisdictions typically refer simply to ‘capacity’. In general, capacity plays a gatekeeper role in Europe similar to that played in the United States.

490   Mary Donnelly and Jessica Berg The standard for capacity is often, though not inevitably, set down in legislation. The England and Wales Mental Capacity Act 2005 (MCA) provides a good example. The MCA adopts a functional test for capacity whereby capacity is assessed in respect of the particular decision to be made at the time the relevant decision must be made. Under the MCA, a person lacks capacity in relation to a matter if ‘at the material time he is unable to make a decision for himself in relation to the matter’.53 A person is unable to make a decision when that person is unable to understand the information relevant to the decision, retain that information, use or weigh that information as part of the proc­ ess of making the decision, or communicate the decision.54 The MCA includes a number of additional elements which are aimed at ensuring that a person is not found to lack capacity simply because he or she proposes to make an unwise or irrational decision. Thus, the MCA establishes a statutory presumption of capacity55 and expressly states that that person is not to be treated as unable to make a decision ‘merely’ because the person makes an unwise decision.56 The MCA also includes a ‘diagnostic threshold’ which requires that a person can only be found to lack capacity where the person’s inability to make a decision is because of ‘an impairment of, or a disturbance in the functioning of, the mind or brain’.57 It also states that a lack of capacity cannot be established ‘merely’ by reference to a person’s age or appearance or ‘a condition of his, or an aspect of his behaviour, which might lead others to make unjustified assumptions about his capacity’.58 The MCA also imposes some positive obligations on capacity assessors. These include the requirement that a person is not to be treated as unable to make a decision unless ‘all practicable steps’ to help the person to make the decision have been taken without success59 and the requirement is to provide the relevant information in an accessible way.60 A feature of the post-­CRPD legislative landscape has been an increased focus on the establishment of frameworks to provide decision-­making support to people with impaired capacity, thus facilitating them in reaching the standard for capacity to consent to healthcare decisions. A relatively long-­established form of supported decision-­ making is the Swedish Personal Ombudsman scheme, which has been operational since 2000.61 The potential impact of decision-­making support mechanisms on processes around capacity and consent to healthcare may be seen in the Irish Assisted Decision-­Making (Capacity) Act (ADMCA). Under the ADMCA, a person ‘whose capacity is in question or may shortly be in question’ (referred to as the ‘relevant person’) may appoint either a Decision-­Making Assistant62 or a Co-­Decision-­Maker63 to provide support in making 53  MCA, s. 2(1). 54  MCA, s. 3(1). 55  MCA, s. 1(2). 56  MCA, s. 1(4). The diagnostic threshold has been criticised as discriminatory on the grounds of disability and as not compatible with the CRPD: see Peter Bartlett, ‘The United Nations Convention on the Rights of Persons with Disabilities and Mental Health Law’ (2012) 75(5) Mod L Rev 752. 57  MCA, s. 2(1). 58  MCA, s. 2(3). 59  MCA, s. 1(3). 60  MCA, s. 3(2). 61  See Andrew Power, Janet Lord and Allison de Franco, Active Citizenship and Disability: Implementing the Personalisation of Support (Cambridge University Press 2013) ch 7. 62  ADMCA, s. 10. 63  ADMCA, s. 17.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   491 decisions (including decisions about healthcare treatment). Both forms of support require the supporter to assist the relevant person by obtaining and explaining information relevant to the decision, to ascertain the relevant person’s will and preferences and assist the relevant person in communicating them, and to endeavour to ensure that the relevant person’s relevant decisions are implemented.64 A Decision-­Making Assistant cannot make the decision on behalf of the relevant person,65 and the decision is taken to be the decision of the relevant person for all purposes.66 In contrast, the Co–Decision-­ Maker must reach the relevant decision jointly with the relevant person.67 As the ADMCA is not yet operational, it is not clear how many people will choose to avail themselves of these support frameworks. For those who do, the process for consent to a healthcare intervention will require a trialogue between clinician, relevant person, and the chosen supporter(s). Of course, this may well represent that way many medical professionals already deal with people with impaired capacity and their families. The difference, once the ADMCA comes into force, is that this will be now legally mandated where a decision-­making supporter is in place.

4  Advance Directives Once a determination is made, either by a medical professional that the patient lacks capacity or by a court that the patient is incompetent, how are medical decisions to be made for that person? One place to begin is with any instructions left by the patient while he or she was competent. Such instructions may be given orally or in writing. The legal framework for advance directives is primarily concerned with written instructions (although verbal instructions may still be highly significant when it comes to making decisions for the patient). While ‘advance directive’ is often used as a generic term, technically there are three legal forms, at least in the United States: the advance care document (which provides instructions for care), a living will (which is an older format and provides instructions if the individual is determined to be ‘terminal’), and durable powers of attorney for healthcare (which designate a decision-­maker and are now almost always incorporated into the prior two documents).

4.1  The United States All US states allow advance healthcare decision-­making in some format. Traditionally, living wills were documents that allowed patients who are terminal (historically defined as having less than 6 months to live, but now may either speak more broadly about those expected to die within a short time period regardless of treatment or may be silent on 64  ADMCA, s. 14(1) (DMA); s. 19(1) (CDM). 65  ADMCA, s. 14(2). 66  ADMCA, s. 14(3). 67  ADMCA, s. 19(1)(e).

492   Mary Donnelly and Jessica Berg the timeline) to provide instructions regarding their end-­of-­life care, including any requests to limit treatment should they become incompetent. Because the original living wills were focused on terminal patients, many states enacted statutes allowing for broader written ‘advance directives’, which could specify care in any of a number of different situations. Other states expanded traditional powers of attorney (which automatically expired in cases of incompetence) to create ‘durable’ powers of attorney that remain in effect and could cover healthcare situations in which the individual was incompetent, allowing patients to designate a decision-­maker and provide written instructions. In other cases where an older living will document was used, a court might remove (or ignore) the terminal requirement. It is most common now (and easier) simply to refer to all written instructions as ‘advance directives’, and, in fact, most states have a specific statute delineating the form for such instructions. The most significant limit on the helpfulness of advance directives is their scarcity. Few people draft a will addressing disposition of their property after death, and even fewer document their healthcare wishes in an advance directive. Even among populations likely to face significant healthcare decisions, such as critically ill or chronically ill persons, advance directives are still used in only a minority of the general population, although their uptake in certain populations is growing and in some subgroups—such as elderly patients, individuals with terminal illnesses, individuals with unusual end-­ of-­life preferences—may be much more widespread. The failure of large numbers of healthy people to complete an advance directive is not surprising as many people may feel uncomfortable talking about their end-­of-­life care preferences. Not only is it hard to talk about the issues, but it is also hard to articulate, in the abstract, what one’s preferences are. Those who draft an advance directive face other difficulties. One problem is locating the directive. Although it is sometimes present in a patient chart, at other times it is filed at home, with family, with attorneys, or with a primary care physician who may not be involved in hospital care. The increasing availability of electronic health records has helped in this area, but has not yet resolved the issue. Various states and organisations have proposed online databases for advance directive documents, but no one particular model has widespread adoption and all have limitations (e.g., how would you know the document was to be found on the database if there are multiple options?). A second and more significant problem is interpretation. Rarely do the forms address all potential healthcare decisions or with a high degree of specificity. As a result, someone must interpret the words used in the form and apply them to the particular healthcare decision at hand. A third problem is enforcement. Courts have been reluctant to burden healthcare providers with liability for failure to withhold treatment in accordance with the patient’s wishes. It may be that these cases look too much like ‘wrongful living’ causes of action in which courts have been loath to quantify the harm to an individual of an impaired life as opposed to no life at all. Alternatively, courts may be unwilling to create legal pressures on physicians to withhold medically beneficial treatment.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   493 Despite the limits, advance directives may play an important role in certain kinds of decisions—particularly for patients with unusual preferences or unusual proxy designations. They may also function to structure care discussions between healthcare providers and patients and between patients and families. Finally, written documentation of preferences may be extremely important in cases where proxies are making certain types of decisions, such as withholding food and fluids. A number of states now require specific written authorisation for proxies to make these kinds of decisions. For example, Ohio (like a number of other states) requires that the document designating a decision-­ maker state clearly that the individual can make decisions regarding withdrawing or withholding food or hydration.

4.2 Europe Legislation for advance healthcare planning was introduced later in European jurisdictions than in the United States. In 2009, the CoE issued a Recommendation setting out principles for the development of legal frameworks in this area,68 and most European states have now introduced some form of legislation.69 There is a wide variation in approach, from the relatively simplistic model adopted in the England and Wales MCA to the more complex model in the more recent Irish ADMCA (which has not yet come into force). A comparison of these two models gives a flavour of the importance of the detail of the legislation in framing legal models for advance healthcare decision-­making. Both the MCA and the ADMCA allow for legally enforceable advance refusals of treatment (including life-­sustaining treatment). However, there are some notable differences in approach, especially in terms of accessibility and enforceability.70 The MCA takes a minimalist approach. A person may either make a (legally binding) advance refusal of treatment,71 or he or she may utilize a lasting power of attorney to authorise someone to make personal welfare decisions (including consent to/refusing medical treatment) on his or her behalf.72 There are no statutory formalities (and no requirement for writing) for an advance refusal of treatment except where the refusal

68  Principles concerning continuing powers of attorney and advance directives for incapacity, Recommendation CM/Rec (2009) 11. 69  See Denard Veshi and Gerald Neitzke, ‘Advance Directives in Some Western European Countries: A Legal and Ethical Comparison Between Spain, France, Italy and Germany’ (2015) 22(4) Eur J Health L 321; D Veshi and G Neitzke, ‘Living Wills in Italy: Ethical and Comparative Law Approaches’ (2015) 22(1) Eur J Health L 38; Monica Navarro-­Michel, ‘Advance Directives: The Spanish Perspective’ (2005) 13(2) Med L Rev 137. 70  For a more detailed comparison, see Mary Donnelly, ‘Developing a Legal Framework for Advance Healthcare Planning: Comparing England & Wales and Ireland’ (2017) 24(1) Eur J Health L 67. 71  MCA, s. 24(1). 72  MCA, s. 11(7)(c). In order for a lasting power of attorney authority to extend to life-­sustaining treatment, it must specifically authorise this: MCA. s. 11(8)(a).

494   Mary Donnelly and Jessica Berg relates to life-­sustaining treatment.73 Thus, under the MCA, it is very easy to make an advance refusal of treatment. However, the MCA allows the advance refusal to be overridden in a wide range of circumstances, and an unwritten advance refusal is highly vulnerable if a dispute arises. The grounds for overriding an advance refusal are, first, where there are reasonable grounds for believing that circumstances exist which the person making the decision did not anticipate at the time of the decision and which would have affected the decision if the person had anticipated these circumstances74 and, second, where the decision-­ maker has done anything clearly inconsistent with the advance decision remaining his or her fixed decision.75 In the latter situation, there is no requirement that the decision-­ maker have capacity at the time of the inconsistent act and no indication of how inconsistency is to be measured. The result is that advance decisions made under the MCA can reasonably be regarded as ‘vulnerable to challenge’ in a way which ‘undermines their value as protection for precedent autonomy’.76 Perhaps reflecting this perception, as in the United States, take-­up on both advance refusals and health lasting powers of attorney has been limited.77 Data cited to the UK’s House of Lords Select Committee in 2014 indicated that only 3% of the public had made an advance refusal of treatment (notwithstanding that 82% of the public had indicated clear views about their end-­of-­ life preferences), and only 20% of lasting powers of attorney which were made related to health and welfare.78 The ADMCA offers a wider range of options and, in theory at least, a more robust and reliable framework. In an advance healthcare directive, the directive-­maker may both refuse and request treatment and may also appoint a Designated Healthcare Representative who has power to ensure that the terms of the advance healthcare directive are complied with.79 A valid and applicable refusal of treatment is legally binding,80 with a small number of exceptions.81 A request for treatment is not legally binding. However, the request must be ‘taken into consideration’ during any relevant decision-­ making process if the treatment specified is relevant to the directive-­maker’s medical condition. If the request is not complied with, the healthcare professional must record the reasons for not complying and give a copy of these to the person’s Designated Healthcare Representative.82 In this way, the ADMCA strives to protect clinical discretion

73  MCA, s. 25(5). In this case the decision must explicitly state in writing that it is to apply even if the decision-­maker’s life is at risk and must be signed by the decision-­maker or at his or her direction and by one witness: MCA, s. 25(6). 74  MCA, s. 25(4). 75  MCA, s. 25(2). 76  Alasdair Maclean, ‘Advance Directives and the Rocky Waters of Anticipatory Decision-­Making’ (2008) 16(1) Med L Rev 1, 16. 77  House of Lords Select Committee, Mental Capacity Act 2005: Post Legislative Scrutiny (The Stationery Office, 2014) para 193. 78 Id. 79  ADMCA, s. 88(1)(a). 80  ADMCA, s. 84(2). 81  These relate to basic care; mental health, where a person has been formally admitted under the Mental Health Act 2001, and pregnancy. 82  ADMCA, s. 84(3)(b).

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   495 while at the same time affording a degree of recognition to patient autonomy as a positive right to choice. The possibility of appointing a trusted person as Designated Healthcare Representative offers potential for a richer and more integrated approach to advance healthcare directives than that offered by the MCA.83 By placing the power of appointment in the hands of the relevant person, it is also more in line with the CRPD requirements. In addition to the general power to ensure that the terms of the advance healthcare directive are complied with, a Designated Healthcare Representative may be specifically empowered to advise on and interpret the directive-­maker’s will and preferences about treatment, including life-­sustaining treatment, based on the advance healthcare directive.84 Rather confusingly, however, the ADMCA also allows a person to appoint someone (an ‘attorney’) to make healthcare decisions under an enduring power of attorney.85 While the decisions covered may overlap, there is one notable difference in scope. An attorney may not make a decision relating to the refusal of life-­sustaining treatment.86 In line with its more expansive approach, the ADMCA imposes more demanding formalities in making an Advance Healthcare Directive. The Advance Healthcare Directive must be made in writing, signed, and witnessed, and it must include designated details.87 The aim of these demands is to reduce the possibility of carelessly made Advance Healthcare Directives. Notably however there is no statutory requirement to consult a healthcare professional in making an Advance Healthcare Directive. This means that it is still possible for someone to make an Advance Healthcare Directive without an adequate understanding of his or her medical situation.88 Possibly reflecting the greater normative force of more formalised Advance Healthcare Directives, the ADMCA restricts the circumstances in which an Advance Healthcare Directive may be overridden. This may happen only where an Advance Healthcare Directive is not valid or applicable. An Advance Healthcare Directive is not valid where it was not made voluntarily or where a person has done something clearly inconsistent with the Advance Healthcare Directive. However, unlike under the MCA, an inconsistent act only overrides the Advance Healthcare Directive if it takes place while the person has capacity.89 An Advance Healthcare Directive is not applicable where the directive-­maker has capacity, where the treatment in question is ‘not materially the same’ as the treatment set out in the Advance Healthcare Directive, or where the circumstances set out in the Advance Healthcare Directive are absent or are not materially the same as those arising.90 The ADMCA also requires that where there is ambiguity as regards the validity or 83  It avoids at least some of the ethically significant differences (in respect of information, evaluation, and safeguarding) between advance and contemporaneous healthcare decisions. 84  ADMCA, s. 88(1)(b). 85  ADMCA, s.59(1). 86  ADMCA, s. 62(5). 87  ADMCA s. 84. 88  Although whether a formal requirement for prior consultation would actually enhance understanding is open to debate: for a sceptical view, see Angela Fagerlin and Carl Schneider, ‘Enough: The Failure of the Living Will’ (2004) 34(2) Hastings Center Rep 30. 89  ADMCA, s. 85(1). 90  ADMCA, s. 85(2).

496   Mary Donnelly and Jessica Berg applicable of an Advance Healthcare Directive, the healthcare professional must consult with the directive-­maker’s Designated Healthcare Representative if there is one or, if there is none, the directive-­maker’s family and friends and seek the opinion of a second healthcare professional.91 Based on the experience in the United States (and England and Wales), it seems unlikely that, once operational, the ADMCA will lead to wide-­scale adoption of Advance Healthcare Directives. Nonetheless, as in the United States, this form of decision-­making can still play an important role for those people who chose to avail themselves of it. Based on a textual analysis, the more robust framework ADMCA model seems to offer a richer protection for autonomy than the more anaemic MCA framework, although whether this potential is delivered on in practice will depend on how the ADMCA model is operationalised.

5  Decisions Made for Others Even with an advance directive, a decision-­maker is usually needed. In the absence of an advance care document, identifying a decision-­maker is even more important.

5.1  The United States There are at least three types of legally recognised substitute decision-­makers for incompetent patients: guardians/conservators (court-­appointed), proxies (legally appointed through a proxy designation form by the patient or a decision-­making statute), and surrogates (informal decision-­makers identified by the treatment team). The ideal individual is an appointed proxy, chosen and authorised by the patient. By making a proxy designation while competent, the patient is able to indicate both the person whom she would like to have make decisions in her stead, as well as provide any specific instructions about the scope of the proxy’s power (e.g., clearly including the power to make end-­of-­life decisions or withdraw life-­sustaining treatments). As noted earlier, however, few people have recorded legal documents, and thus a less formal mechanism of surrogate decision-­makers is necessary. In these situations, a next-­of-­kin hierarchy may be used to identify the appropriate decision-­maker: spouse (or possibly a domestic partner), adult children, parents, adult siblings, and sometimes other relatives or a close friend. While extremely useful, the informal surrogate hierarchy can also be problematic as it may not differentiate between individuals of equal status (e.g., children or siblings) and also may not accommodate situations in which the person with legal authority is not actually the best person to make the decisions (e.g., an estranged child vs. a nephew 91  Id. If, after this consultation, the ambiguity remains, the healthcare professional is required to resolve the ambiguity in favour of the preservation of life.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   497 caretaker). Individuals lower down on the hierarchy (or outside the hierarchy) can go to court to challenge a decision-­maker. But these challenges can be difficult to win. More likely, where there is disagreement among potential decision-­makers, a court would take issue with the decision made by the designated surrogate, perhaps indicating that additional evidence is needed before proceeding, instead of removing the surrogate and replacing her with another person. But the latter does happen. Despite these limitations, it is widely understood that the informal next-­of-­kin hierarchy plays an important role in decisions for incompetent patients. For ease in the following discussion, we use the term ‘surrogate’ to cover all three types of decision-­ makers: guardians, proxies, and surrogates. Once a decision-­maker has been identified, the next question is how the decision should be made. There are two formal standards, but at least three decision-­making approaches; the appropriateness of each dependant on the available information. Where there is a clear written or verbal advance directive, the surrogate should effectuate the patient’s decision. The role in these cases is to ensure that the advance directive covers the circumstance at hand and to monitor the patient’s situation in case her medical state changes in such a way that would make the advance directive inapplicable. As noted earlier, these situations should be rare either because there are few written advance directives or because the ones that exist are unlikely to provide clear step-­by-­step instructions that can be easily applied to specific healthcare decisions. In most situations then, the surrogate is making a decision based on a variety of information about what the patient would have wanted. In these cases the ‘substituted judgment’ standard is applied, and the surrogate must figure out what the patient would have chosen should she have been competent. In addition to information gathered from written forms and verbal discussions, surrogates may also draw from knowledge about other healthcare decisions, reactions to treatment of other people, general personality, religious preferences, or any other pertinent information. Furthermore, surrogates should evaluate the information for its remoteness (how long ago), consistency (was this just one random comment?), thoughtfulness, maturity, and specificity. In the absence of adequate information about patient preferences, a ‘best interests’ standard should be used. The question here is what treatment decisions would be in the patient’s best interests, with ‘interests’ defined rather broadly and likely including more than simply best medical interests. Likewise, for patients who have never been competent, a best interests standard should be used. This can include children, adults with developmental disabilities, or perhaps even adults who fail to leave any real evidence of their preferences such that using a substituted judgment standard would be even more of a legal fiction than otherwise is the case. One of the most thoughtful court decisions in this area is In re Conroy where the court articulated a hierarchy of decision-­making standards: from the preferred subjective ‘substituted judgment’ to the ‘limited-­objective’ and ‘pure-­objective’ best interests tests.92 In the latter situations, according to the Conroy court, treatment could only be withheld if the pain and suffering of continued treatment were severe. While some 92  In re Conroy, 98 N.J. 321, 486 A.2d 1209 (1985).

498   Mary Donnelly and Jessica Berg courts have not accepted (or even explicitly rejected)93 the Conroy hierarchy or description of the two different best interests tests, all states use some kind of best-­interests standard in certain situations. The precise application of the standard in the context of withholding medically recommended treatment, however, is one where additional legal analysis is needed. Decisions by surrogates to withdraw life-­sustaining treatment from incompetent patients have resulted in the most legal cases and media attention in the United States. Decisions regarding withdrawing or withholding of respiration support, food, and hydration (water) have been of particular concern. One of the most important cases is the 1990 Supreme Court decision Cruzan v. Director, Missouri Department of Health.94 The case involved a young woman in a persistent vegetative state (i.e., having motor reflexes but no significant cognitive function and no hope for recovery). A plurality decision, it is most often cited for the propositions that patients have a federal Constitutional Fourteenth Amendment right to refuse life-­sustaining medical treatments, that the right includes refusal of artificial feeding and hydration, and that a surrogate may exercise the right for incompetent patients. In fact, the majority opinion only referred to the patient’s ‘liberty interest’ and specifically held that a state could put in place safeguards to ensure that a decision made by a surrogate for an incompetent patient was, in fact, a reflection of what that patient would have wanted. As a result, states can require a higher degree of evidence of an incompetent patient’s wishes in cases involving the withdrawal or withholding of life-­sustaining treatment, specifically ‘clear and convincing evidence’, rather than simply rely on the more common civil case stand­ ard of ‘preponderance of evidence’. On remand, the lower court determined that the evidence raised in the case did not meet the higher standard, although later evidence was then discovered, submitted to the court, and a ruling issued allowing treatment to be removed. A handful of state cases over the next few years focused on the type of evidence necessary to meet the clear and convincing standard. In the absence of sufficient evidence or a statute which authorizes families (or others) to make a decision in the absence of evidence, the default position is continued treatment. There are a variety of reasons for this, including concerns that the medical professionals’ role is to provide treatment and thus they should be encouraged to do so in the absence of clear indications of contrary patient wishes. Despite widespread acceptance of this default rule, there are some people who argue that we should reconsider and make the default position ‘minimizing suffering’, which is likewise compatible with medical professional obligations. Nonetheless, the legal default remains continued treatment. From a practical perspective, in the absence of disagreement between family members or between family members and the treatment team, withholding or withdrawal of treatment is highly unlikely to lead to either court involvement or legal liability. The cases that do make it into court arose in the context of a disagreement between family members—often the spouse on one side supporting treatment limitation and the 93  In re Martin, 450 Mich. 204 (1995). 94  Cruzan v Director, Missouri Department of Health, 497 US 261 (1990).

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   499 parents on the other supporting treatment continuation. Such was the case with Terri Schiavo, which drew international attention in the early 2000s.95 The Schiavo case involved a women who suffered a cardiac arrest at age 26 and subsequent severe, permanent brain damage. She was described as being in a ‘persistent vegetative state’ for many years. When her husband petitioned a court to remove her feeding tube, her parents challenged the decision. The case gained national attention for weeks while it made its way up and down the state and federal court systems and into the Florida Legislature and governor’s office. In the end, the tube was removed and Terri Schiavo died. A number of interest groups formed in her memory, which advocate on all sides of the debate.

5.2 Europe In 1999, the CoE adopted a recommendation on principles relating to adults lacking capacity.96 The recommended principles include respect for human rights and the maximisation of capacity. The latter principle requires that legal measures should not result automatically in a complete removal of legal capacity. In particular, the principles state that a person should not be automatically deprived of certain basic legal rights, including the right to consent to or refuse consent to any intervention in the health field when his or her capacity permits the person to do so. There is a wide variety of legal responses to decision-­making where a person lacks capacity. The most common response involves a judicial declaration of incapacity and the appointment of a guardian/curator/trustee.97 Notwithstanding the CoE recommendation (and the CRPD), in some jurisdictions, the person may be declared to lack capacity generally. In other jurisdictions, the declaration will relate to specific decisions or categories of decision only (e.g., a person might be declared to lack capacity to make business or financial decisions but retain the capacity to make personal welfare or healthcare decisions98). Where a guardian/curator has been appointed in respect of healthcare decisions, he or she will generally be authorised to provide consent to treatment on behalf of the person lacking capacity. Where a person lacks capacity but a guardian or equivalent has not (yet) been appointed, consent is typically given by a representative, often a next-­of-­kin. Depending on the applicable legal framework, the status of next-­of-­kin consent can be problematic. For example, in Ireland, next-­of-­kin

95  For a description of the case and resources, see https://msu.edu/course/hm/546/schiavo_case. htm; accessed 1 January 2018. 96  Recommendation no R (99) 4 of the Committee of Ministers to Member States concerning the Legal Protection of Incapable Adults. 97  See, e.g., the Spanish Civil Code, Article 222–285 (guardianship) and Article 286–294 (curatorship); the Netherlands Civil Code, Article 1: 450–462 (personal guardianship). 98  See, e.g., the appointment of a Decision-­Making Representative under the Irish ADMCA, s. 38.

500   Mary Donnelly and Jessica Berg consent is still widely relied on by healthcare providers notwithstanding that it has no legal basis whatever.99 The MCA (England and Wales) provides a notable contrast to the guardianship-­ based approach to healthcare decision-­making for adults lacking capacity. Although there is statutory provision for the appointment of a guardian in some limited situations,100 the authority of a guardian does not extend to giving consent to healthcare (outside of treatment for a mental disorder). This created a legal lacuna which became evident in the late 1980s and was exacerbated by the fact that the courts’ parens patriae jurisdiction had been removed by the Mental Health Act 1959. The difficulty was ultimately addressed in Re F: (An Adult: Sterilisation).101 Here, the House of Lords held that medical treatment which is in the best interests of a person lacking capacity could be provided on the basis of the common law doctrine of necessity so long as the proposed treatment was in the person’s best interests. It was up to the healthcare professional who proposed to provide the treatment to determine if the treatment was in the best interests of the person lacking capacity, although an application could be made to the court for a declaration on the matter. This position was put on a statutory basis in the MCA. This authorises the performance of an act in connection with the care or treatment of a person provided that the person performing the act has taken reasonable steps to establish whether the person being treated lacks capacity in relation to the matter and reasonably believes that the person lacks capacity and that the act is in the person’s best interests.102 In making a decision as to the person’s best interests, the healthcare professionals must take account ‘if it is practicable and appropriate to consult them’ of the views of anyone named by the person to be consulted, anyone engaging in caring for the person lacking capacity or with an interest in his or her welfare, or anyone with a power of attorney or any deputy appointed by the court to determine what would be in the person’s best interests.103 Although there is limited empirical evidence, it appears that there is consultation with family members, albeit to varying degrees.104 As in the United States, court involvement becomes necessary where there is a dispute between carers/family members and healthcare professionals105 or where a serious decision has to be made and there is doubt

99  See Mary Donnelly, ‘The Assisted Decision-­Making (Capacity) Act 2015: Implications for Healthcare Decision-­Making’ (2016) 22(2) Medico-­Legal J Irel 65, 66. 100  Under the Mental Health Act 1983, s. 7 as amended by Mental Health Act 2007, a guardian may be appointed to a person with a ‘mental disorder’ in order to enable the person to live in the community and receive care and treatment for his or her mental disorder in this setting. 101  [1990] 2 AC 1, 74–77. 102  MCA, s. 5(1). 103  MCA, s. 4(7). 104  See, e.g., Simon Halliday, Celia Kitzinger and Jenny Kitzinger, ‘Law in Everyday Life and Death: A Socio-­Legal Study of Chronic Disorders of Consciousness’ (2015) 35(1) Legal Studies 55; Gill Livingston et al, ‘Making Decisions for People with Dementia who Lack Capacity: Qualitative Study of Family Carers in UK’ (2010) 341 BMJ c4184. 105  See, e.g., Abertawe Bro Morgannwg University Local Health Board v RY [2017] EWCOP 2.

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   501 (possibly because there are no carers/family members to consult) about what is in the patient’s best interests.106 The other notable feature of the MCA in a contemporary context is its move away from an ‘objective’ understanding of best interests. Two aspects of the MCA provide the statutory basis for this move. First, section 4(4) requires that any person making a decision for a person lacking capacity must ‘so far as reasonably practicable permit and encourage the person to participate . . . as fully as possible in any act done for him and any decision affecting him’. Second, section 4(6) requires the decision-­maker to consider ‘so far as is reasonably ascertainable’ (a) the person’s past and present wishes and feelings (and, in particular, any relevant written statement made by him when he had capacity), (b) the beliefs and values that would be likely to influence his decision if he had capacity, and (c) the other factors that he would be likely to consider if he were able to do so. This recognition of the subjectivity of the person deemed to lack capacity reflects elements of the US substituted judgment approach as well as emerging human rights norms. Although it took some time for the more subjective approach to become established,107 the wishes and feelings of the person lacking capacity have now assumed a central role in judicial approaches to healthcare decision-­making. The statutory focus on wishes and feelings was reinforced by the UK Supreme Court in Aintree University Hospitals NHS Trust v James.108 In considering an application for a declaration permitting the removal of various life-­sustaining treatments from a man in a minimally conscious state, Lady Hale (who, as a Law Commissioner, had played a leading role in the formulation of the MCA) stated [I]in considering the best interests of this particular patient at this particular time, decision-­makers must . . . try and put themselves in the place of the individual patient and ask what his attitude to the treatment is or would be likely to be; and they must consult others who are looking after him or interested in his welfare, in particular for their view of what his attitude would be.109

Lady Hale recognised that the patient’s wishes need not always prevail ‘any more than those of a fully capable patient must prevail’.110 However, she was clear that ‘insofar as it is possible to ascertain the patient’s wishes and feelings, his beliefs and values or the 106  This approach coheres with the guidance in the MCA Code of Practice (2007) para 8.18. The Code also affirms the need for Court of Protection approval in cases involving nontherapeutic sterilisation or organ or bone marrow donation by a person lacking capacity, although these kinds of cases have not tended to come before the Court of Protection. The requirement for mandatory court approval for removal of life-­sustaining treatment from people in a persistent vegetative state or a minimally conscious state was removed by the UK Supreme Court in An NHS Trust and Others v Y [2018] UKSC 46. 107  See e.g., DH v NHS Foundation Trust [2010] EWHC 1217 (Fam); AVS v A NHS Foundation Trust AVS v A NHS Foundation Trust [2010] EWHC 2746 (COP). 108  [2013] UKSC 67 [45]. 109  Id. [39]. 110  Id. [45].

502   Mary Donnelly and Jessica Berg things which were important to him, it is those which should be taken into account because they are a component in making the choice which is right for him as an individual human being’.111 The greater emphasis on the person’s wishes and feelings in respect of healthcare decision is best seen in the England and Wales Court of Protection decision in Wye Valley NHS Trust v Mr B.112 Mr B was 73 years old, fiercely independent, with a long history of mental illness. Because of poorly controlled type 2 diabetes, he had developed a chronic foot ulcer which had begun to putrefy. Medical evidence was that, without an amputation, he would succumb to overwhelming infection within a matter of days and would die. With the amputation, he would have to move to a nursing home and would have a life expectancy of approximately 3 years. Mr B refused to consent to the amputation. He stated very clearly that he knew the seriousness of the decision, that he was not afraid of dying, and that he did not wish to move to a nursing home (his partner had died in one). Peter Jackson J found that Mr B lacked the capacity to make this decision. However, drawing on the ECHR, he also found that ‘a conclusion that a person lacks decision-­ making capacity is not an “off-­switch” for his rights and freedoms’.113 It was, he said, to state the obvious that ‘the wishes and feelings, beliefs and values of people with a mental disability are as important to them as they are to anyone else, and may even be more important’.114 Given the seriousness of the decision, Peter Jackson J felt it necessary to meet Mr B and he travelled to the hospital where Mr B was a patient. From this meeting, he ‘obtained a deeper understanding of Mr B’s personality and view of the world’,115 and he concluded that it would not be in Mr B’s best interests ‘to take away his little remaining independence and dignity in order to replace it with a future for which he understandably has no appetite and which could only be achieved after a traumatic and uncertain struggle that he and no one else would have to endure’.116 There was, he said, ‘a difference between fighting on someone’s behalf and just fighting them’, and forcing this unwanted treatment on Mr B would constitute the latter.117

6 Conclusion Both European and US law regarding informed consent, capacity, and autonomy continues to evolve. In both contexts, there is a wide variety of approaches. While this is unsurprising in Europe, which comprises clearly distinct legal systems, it is also a feature of the law in the United States. The very nature of the US legal system, which is 111  Id. [45]. 112  [2015] EWCOP 60. For other examples of a judicial focus on wishes and feelings, see A London Local Authority v JH and MH [2011] EWCOP 2420; HT v CK and ALA [2012] EWCOP 4160; Re M (Best Interests: Deprivation of Liberty) [2013] EWCOP 3456; Westminster City Council v Manuela Sykes [2014] EWCOP B9; M v N [2015] EWCOP 76. 113  Id. [11]. 114  Id. [11]. 115  Id. [18]. 116  Id. [45]. 117  Id. [45].

Patient Autonomy, Capacity, and Consent: Vulnerable Adults   503 distributed through the 50 states, lends itself to a variable system in which, paradoxically, some decisions may be permitted in one state but not another, or even more oddly that a person could be considered competent in one state and not another. At the same time, different European jurisdictions and different states in the United States may be considered to be ‘laboratories of innovation’,118 allowing new approaches to autonomy and decision-­making to be developed and tested before widespread adoption. Most healthcare decisions are made in homes, hospitals, and clinics. While this happens against the legal backdrop outlined in this chapter, it does so without the need for court involvement. However, sometimes, most likely in situations of disagreement (whether between the patient or his or her representative and the medical professional/s or between medical professionals), the law has to become more overtly involved. In the United States, while many aspects of the law in this area are considered settled, there are still numerous issues which need to be resolved, and high-­profile cases are likely to arise—such as the Schiavo case or the Charlie Gard case (discussed in the chapter by Jessica Berg and Emma Cave, in this volume)—which will garner international attention and commentary. In Europe also the increasing role of human rights in healthcare decision-­making suggests that court involvement may well be an increasing feature of the law. This comparison between the approaches used in the United States and Europe may provide additional guidance as we navigate these fundamental issues.

118  Supreme Court Justice Louis Brandeis described the states as ‘laboratories of democracy’ in New State Ice Co. v Liebman, 285 US 262 (1932).

chapter 25

I n troduction to Prov ider Con flicts of I n ter est i n Hea lth  Ca r e Sunčana Roksandić and Richard S. Saver

According to medical ethics and professional health community norms, a healthcare provider must maintain zealous, undivided loyalty to the patient.1 However, this idealized view does not fairly comport with the realities of medical practice and the complexity of the medical profession. Healthcare providers and institutions typically balance an array of competing interests, such as economic gain, the desire to favor colleagues, to advance in one’s academic or professional career, or the needs of other patients, and these competing interests can undermine professional judgment and duties to patients. Conflicts of interest are an omnipresent, long-­standing obstacle in the objective rendering of healthcare worldwide. As summarized by a well- known quote from George Bernard Shaw’s introduction to his 1906 play, The Doctor’s Dilemma, And what other men dare pretend to be impartial where they have a strong pecuniary interest on one side? Nobody supposes that doctors are less virtuous than judges; but a judge whose salary and reputation depended on whether the verdict was for plaintiff or defendant, prosecutor or prisoner, would be as little trusted as a general in the pay of the enemy.2 1  See, e.g., American Medical Association, Principles of Medical Ethics § VIII (“A physician shall, while caring for a patient, regard responsibility to the patient as paramount”), https://www. ama-­assn.org/delivering-­care/ama-­principles-­medical-­ethics 2  George Bernard Shaw, Preface, The doctor’s dilemma: A tragedy (Baltimore, MD: Penguin, 1988) as cited in Marc A. Rodwin, Conflicts of Interest and the Future of Medicine: The United States, France, and Japan, Published to Oxford Scholarship Online: May 2011, Introduction, p. 1.

506   Sunčana Roksandić and Richard S. Saver There are varying views as to what constitutes a true provider conflict of interest versus merely the potential for bias or a misaligned incentive. Conflict of interest is considered as a risk factor that increases the probability of occurrence of misconduct or acting contrary to the principle of good faith. Therefore, the problem is not conflict of interest per se, but the biased judgment that can follow. Inevitably, a conflict of interest “provokes a combination of stimuli and motivators that can have harmful effects through their influence on judgments and actions.”3 A conflict of interest can also occur when there is a decision that must be made by a physician or a duty that has to be followed that has more value or wider resonance in the individual case than loyalty to the primary patient, such as, for instance, the interests of a primary patient’s data privacy in the context of the interests of another patient or a threat to public health. A key element common to most interpretations of conflicts of interest is the provider’s exposure to undue influence from a secondary interest. According to one widely used European Union (EU) definition, a conflict of interest refers to a situation where the impartiality and objectivity of a decision, opinion, or recommendation of a subject is or might be perceived as being compromised by a personal interest held or entrusted to a given individual.4 A healthcare provider’s primary interests include caring for patients, engaging in research, and supporting medical education. An influential US Institute of Medicine (IOM) report defines a conflict of interest as “circumstances that create a risk that professional judgment or actions regarding a primary interest will be unduly influenced by a secondary interest,” such as financial gain or other personal, professional, or political gain.5 It is important, however, to recognize that some of these conflicting secondary interests are not per se problematic but appropriate and important. A second key element is that a conflict of interest can occur when there is merely the perception of undue influence by a secondary interest. Perception that a healthcare provider’s independence has been compromised leads to reputational risk and undermines the trust of other stakeholders. Indeed, all conflicts of interest “involve perceptions or appearances because they are specified from the perspective of people who do not have sufficient information with which to assess the actual motives of a decision maker and the effects of those motives on the decisions themselves.”6 Similarly, the European Charter of Medical Ethics from 2011, Principles 11 and 12, provide the following: 3  Mental Health Europe, Shedding Light on Transperent Cooperation in Healthcare: The Way Forward for Sunshine and Transperancy Laws Across Europe, 2019, p. 11. 4  This definition refers to the acts of the EU Decentralized Agencies as provided in the Guidlince for Conflict-­of-­Interest Policies in EU Decentralized Agencies, 10 December 2013, p. 5. Compare to T. L. Beauchamp & J. F. Childress, Principles of Bioemdical Ethics, Oxford University Press, 2013, p. 328. 5  Institute of Medicine of the National Academies, Conflict of Interest in Medical Research, Education, and Practice 6 (Bernard Lo & Marilyn J. Field, eds., 2009). 6  Conflict of interest and medical innovation, Ensuring Integrity While Facilitating Innovation in Medial Research, S. H. Beachy, A. C. Berger, and S. Olson, Rapporteurs, Roundtable on Translating Genomic-­Based Research for Health Board on Health Sciences Policies, IOM, The National Academies Press, Washington, DC, 2014, p. 6, referring here to the IOM observer.

Introduction to Provider Conflicts of Interest in Health care   507 The physician acting as a practitioner towards a patient, or as an expert or member of an institution, must ensure the greatest transparency in what might appear to be a conflict of interest and act in full moral and technical independence. If the moral and technical conditions are such as to prevent the physician from acting in full independence, he must inform the patient of this. The patient’s right to treatment must be guaranteed.7

Which secondary interests wield undue influence over providers? Much attention ­typically focuses on financial conflicts of interest. There is increasing recognition that various reimbursement systems can create troubling economic incentives for healthcare providers to withhold important care (as with controversies over managed care and global budget systems). Yet other reimbursement systems can create opposite incentives to provide wasteful and unnecessary care (as with fee-­for-­service systems). Also, the drug and device industry has partnered with healthcare providers in developing and commercializing new medical products, creating a confusing array of potentially conflicting financial relationships ranging from consulting agreements to royalty and licensing arrangements. Pharmaceutical companies “employ a wide range of promotional activities to increase medicines’ prescription rates and maximize their profit.”8 Financial conflicts also can negatively impact system-­wide decision-­making, such as when providers with financial conflicts work on formulary committees for healthcare institutions or serve on advisory bodies to governmental agencies making regulatory decisions about medical products. Furthermore, according to the latest IOM report on corruption in healthcare in the EU, double practice by providers in public and private clinics is an important issue implicating conflict of interest issues “but is not acknowledged as being a problem by many Member States.”9 A related concern is the referral of patients to medical facilities or services that physicians own or in which they have a financial investment. Despite the large attention paid to financial conflicts, nonfinancial secondary interests can also unduly influence healthcare decision-­making. Apart from direct economic gain, providers’ professional judgment may be swayed by prospects of professional reputation, honors and prestige, political predispositions, institutional loyalty, and social relationships. Notwithstanding their likely powerful influence, nonfinancial interests have not attracted the equivalent amount of regulatory attention as have financial ties. Conflicts of interest pervade most healthcare systems and pose considerable risks, both systemic and patient-­focused, including increasing costs, harming patients, limiting choice, biasing publication decisions, and eroding trust in healthcare providers and institutions. It should be kept in mind when considering the risks that can occur as a 7  European Council of Medical Orders, European Charter of Medical Ethics, 2011, http://www. ceom-­ecmo.eu/sites/default/files/documents/en-­european_medical_ethics_charter-­adopted_ in_kos.pdf 8  A. la Santos, The Sun Shines on Europe, Transparency of Financial Relationships in the Healthcare sector, Health Action Int’l. (May 2017). 9  Id., p. 149.

508   Sunčana Roksandić and Richard S. Saver consequence of conflicts of interest that “a key reason that relationships between industry and physicians are so pervasive is that physicians believe that they are invulnerable to undue influence from industry.”10 Relatedly, and making matters even more complex, a phenomenon called “bias bias” contributes to conflicts of interest situations: “while most people are ready to accept the possibility of bias in others, few are ready to acknowledge that they themselves might be biased.”11 The legal obligation for providers to avoid or minimize certain conflicts of interest may seem self-­evident. But such a rule is not always firm and uniform and, indeed, arises from a confusing variety of possible overlapping sources, both doctrinal case law governing ordinary medical care and statutory/regulatory law. First, the special nature of the doctor–patient relationship, which is seen as fiduciary or quasi-­fiduciary in many legal regimes, imposes obligations on the physician, acting as a fiduciary, to make the patient’s interests primary. Acting as a fiduciary, and in the patient’s best interests, may require the physician to avoid or, at least disclose, certain interests secondary to the patient’s direct welfare. In many European countries, especially those with legal systems based on continental/civil law, the main principle of acting in good faith demands the same. Second, the physician’s obligation to obtain the patient’s informed consent for treatment or research may require that the physician disclose material conflicts related to the physician’s recommendations or the study opportunity.12 Third, lawmakers have partially responded to concerns about conflicts of interest by imposing additional statutory and regulatory rules to address particular secondary interests. For example, referrals by physicians to facilities with which they have certain financial relationships may be statutorily prohibited. Fourth, conflict of interest rules may come into play, albeit more indirectly, in tort liability suits alleging substandard care. A provider’s disregard of professional ethics rules counseling avoidance of certain conflicts can be used as evidence to support malpractice claims of potential breaches of the standard of care. Fifth, anticorruption laws and criminal regulations could be invoked as well, such as medical malpractice offences/unconscientious rendering of healthcare and/or passive bribery. Finally, in recent decades, formal codifications of and instruction in professional morality have increased, not only through amending codes of medical ethics, but also through “codes of research ethics, corporate policies of bioethics, institutional guidelines governing conflict of interest, and the reports and recommendations of public commissions.”13 10 P. A. Ubel, Commentary: How Did We Get into This Mess? in Conflicts of Interest, Challenges and Solutions in Business, Law, Medicine and Public Policy (D. A. Moor, D. M. Cain, G. Loewenstein, & M. H. Bazerman, eds. 2005), Cambridge University Press, pp. 142–152, at 143, 145. 11  Z. Sharek, R. E. Schoen, & G. Lowenstein, Bias in the Evaluation of Conflict of Interest Policies, 40 J. L. Med 6 Ethics 368 (2012), pp. 368–382, at 368. 12  See more on EU regulation in Chapter 11. Informed Consent, Garwood-­Gowers A. and Olsena, S. in European health law, den Exter, A (ed.), Maklu, 2017., pp. 245–272, at 257 13  T. L. Beauchamp & J. F. Childress, Principles of Bioemdical Ethics, Oxford University Press (2013), p. 7.

Introduction to Provider Conflicts of Interest in Health care   509 The law’s typical way of addressing provider conflicts of interest has been through disclosure. Disclosure obligations can arise, for example, via fiduciary and informed consent doctrines. While disclosure may seem a relatively straightforward response to conflicts, the law has struggled at times with articulating precisely which conflicts require disclosure because some secondary interests may seem too extrinsic to the core of the doctor–patient relationship to require discussion with the patient. Some courts have found, for example, that the way a provider is paid, including economic incentives for practicing cost-­effectively, is beyond the scope of ordinary disclosure. In recent years, several countries have enacted statutory laws to increase the transparency of financial relationships between the healthcare industry and providers. In the United States, the Physician Payments Sunshine Act is a broad statutory disclosure regime in which payments between the drug and device industry and physicians and hospitals are disclosed through a publicly available database. EU laws regulating pharmaceutical promotion do not include provisions like those found in the US Sunshine Act, but some Member States, such as France, Belgium, Portugal, Romania, Slovakia, and Denmark, have adopted these types of rules in their national legislative framework.14 Some European countries have “partial sunshine provisions”—legal obligations are included in more general legal frameworks or only certain obligations are regulated by certain stakeholders.15 EU Directive 2001/83/EC, known as the EU’s Pharmaceutical Code, regulates all stages of a medicinal product’s marketing, including its advertising. According to this Directive, medicine advertising should be subject to stringent conditions and effective monitoring because health professionals must be able to perform their functions objectively, without being influenced by direct or indirect financial incentives from the industry. However, it is up to EU Member States to take all measures necessary to this end in light of their own individual context. More generally, consensus exists that a “need of transparency”16 is apparent as key international health organizations, including the World Health Organization, World Medical Association, and the Council for International Organizations of Medical Sciences, have emphasized the importance of disclosing conflicts of interest. But the law’s heavy reliance on disclosure has been criticized. Among other reasons, disclosure may be an incomplete remedy for managing conflicts, conflicts disclosure may have limited sway over patient decision-­making and can lead to unintended effects, and disclosing conflicts in a manner that is material, meaningful, and provides ­necessary context but is also comprehensible and accessible to ordinary patients has 14  In other EU countries, disclosure is primarily managed thorugh self-­regulation but also prescribed by special laws that govern the heath profession or by application of “acting in good faith” principle as found in laws governing obligations. Also, in anticorruption legislation on the national level, including Criminal Codes, disclosure could be managed by prescribing sanctions for opposite behavior, especially if negative consequence occured to the primary patient. Additionally, an important role is played by the European Federation of Pharmaceutical Industry Associations, op.cit., whose Disclocure Code was adopted in 2013. 15  See more details in Mental Health Europe, Shedding Light, supra 3, at 26–29. 16  P. D. Young, D. Xie, & H. Schmidt, Toward Patient-­Centred Conflicts of Interest Policy, 7 Int'l. J. Health Pol. Mgmt. 2 (2018), pp. 112–119, at 112.

510   Sunčana Roksandić and Richard S. Saver been pragmatically difficult. Beyond disclosure, some conflicted transactions may be prohibited altogether under varying legal rules, but strict avoidance requirements have tended to play a more limited, secondary role in conflicts regulation. Regulation of provider conflicts of interest has proved particularly challenging for health law. First, not all conflicts are easily avoidable or disclaimed. All methods of paying physicians involve a conflict of interest—either an incentive to do too much or an incentive not to do enough. Moreover, the mere existence of a conflict does not necessarily indicate a problem. Physicians play an important role as stewards of society’s limited healthcare resources, and that can come into conflict with the needs of an individual patient. Second, limited empirical data exist on the effects of conflicts of healthcare on medical decision-­making. Third, health law struggles with how to accommodate other legitimate interests beyond singular loyalty to the patient, such as advancing needed biomedical research or generating efficiencies in healthcare delivery. The danger is that conflicts regulation may over-­deter, thus chilling important secondary incentives that are needed to advance important societal interests regarding healthcare. Fourth, there are fairness considerations. Many healthcare providers are reluctant to view their own secondary interests, such as reimbursement incentives or professional prestige, as problematic when their overt intent is to help patients. They believe that they still can act in the interests of their patients even if also advancing secondary interests. They may perceive conflicts regulation as heavy-­handed, which can encourage disregard and noncompliance. Fifth, because conflicts of interest are now pervasive, a zero-­tolerance legal regime may be too costly and impractical. Yet locating the proper boundary between which conflicts to regulate and which to permit remains difficult. The pervasiveness of conflicts further raises identification problems. Correctly tracking particular financial conflicts and which parties are at risk of undue influence is becoming increasingly arduous as the underlying business arrangements in the healthcare system become more complicated. Sixth, once a conflict has been identified, law often provides ­incomplete guidance about what to do about it. Between disclosure, which may be an incomplete remedy, and prohibition, which may lead to overdeterrence, law has not provided clear and consistent standards for devising intermediate conflict management strategies. In the following chapters on provider conflicts of interest in healthcare, we will discuss important principles essential to understanding conflicts of interest regulation in different legal systems.



1. Conflicts of interest are generally viewed as interests secondary to the healthcare provider’s core role and obligations that may unduly influence professional judgment. Perceptions that secondary interests are exerting undue influence on a provider can also constitute a conflict of interest. But this leaves considerable room for interpretation and debate, such as how broadly conflicts regulation should extend to certain secondary interests, including nonfinancial interests. 2. The obligation for healthcare providers to avoid or minimize conflicts of interest arises from a variety of legal sources, including informed consent and fiduciary

Introduction to Provider Conflicts of Interest in Health care   511



duty doctrine or acting according to good faith, as well as statutory/regulatory law, but the rules are not always uniform and can be very context-­specific, making enforcement somewhat haphazard. Additionally, enforcement of conflicts of interest rules becomes difficult for the further reason that it depends to a large degree on self-­regulation: patients and regulators may be unaware of a provider’s conflicts in order to raise timely challenges. 3. The law heavily relies on disclosure, for better or worse, as a regulatory tool for addressing conflicts, and this approach often remains incomplete. 4. Competing secondary interests are now pervasive, but the mere existence of a conflict does not always indicate a problem. Conflicts regulation should not be viewed as a zero-­sum game. Indeed, some secondary interests are inevitable, legitimate, and necessary. The law struggles to find optimal balance in addressing conflicts that pose real risk of bias while not overly deterring important secondary incentives that advance larger societal goals beyond the immediate patient’s welfare.

chapter 26

Prov ider Con flicts of I n ter est i n US Hea lthca r e The Intractable Regulatory Challenge Richard S. Saver

Physicians in the United States command significant public trust.1 Their high standing arises in part because of expectations, underscored by medical ethics, that physicians act with zealous, undivided loyalty to their patients.2 However, the independent, unfailingly devoted doctor increasingly seems a relic of the past. Physicians practice within a turbulent US healthcare system featuring expanding complexity and commercialization. Potential conflicts of interest (COIs) abound, including financial arrangements with drug companies and third-­party payers, that threaten to undermine clinical judgment and steadfast devotion to the patient. How is health law responding? This chapter examines the role of law in addressing provider COIs within US healthcare and why, dismayingly, optimal regulation proves such an intractable challenge. Current rules regulating COIs are limited in application and primarily rely on disclosure, an insufficient management tool. The path to more optimal regulation remains fraught with difficulties. There is no consensus about which sources of influence should be treated as COIs, not all conflicts are easily avoidable, and, for certain conflicts, appropriate management techniques remain elusive. Moreover, health law continues to struggle with the feasibility and desirability of commanding 1  Megan Brennan, Nurses Keep Health Lead as Most Honest, Ethical Profession, Gallup, Dec. 26, 2017, http://news.gallup.com/poll/224639/nurses-­keep-­healthy-­lead-­honest-­ethical-­profession.aspx?g_ source=CATEGORY_SOCIAL_POLICY_ISSUES&g_medium=topic&g_campaign=tiles 2  See, e.g., Amer. Med. Ass’n, Principles of Medical Ethics § VIII (“A physician shall, while caring for a patient, regard responsibility to the patient as paramount.”), https://www.ama-­assn.org/ delivering-­care/ama-­principles-­medical-­ethics

514   Richard S. Saver healthcare providers’ singular loyalty to patient welfare as it tries to accommodate other legitimate interests, such as using economic incentives to generate efficiencies in healthcare delivery. Section 1 of this chapter reviews definitional issues about COIs and their associated risks and prevalence, and it considers ongoing debates over whether to construe and regulate certain secondary interests as conflicting. Section 2 analyzes the legal background, exploring how the obligation to avoid or manage COIs is not always clear and uniform under US law; instead, it arises from a confusing patchwork of overlapping legal sources, including common law doctrines and regulatory provisions, that are often limited in application or impose difficult procedural and proof burdens. This section, however, does not consider potential application of the US healthcare fraud and abuse laws to COI transactions, because fraud and abuse regulation is addressed in a separate chapter of this volume.3 Section 3 considers how the typical legal response to COIs has been required disclosure, but disclosure proves a necessary yet insufficient response with potentially unintended results. Section 4 explains why regulation of provider COIs remains particularly challenging for US health law, including limited empirical evidence about the effect of COIs on healthcare, difficulties in developing proportionate responses, and uncertainty in how to address secondary interests that advance important societal goals beyond the immediate patient’s welfare.

1  Background on Conflicts of Interest 1.1  Definitional Issues Under leading US interpretations, a COI exists when secondary considerations, including provider self-­interest, threaten to interfere with the healthcare provider’s primary obligations. These primary interests include caring for patients, engaging in research, and supporting medical education. A highly influential 2009 Report by the Institute of Medicine defines a COI as “circumstances that create a risk that professional judgment or actions regarding a primary interest will be unduly influenced by a secondary interest.”4 A second key element of COIs is that mere perception of undue influence can create real conflict concerns because “the confidence of others in [provider] judgment is of paramount importance. . . . When trust in [provider] judgment is impaired, the role of [providers] is diminished.”5

3  See the chapter by Krause in this volume. 4  Inst. of Med. of the Nat’l Acads., Conflict of Interest in Medical Research, Education, And Practice 6 (Bernard Lo & Marilyn J. Field, eds., 2009) [hereinafter IOM Conflict of Interest Report]. 5 Harvey V. Fineberg, Conflict of Interest: Why Does it Matter? 317 JAMA 1717 (2017).

Provider Conflicts of Interest in US Healthcare   515 The typical COI involves potential financial gain. Financial conflicts can arise in many different contexts, ranging from how providers are reimbursed to financial ties between providers and drug and device manufacturers in the development and commercialization of new medical products. Financial COIs can also negatively impact system-­wide decision-­making, such as when providers with financial ties to industry serve on formulary committees for healthcare institutions or on advisory bodies to governmental agencies. Other traditionally recognized COIs include the dual loyalty or conflicts of commitment situation. A physician may have competing loyalties because she occupies more than one role. For example, the physician investigator has a conflict of commitment when her patient enrolls in a clinical trial involving the physician. Providing individually tailored treatment may conflict with the physician-­investigator’s obligations to follow the trial protocol faithfully and generate valid, replicable research results.6 More controversial is whether nonfinancial interests are suitable for equivalent legal regulation in the US as financial COIs. Desire for career advancement, professional reputation, and prestige, even when removed from the prospect of immediate financial gain, can unduly influence provider decision-­making. Likewise, social interdependence with colleagues, including both collaborative and competitive relationships, and allegiance to particular institutions, can drive physician referral decisions. Other nonfinancial interests that raise COI concerns include intellectual, political, and ideological predispositions. A study, for example, found that physicians’ political affiliations as Republican or Democrat correlated highly with their treatment recommendations on socially contested medical care issues such as abortion and medical marijuana.7 Some commentators caution that nonfinancial interests should remain beyond the reach of ordinary COI regulation. Financial ties are thought more suitable for management because economic gain is supposedly tangible, objective, discretionary, and avoidable.8 Other commentators maintain that a COI has traditionally been limited in US conceptions, reflecting its origins in fiduciary law, to direct conflict between a secondary interest and the physician’s loyalty to a party or a particular role-­conceived mission.9 Expanding COI oversight more broadly to any source of potential bias can lead to overly burdensome regulation and may dilute necessary attention to financial conflicts.10 Yet several of these concerns about COI creep seem misplaced. First, the supposed tangibility/nontangibility distinction between financial and nonfinancial interests is not uniform. Some nonfinancial interests can be readily identified, such as published work 6  Miriam Wiersma, et al., Should We Try To Manage Non-­Financial Interests?, 361 Brit. Med. J. k1240 (2018). 7  Eitan D. Hiersh, et al., Democratic and Republican Physicians Provide Different Care on Highly Politicized Health Issues, 113 Proc. Nat’l Acad. Sci. USA. 11811 (2016). 8  See, e.g., Inst. of Med. of the Nat’l Acads., Conflict of Interest in Medical Research, Education, And Practice 47 (Bernard Lo & Marilyn J. Field, eds., 2009). 9  See, e.g., Marc A. Rodwin, Attempts to Redefine Conflicts of Interest, 25 Accountability in Research 67 (2018). 10 Id.

516   Richard S. Saver reflecting a physician’s previous intellectual position on use of a particular medical product. Meanwhile, not all financial interests are readily identified in the complex web of financial arrangements underlying the modern healthcare system. Often unclear is how payments from payers and manufacturers to institutions get blended with other revenue sources and trickle down to influence the economic gain of particular departments and specific staff physicians.11 Second, the discretionary/nondiscretionary distinction fuels the perception that nonfinancial interests, because nondiscretionary, are so pervasive that they remain too difficult to regulate. However, regulatory difficulty has also been experienced with financial ties. Nonfinancial interests may be worth the extra regulatory effort because of the serious risks of undue influence, and the point of much COI regulation is ultimately effective management, not necessarily elimination of the conflicts.

1.2  Risks and Prevalence Industry–medicine financial collaboration can yield many benefits, such as accelerating development of novel therapies and educating front-­line clinicians about complex medical products. However, the conflicting financial arrangements have been correlated with both systemic and patient-­focused risks, including harmful patient care, cost escalation, limits on patient choice, erosion of trust, publication bias, and threatened integrity to medical research.12 The evidence about harms posed by nonfinancial COIs remains far less developed. Nonetheless, studies suggest that certain nonfinancial interests, such as prior allegiance to a particular treatment method, raise comparable concerns to competing financial ties. In the context of research, both are associated with similar bias effects, such as failure to publish negative trial results.13 Moreover, the risk calculus is further complicated because “apparent conflicts can cause as much harm as real conflicts. . . . ”14 Because the public may lack expertise to evaluate the substance of many medical recommendations, neutral avoidance of competing interests becomes all the more important to maintaining trust. COIs are insidious because they may, unwittingly, bias decision-­making of well-­ intentioned providers who believe that they are rendering quality care. Physicians regularly face clinical situations involving substantial uncertainty, where considerable professional discretion exists as to what to recommend. Accordingly, secondary interests can hold considerable sway. Furthermore, some COIs create psychological

11 Richard S. Saver, Is It Really All About the Money? Reconsidering Non-­Financial Interests in Medical Research, 40 J. L. Med. & Ethics 467, 470 (2014). 12  IOM Conflict of Interest Report, supra note 4, at 23, 27. 13  See, e.g., M. Maj, Non-­Financial Conflicts of Interest in Psychiatric Research and Practice, 193 Brit. J. Psych. 91 (2008). 14  James L. Bernat, et al., Relationships Between Neurologists and Industry, 90 Neurology 1047 (2018).

Provider Conflicts of Interest in US Healthcare   517 pressures of indebtedness and reciprocity among physician recipients, which can be subtle but powerful motivators of human behavior.15 In addition to the risks posed, financial conflicts now pervade US healthcare. For example, manufacturers pay a staggering amount, about $7 billion each year, to physicians and teaching hospitals for consulting, speaking fees, educational grants, research support, and other services.16 Nearly half of all physicians (48%) received some form of payment from drug companies and device manufacturers in 2015.17 Patients have broad exposure to physicians under financial influence, with estimates suggesting that about two-­ thirds of patients have regular contact with physicians receiving industry payments.18 The evidence base is, unfortunately, far less developed concerning the prevalence of nonfinancial COIs. However, anecdotal reports suggest that not only are nonfinancial COIs important determiners of physician conduct, but that concerns also continue to recur about their influence.19 And commentators warn that nonfinancial interests, such as professional rivalry and institutional loyalties, are both widespread and may prove more potent than the prospect of economic gain.20

2  Obligations to Avoid or Manage COIs Under US Law Because of their prevalence and significant risks, direct legal oversight of provider COIs takes on increased importance. However, US law is a bit muddled. The legal obligation for healthcare providers to avoid or manage COIs may seem self-­evident. But such a rule is not firm and uniform under US law; it arises from a confusing patchwork of possible overlapping sources, including common law governing the provider–patient relationship and statutory/regulatory requirements. These rules are often limited to specific

15  See, e.g., J. Dana, A Social Science Perspective on Gifts to Physicians From Industry, 290 JAMA 252 (2003). 16  Ctrs. for Medicare & Medicaid Servs., The Facts About Open Payments Data: 2015 & 2016 Totals, https://openpaymentsdata.cms.gov/summary [https://perma.cc/W5TA-­VD6U] (detailing about $2.80 billion in general payments and $4.36 billion in research payments for 2016 and $2.68 billion in general payments and $4.45 billion in research payments for 2015). These estimates refer to payment transfers and do not include ownership and investment interests held by physicians and teaching hospitals. 17  Kathryn R. Tringale, et al., Types and Distributions of Payments from Industry to Physicians in 2015, 317 JAMA 1774, 1774 (2017). 18  Genevieve Pham-­Kanter, et al., Public Awareness of and Contact with Physicians Who Receive Industry Payments: A National Survey, J. Gen. Internal Med. (March 6, 2017), https://link.springer. com/article/10.1007/s11606-­017-­4012-­3. 19 Saver, supra note 11, at 469. 20  See, e.g., David Korn, Conflicts of Interest in Biomedical Research, 284 JAMA 2234 (2000).

518   Richard S. Saver contexts or impose difficult procedural and proof requirements, thus complicating efforts for comprehensive, robust oversight.

2.1  Fiduciary Duties The usual starting place in understanding healthcare providers’ obligations regarding COIs is fiduciary law. Numerous US courts recognize the doctor–patient relationship as fiduciary in nature because physicians, as trusted professionals, may abuse their superior position over dependent patients.21 Perhaps the most important duty in a fiduciary relationship, and most central to resolving questions about COIs, is the duty of loyalty. This duty obligates the fiduciary to focus on the beneficiary’s interests and avoid conflicts where self-­interest and the interest of other parties detract from pursuit of the beneficiary’s goals.22 However, the reach of fiduciary law to police COIs in the doctor–patient relationship is limited for several reasons. To start, there is considerable debate about how broadly fiduciary doctrine even extends to the doctor–patient relationship. Some leading ­treatises refer to it as a “confidential” relationship, rather than a fiduciary one.23 Furthermore, while numerous US court rulings affirm that the doctor–patient relationship has fiduciary characteristics, courts in a small number of states such as Alabama and Minnesota have expressly stated that the doctor–patient relationship is not a fiduciary one.24 Other caselaw recognizes fiduciary duties for physicians in only very limited circumstances, such as obligations to maintain confidentiality, but has stopped short of applying fiduciary principles to all aspects of the doctor–patient relationship including, importantly, financial conflicts of interest.25 Under some commentators’ views, physicians cannot be classic fiduciaries because this would unduly interfere with their competing responsibilities to society, including to control healthcare costs.26 Meanwhile, other scholars insist that physicians are traditional fiduciaries and criticize the nonfiduciary interpretation as based on a misunderstanding of precedent and

21  See, e.g., M.A. v. United States, 951 P.2d 851, 854 (Alaska 1998) (“We have recognized that the unique nature of the physician-­patient relationship confers upon physicians a fiduciary responsibility toward their patients.”); Witherell v. Weimer, 118 Ill. 2d 321, 331 (1987) (“A physician and his patient stand in a fiduciary relation . . .”). 22  See Restatement (Third) of Trusts § 2 cmt. b (“Despite the differences in the legal circumstances and responsibilities of various fiduciaries, one characteristic is common to all: a person in a fiduciary relationship to another is under a duty to act for the benefit of the other as to matters within the scope of the relationship”). 23  See id., § 2 cmt. (b)(1). 24  See Gunter v. Huddle, 724 So. 2d 544, 546 (Ala. Civ. App. 1998); Carlson v. SALA Architects, Inc., 732 N.W. 2d 324 331 (Minn. App. 2007). 25  See, e.g., Marc A. Rodwin, Money, Medicine, and Morals: Physicians’ Conflicts of Interest 210 (1993). 26  E. Haavi Morreim, Redefining Quality by Reassigning Responsibility, 20 Am. J. L. & Med. 79, 92 (1994).

Provider Conflicts of Interest in US Healthcare   519 confusion about what types of personal interests fiduciaries may still pursue.27 In any event, far more doubtful is that physicians wearing the hat of clinical investigator are fiduciaries to their research subjects. Research projects require strict fidelity to follow the study protocol to produce generalizable results, thus significantly limiting the ability of investigators to provide individually tailored care. Also, the principal aim of a study is to benefit future patients, not the subject, from the knowledge gained.28 The differing positions in commentary and uneven caselaw engender lingering confusion, undercutting the use of fiduciary duty rules to police provider COIs. For example, in classic fiduciary relationships, self-­dealing transactions by the fiduciary are usually automatically voidable by the principal, but if the doctor–patient relationship is merely a confidential relationship, an aggrieved patient seeking to undo an action must first prove that she placed confidence in the physician and that this confidence was abused.29 Furthermore, despite the strong rhetoric surrounding the fiduciary’s duty of loyalty, absolute fidelity to the beneficiary is not always required as courts occasionally modify the trust-­based rules when the maximalist position is at odds with policy and societal concerns.30 Fiduciary law in the healthcare context has, likewise, been applied flexibly to allow for financial incentives directed at physicians that otherwise might present duty of loyalty concerns because the incentives help advance other societal imperatives.31 For example, claims by patients that their physicians breached fiduciary duties by receiving financial incentives to reduce care utilization have been rejected by US courts on the grounds that the capitation payments were elsewhere endorsed by applicable law as a method of ­controlling healthcare costs.32 Similarly, the US Supreme Court ruled in Pegram v. Herdrich33 that physicians receiving shared savings incentives under a prepaid clinic model did not violate their statutory fiduciary obligations under the Employee Retirement Income Security Act (ERISA) to act for the benefit of plan beneficiaries. The Court noted that the financial incentives complained about were endorsed by Congress in legislation regulating health maintenance organizations.34 Finally, even legal acknowledgment that the physician owes a fiduciary duty of loyalty has not always resulted in pragmatic success for patients. US courts have been reluctant to recognize broad causes of action for breach of fiduciary claims. In one leading case, Neade v. Portes,35 the Illinois Supreme Court considered a breach of fiduciary duty claim alleging that the physician had financial incentives from the patient’s managed care plan 27  See Maxwell J. Mehlman, Why Physicians Are Fiduciaries For Their Patients, 12 Ind. Health L. Rev. 1, 17–19 (2015). 28  See Suthers v. Amgen, Inc., 372 F. Supp. 2d 416, 427 n.9 (S.D.N.Y. 2005). 29  See Mehlman, supra note 27, at 17–18. 30  D. Gordon Smith & Jordan C. Lee, Fiduciary Discretion, 75 Ohio St. L. J. 609, 621 (2014). 31  See, e.g., Robert Gatter, Communicating Loyalty: Advocacy and Disclosure of Conflicts in Treatment and Research Relationships in Oxford Handbook of US Health Law 242–247 (2017). 32  Id., at 248. 33  530 US 211 (2000). 34  Id., at 233–234. 35  193 Ill. 2d 433 (2000).

520   Richard S. Saver to deny specialist referrals and tests outside the physician’s office. The court viewed the breach of fiduciary duty claim as duplicative of a medical malpractice negligence claim and limited the plaintiff to arguing the malpractice cause of action.36 Likewise, in D.A.B.  v. Brown,37 a class of patients challenged an alleged kickback scheme involving payments to induce their physician to prescribe a synthetic growth hormone. The Brown court found that creation of a new cause of action for breach of fiduciary duty was unnecessary, reasoning that “[a]ny breach of fiduciary duty that may have occurred . . . arose while the doctor was examining, diagnosing, treating, or caring for his patients” and thus constituted a claim for medical negligence.38 Part of the concern with duplicative causes of action is that litigating fiduciary duty breaches enables medical malpractice patients to bypass proving that physical injury occurred, a necessary element of a malpractice claim, or to avoid important statutes of limitations applicable to malpractice actions.39

2.2  Informed Consent A physician’s informed consent duty under US law generally obligates disclosure of the proposed treatment, its likely benefits and material risks, and the medically viable or medically reasonable alternatives to the physician’s recommendation.40 To succeed in a negligent informed consent claim, the plaintiff generally must show that the physician failed to disclose the required information and that the reasonable patient would have decided differently if the omitted information had been communicated. The patient must also demonstrate that the disclosure failure caused actual harm, meaning the patient would have likely ended therapeutically better off if the different treatment choice had been made.41 US courts struggle with how to fit COIs within classic informed consent doctrine. As noted, informed consent claims typically involve nondisclosure of physical risks associated with a proposed treatment. The fact that a physician proposing treatment has a potential COI is a more attenuated risk, not tied to physical injuries emanating from the proposed procedure itself. Informed consent doctrine generally does not require open-­ ended disclosure of all possible risks because this could be unduly costly, overwhelm patients with vast risk information, and end up being counterproductive. 36  Id., at 440 (“Illinois courts have recognized a fiduciary relationship between a physician and his patient, . . . but Illinois courts have never recognized a cause of action for breach of fiduciary duty against a physician.”). But see Moore v. Regents of the Univ. of Cal. 51 Cal. 3d 120, 129 (Cal. 1990) (“[A] physician’s failure to disclose such interests may give rise to a cause of action for performing medical procedures without informed consent or breach of fiduciary duty” [emphasis added]). 37  570 N.W.2d 168 (Minn. Ct. App. 1997). 38  Id., at 172. 39  See id., at 171–172. 40  See, e.g., Matthies v. Mastromonaco, 160 N.J. 26, 36–37 (N.J. 1999). 41  See Joan H. Krause, Reconceptualizing Informed Consent in an Era of Health Care Cost Containment, 85 Iowa L. Rev. 261, 317–322 (1999).

Provider Conflicts of Interest in US Healthcare   521 Some US courts have held that financial incentives may rise to the level of “material” risks requiring disclosure. Presumably, this is because the COI can influence the provider’s medical judgment in a way that may physically harm patients. The leading case supporting disclosure of COIs is Moore v. Regents of the University of California.42 In Moore, the plaintiff alleged that the physician treating him for hairy cell leukemia withdrew his spleen, blood, bone marrow aspirate, and other bodily materials without disclosing plans to use the plaintiff ’s cells in potentially lucrative clinical research.43 The California Supreme Court held that the patient could state a possible breach of fiduciary duty claim and also an informed consent claim. As for informed consent, the court reasoned that a physician must “disclose personal interests unrelated to the patient’s health, whether research or economic, that may affect his medical judgment.”44 In deciding on a course of treatment, the court explained, a reasonable patient would want to know whether “an interest extraneous to the patient’s health has affected the physician’s judgment.”45 Moore seemingly opened the door to all sorts of “extraneous” interest information requiring disclosure as part of the material risks under informed consent. Some courts have followed the Moore approach to require disclosure of potentially conflicting financial incentives that might affect the physician’s judgment, such as payments from a drug company for prescribing a particular medication.46 However, in cases involving more typical physician–industry financial relationships, such as consulting arrangements, other courts frequently distinguish Moore on the grounds that it involved the research setting.47 COIs can be viewed as part of the larger category of physician-­specific information that meshes awkwardly, if at all, with informed consent doctrine. For example, in Duttry v. Patterson,48 the Supreme Court of Pennsylvania declined to recognize an informed consent claim by a patient alleging her physician misrepresented his experience in performing a particular surgical procedure. The court reasoned that such information was “irrelevant” for informed consent purposes because it was beyond information relative to the risks of the surgery itself.49 Consistent with this approach, the court in Corrigan v. Methodist Hospital50 ruled that a patient did not need to be informed of a financial relationship between her surgeon and the manufacturer of the pedicle screws used during spinal surgery because the financial tie was not a “surgical risk.”51 What seems clear is that to even plausibly succeed in challenging COIs under an informed consent claim, patients must present good evidence that the extraneous 42  51 Cal.3d 120 (Cal. 1990). 43  Id. at 125. 44  Id. at 131–132. 45  Id. at 130. 46  See, e.g., Darke v. Estate of Isner, No. 02-­2194, 2004 WL 1325635, at *2 (Mass. Super. 2004). 47  See, e.g., Dimmick v. United States, No. C 05-­0971 PJH, 2006 WL 3741911, at *19 (N.D. Cal. 2006); Hecht v. Kaplan, 645 N.Y.S.2d 51, 53 (Supr. Crt. App. Div. 1996). 48  565 Pa. 130 (Pa. 2001). 49  Id., at 137. But see Howard v. University of Medicine & Dentistry of New Jersey, 172 N.J. 537 (N.J. 2002) (significant misrepresentation of provider’s experience and credentials might sustain an informed consent claim as it could be material to determining actual medical risks of the procedure to which the patient consented). 50  874 F. Supp. 657 (E.D. Pa. 1995). 51  Id., at 659.

522   Richard S. Saver interest affected clinical judgment in a way that increased the medical risks of the procedure recommended, rather than more generalized assertions that the conflict was wrong or eroded trust and confidence in the physician. However, hard causal evidence linking COIs to actual risk of harm is debatable52 and may not be sufficient in many cases to maintain viable informed consent actions. Even in states that allow informed consent claims based on failure to disclose financial conflicts, these claims are rarely successful due to the short statute of limitations53 and the difficulties of establishing causation and damages.54 In his dissent to Moore, Justice Mosk of the Supreme Court of California articulated this dilemma for plaintiffs, noting “here we deal not with a risk of physical injuries such as a stroke, but with the possibility that the doctor might later use some of the patient’s cast-­off tissue for scientific research or the development of commercial products. Few if any judges or juries are likely to believe that disclosure of such a possibility of research or development would dissuade a reasonably prudent person from consenting to the treatment.”55 Indeed, the empirical evidence suggests that many patients, while desiring information about financial conflicts, also indicate that this information would not sway their decision to consent to treatment or research.56 This makes it all the harder, therefore, for patients to satisfy the required causation element in financial COI cases alleging informed consent violations.

2.3  Ordinary Medical Malpractice COIs can also come into play in malpractice cases alleging negligent delivery of care, as opposed to informed consent failures, although this occurs far less often in US litigation. In an ordinary malpractice case, the patient must show that a therapeutic relationship existed and that the physician breached the duty of care by deviating from applicable care standards. In theory, COI information should be regarded as irrelevant to the typical medical malpractice claim. To prevail, the patient need only prove that the physician’s course of treatment fell below the standard of care and that it caused the 52  See Section 4 infra. 53  See, e.g., D.A.B. v. Brown, 570 N.W.2d 168, 171–172 (Minn. Ct. App. 1997). 54  See, e.g., Dimmick v. United States, No. C 05-­0971 PJH, 2006 WL 3741911, at *20 (N.D. Cal.2006) (“[E]ven if the court were to find that Dr. Lampiris had a duty to disclose his role as an Abbott consultant, Dimmick’s claim fails for a lack of causation”). 55  51 Cal. 3d. 120, 180 (Cal. 1990) (Mosk, J. dissenting); Id.; See also Darke v. Estate of Isner, No. 02-­2194, 2004 WL 1325635 (Mass. Super. 2004). In Darke, a Massachusetts Superior Court held that the plaintiff had a valid informed consent claim against a physician based on failure to disclose that the physician had a financial interest in the gene therapy treatment that he recommended. Id. The jury, however, rejected plaintiff ’s claims on the merits, failing to find the nondisclosure material to the plaintiff ’s decision to undergo gene therapy. Jury Verdict Form, Darke, v. Isner, 2007 WL 2219825 (Mass. Super. 2007). 56  See, e.g., Adam Licurse, et al., The Impact of Disclosing Financial Ties in Research and Clinical Care: A Systematic Review, 170 Archives Intern. Med. 675 (2010); Mark W. Camp, et al., Patients’ Views on Surgeons’ Financial Conflicts of Interest, 95 J. Bone Joint Surg. e9(1), e9(6) (2013).

Provider Conflicts of Interest in US Healthcare   523 patient’s injury. Because the focus in negligence is on conduct, not intent, the reasons why the physician’s actions deviated from the standard of care, whether due to COIs or otherwise, are not a necessary element. However, in some US medical malpractice cases, evidence of competing financial incentives has figured in resolving the ultimate question whether a breach of the stand­ ard of care occurred. For example, in Martinez v. Elias,57 the court allowed evidence of an orthopedic surgeon’s financial motives in performing and interpreting discograms as relevant to the surgeon’s compliance with the standard of care. The court reasoned that “[t]he evidence of financial incentive supported the plaintiff ’s claim that the surgical procedures were unnecessary.”58 In other ordinary medical malpractice cases in the United States, COI information has been useful to plaintiffs for more strategic reasons, such as impeaching the physician’s testimony.59 However, because of its possible prejudicial influence on the jury, courts have shown reluctance in allowing COI information to be used broadly. Among other reasons, there is concern that plaintiffs not be allowed to rely on COI information to bootstrap claims and avoid having to produce evidence through expert testimony and other traditional means showing that the treatment rendered actually did deviate from the standard of care.

2.4  Statutory and Regulatory Rules in Specific Settings Provider COIs may also trigger US statutory and regulatory rules for managing conflicts in particular settings, such as occurs within medical research and health plan operations. These obligations typically fall not on the provider but on the entities making the payments or the institutions where the physicians practice.

2.4.1  Research: Public Health Service and Food and Drug Administration Rules National Institutes of Health (NIH) regulations require that physician-­investigators receiving NIH grants disclose to their institutions “significant financial interests”60 (generally $5,000 or more), such as consulting payments or equity interests, that relate to their research-­related activities.61 The institution must then evaluate the disclosed financial relationship and report to the US Public Health Service (PHS) any interests identified as conflicting and the institution’s management plan for dealing with the conflict.62 Internal conflict of interest committees (COICs) usually perform this important required review at the institutional level.

57  397 Ill. App. 3d 460 (Ill. App. Ct. 2009). 58  Id., at 469. 59  See, e.g., Neade v. Portes, 303 Ill. App. 3d 799, 816–817 (1999), rev’d on other grounds, 193 Ill. 2d 433 (2000). 60  42 C.F.R. § 50.604(e). 61  42 C.F.R. § 50.603. 62  42 C.F.R. §§ 50.604(g), 50.605.

524   Richard S. Saver Having researchers disclose COIs to their institutions for further response and management has been questioned as an ill-­conceived oversight mechanism because frequently the research institutions have their own financial conflicts as well.63 In any event, in coming up with management plans, the institutional COICs have tremendous discretion. For example, COICs can choose to require the investigator to disclose the conflict in all published studies resulting from the research, move the research to another site, or ban the conflicted investigator from directly enrolling subjects for the research.64 Unfortunately, minimal to no regulatory guidance exists on basic matters such as how COICs are to be constituted, the best practices they should follow, and the substantive criteria that should guide their reviews. This can leave COICs woefully unprepared for simply assessing the financial relationships, let alone determining appropriate oversight steps. Likewise, the US Food and Drug Administration (FDA) rules require a company seeking FDA approval of its product to report to the agency certain financial relationships (payments greater than $25,000 and equity interests greater than $50,000) with clinical investigators testing the product in FDA-­regulated clinical trials.65 The rules do not prescribe a particular course of action for the FDA upon learning of such financial relationships. As with the NIH rules, tremendous discretion is left to the ultimate reviewing body for how to address the conflict, leading to concerns of inconsistent and ad hoc management of COIs under the research regulations.

2.4.2  Accountable Care Organizations and Health Plans In certain instances, provider reimbursement implicates statutory and regulatory rules under US law regulating the use of financial incentives. However, “[b]ecause law has endorsed the use of financial incentives in these arenas as a way of promoting economic efficiency, mandatory disclosure rather than prohibition is the primary mechanism to account for the risks associated with potential COIs.”66 Moreover, these disclosure regimes tend to be light disclosure in the form of sharing information with regulators or on the entities’ own websites, rather than centralized reporting and detailed information disseminated to each patient. For example, accountable care organizations (ACOs), the large integrated provider networks encouraged by the Affordable Care Act, contract with payers to provide a broad spectrum of care to a group of patients. When contracting with the US Medicare program, ACOs are eligible to share in savings they generate for the governmental program by controlling costs. Federal regulations require that each Medicare ACO publicly report on its website how any shared savings are distributed to its network providers,67 63 Gatter, supra note 31, at 250. 64  42 C.F.R § 50.605(a)(1). 65  21 C.F.R. pt. 54. 66 Gatter, supra note 31, at 248. 67  42 C.F.R. § 425.308(b); Centers for Medicare and Medicaid Services, Medicare Shared Savings Program: Public Reporting Guidance 11 (updated 2017), https://www.cms.gov/Medicare/ Medicare-­Fee-­for-­Service-­Payment/sharedsavingsprogram/Downloads/ACO-­Public-­Reporting-­ Guidance.pdf.

Provider Conflicts of Interest in US Healthcare   525 post signs in its facilities notifying Medicare patients that the ACO participates in a shared savings program, and make standardized written notices available upon request.68 Likewise, some state laws require health insurance plans to disclose to state regulators particular financial arrangements they have with network providers. There is no guarantee that these various forms of public disclosure actually reach individual patients, however. Under most applicable laws, ACOs and health plans are not required to disclose their financial incentives to individual patients unless specifically asked for such information.69

2.5  Physician Payments Sunshine Act The most significant recent legislative development in the United States for regulation of COIs is the Physician Payments Sunshine Act (Sunshine Act), which mandates public reporting of payments between drug companies, device manufacturers, and medicine.70 The mixed record of the Sunshine Act and the limits of its disclosure approach are discussed in more detail in Section 3.

2.6  Self-­Regulation: Institutional and Professional Society Policies Finally, self-­regulation plays an important role in oversight of provider COIs in the United States. Policies and ethics guidance of professional provider and industry groups offer additional instruction and professional expectations for dealing with COIs.71 Consistent with Association of American Medical Colleges (AAMC) recommendations72 and the NIH research rules, many US medical centers now have institutional COI policies that staff physicians must adhere to as a condition of maintaining their clinical privileges. For example, the COI policies of Partners Healthcare System, which includes Harvard-­affiliated Brigham and Women’s Hospital and Massachusetts General Hospital, make it generally impermissible for staff physicians engaged in research to have equity interests of greater than $50,000 in a company or to receive income of greater than $25,000 from a company when engaged in research on that particular firm’s

68  42 C.F.R. § 425.312(a). 69 Gatter, supra note 31, at 252. 70  42 USC. § 1320a-­7h (2012). 71  See, e.g., AMA Code of Medical Ethics § 7.1.4 (Conflicts of Interest in Research); AMA Code of Medical Ethics § 9.6.2 (Gifts to Physicians from Industry); PhRMA Code on Interaction with Health Care Professionals § 6 (Consultants). 72  Assoc. of Amer. Medical Colleges, In the Interest of Patients: Recommendations for Physician Financial Relationships and Clinical Decision Making 23 (2010), https://www.aamc. org/download/482222/data/intheinterestofpatients.pdf.

526   Richard S. Saver technology.73 Partner physicians are also supposed to disclose their financial interests with industry to an institutional oversight committee that has wide discretion to implement management plans including required disclosure to patients, independent corroboration of the physician’s prescriptions, review of the physician’s practice patterns, or transfer of care to another physician.74 While self-­regulation remains an important part of the regulatory landscape, concerns persist about its ultimate effectiveness. As previously noted, there are questions about how independent and forceful COIC reviews can be when the institutions themselves may also have conflicting financial interests with industry. Also, much self-­ regulation depends on, in the first instance, physicians voluntarily reporting their activities to their institutions, and there is reason to question whether physicians faithfully report conflicting financial ties.75 Furthermore, beyond required disclosure to patients, it is not clear that COICs are taking more forceful and creative management actions, such as independently reviewing prescribing decisions of providers with conflicts. In any event, the lack of firm guidance for COICs and the wide discretion afforded to them under various policies means that self-­regulation remains largely ad hoc and varied between institutions, leaving significant gaps and inconsistencies in oversight.

3  The Promise and Pitfalls of Disclosure As noted, many of the applicable legal rules and institutional policies in the United States for addressing COIs rely primarily on required disclosure. Enactment of the Sunshine Act in 2010 significantly amplified the trend toward increased transparency. Regulating COIs through disclosure offers many theoretical benefits for the healthcare system, including facilitating efficient market operations, strengthening fiduciary relationships, and enhancing evaluation of financing and delivery.76 However, the Sunshine Act’s mixed track record illustrates more generally why transparency raises considerable implementation and design challenges and brings its own potential downsides and unintended effects.

73  Partners Healthcare, Partners Policy for Interactions with Industry and Other Outside Entities §1.2.3 (Approved Nov. 16, 2016), https://www.partners.org/Assets/Documents/About-­Us/OII/OII_ Policy.pdf. 74  Id. §2.3.2. 75  See, e.g., IOM Conflict of Interest Report, supra note 4, at 63. 76 William M. Sage, Regulating Through Information: Disclosure Laws and American Health Care, 99 Colum. L. Rev. 1704–1707, 1710–1711 (1999); Kristin Madison, Regulating Health Care Quality in an Information Age, 40 U.C. Davis L. Rev. 1577, 1589–1592 (2007).

Provider Conflicts of Interest in US Healthcare   527

3.1  Sunshine Act Basics and Implementation Challenges The Sunshine Act is the first comprehensive federal legislation mandating public reporting of payments between drug companies, device manufacturers, and medicine. It requires that drug and device manufacturers of products covered by Medicare, Medicaid, or the Children’s Health Insurance Program report annually to the Centers for Medicare and Medicaid Services (CMS) information about their physician investors and their payments to physicians and teaching hospitals.77 CMS releases most of the data on a publicly available, searchable website known as the Open Payments Database.78 Any transfers of value to physicians and teaching hospitals of greater than $10 per instance or $100 per year must be disclosed.79 Manufacturer reports must characterize each transfer as a research payment or general payment, and place each general payment in a limited number of categories, such as consulting fees, travel and lodging, royalty and license fees, and education.80 The Sunshine Act thus goes far beyond many previous disclosure programs because it is a uniform federal law, has the very low dollar threshold for triggering reporting, and applies to an extensive range of financial relationships. However, inherent challenges arise in data selection and communication. The Sunshine Act initially did not account for all important recipients of industry payments. It excluded reporting about physician assistants and nurse practitioners, although Congress has tried to fix this gap moving forward.81 The legislation is also under-­ inclusive about the other side of the transaction: the entities making payments. It imposes payment reporting obligations only on manufacturers, not on independent wholesalers and distributors,82 even though these entities also have a direct economic interest in encouraging greater provider utilization of the product. Even more troubling is the lack of needed context and comparative information. Studies of physician financial incentives indicate that a financial incentive’s actual risk of creating bias and undue influence depends on a number of important variables including the amount of money at stake, how this compares relative to the physician’s other sources of income, the nature of the physician’s services involved in the transaction, how long the financial relationship will continue, the length of any measurement periods to

77  42 USC. § 1320a-­7h(a). 78  See Open Payments Database, Ctrs. for Medicare & Medicaid Servs., https://openpaymentsdata. cms.gov. 79  42 USC. § 1320a-­7h(e)(10)(B). 80  42 USC. § 1320a-­7h(a)(1)(A)(v), (vi); 42 C.F.R. § 403.904(e)(2). 81  The SUPPORT for Patients and Communities Act, enacted in 2018, expanded the Sunshine Act reporting requirements to include these ancillary providers for data submitted starting in 2022. Pub. Law. No. 115–271 (Oct. 24, 2018). 82  See Medicare, Medicaid, Children’s Health Insurance Programs; Transparency Reports and Reporting of Physician Ownership or Investment interests, 78 Fed. Reg. 9458, 9461–9462 (Feb. 8, 2013).

528   Richard S. Saver earn compensation, whether explicit quality of care measurements also impact the physician’s receipt of compensation, and the number of patients affected.83 The architects of the Open Payments Database opted to omit much of this important contextual information, keeping the disclosures deceptively simple. Not only does this exclude needed context as well as important comparative information about how a physician’s payments compare to regional peers, but the remaining data reported also may skew patient responses. Matters such as payment amount may mislead audiences into thinking that the greater the payment, the greater the reason for concern. But the scholarly literature on financial conflicts urges increased attention to restricting even small gifts and de minimis payments. Even low-­dollar financial relationships can create indebtedness, unconscious gratitude, and reciprocity obligations for the physician recipients, pressures which exert potential bias in subtle but powerful ways.84

3.2  Disclosure’s Downsides The Sunshine Act reflects several of the more general downsides associated with required disclosure. Many transparency programs generate information too complex for end users to process successfully.85 Indeed, the Open Payments Database may mislead users with informational noise, skew patient responses, and obscure instances of actual undue influence and bias. Second, disclosure regulation may be, ironically, too attractive because of its apparent low cost and political acceptability.86 In this respect, the Sunshine Act may be crowding out consideration of additional financial conflicts regulation in the United States. Third, transparency’s apparent low cost may be misleading. CMS estimates a not insignificant $180 million in annual compliance costs arising from the Sunshine Act.87 The overall cost effects of transparency are uncertain. Deterring problematic industry– medicine financial ties through public disclosure may help lower certain healthcare expenditures. On the other hand, other costs may escalate because certain industry– medicine financial relationships will be continued, with necessarily higher payments offered to the providers to offset the reputational harms incurred for transacting with industry.88 Fourth, transparency runs the risk of, counterintuitively, magnifying overly 83  See, e.g., Richard S. Saver, Squandering the Gain: Gainsharing and the Continuing Dilemma of Physician Financial Incentives, 98 Nw. U. L. Rev. 145, 207–210 (2003); Brian Armour, et al., The Effect of Explicit Financial Incentives on Physician Behavior, 161 Arch. Intern. Med. 1261, 1265–1266 (2001). 84  See generally Jason Dana, How Psychological Research Can Inform Policies for Dealing with Conflicts of Interest in Medicine, in IOM Conflict of Interest Report, supra note 3, app. D at 358–374. 85  Omri Ben-­Shahar & Carl E. Schneider, More Than You Wanted to Know: The Failure of Mandated Disclosure 3 (2014). 86 Id. 87  See Medicare, Medicaid, Children’s Health Insurance Programs; Transparency Reports and Reporting of Physician Ownership or Investment interests, 78 Fed. Reg. 9458, 9466–9467 (Feb. 8, 2013). 88  Daniel L. Chen, at al., Mandatory Disclosure: Theory and Evidence from Industry-­Physician Relationships (2017), http://users.nber.org/~dlchen/papers/Mandatory_Disclosure.pdf.

Provider Conflicts of Interest in US Healthcare   529 negative public views about industry–medicine collaboration because of the public’s inclination to assume that transactions needing disclosure are worrisome.

3.3  Disclosure’s Uncertain Audience Effective transparency accounts for end users expressly in questions of format, translation, framing, and operability to ensure useful engagement with the information.89 The Sunshine Act, like many COI disclosure programs, lacks this fine-­tuned design. If, as may be self-­evident, patients are part of the Sunshine Act’s primary audience, one must question whether they have the inclination, resources, and capabilities to use the disclosed information productively. Indeed, the early Sunshine Act data indicate that patients are engaging with the information in only modest numbers.90 Moreover, research suggests that financial conflicts disclosure can have unintended consequences with patients. Behavioral theorists posit that learning of a financial conflict may create new pressures, or “insinuation anxiety,” for a patient to follow the physician’s recommendation because to not follow would send an implied message that the patient believes the physician is dishonest, something the patient prefers to avoid.91 Also, rather than evaluating the physician’s recommendations more critically, the patient may become more trustworthy of the physician simply because of the act of disclosure. Or, if inclined to discount the physician’s recommendation, the patient may not know how to do so and to what degree.92 In any event, a growing body of research suggests that financial ties disclosures have limited sway over patient and research subject decision-­making. While a majority of patients and subjects say they wish to know about financial ties, only smaller numbers indicate that this information would affect their decision-­making about type of clinical care or whether to participate in research.93 Physicians are another problematic audience for COI disclosure. Faced with transparency, they may, in theory, decline certain transactions and reevaluate their own prescribing patterns. But it remains to be seen whether increased transparency really impacts physician behavior in this manner. A 2017 focus group study found that many of the participating physicians had only limited interactions with the Open Payments Database, reflecting in part their distrust and skepticism about the utility of Sunshine

89  See generally Nathan Cortez, Regulation by Database, 89 U. Colo. L. Rev. 1 (2018). 90  See Ctrs. for Medicare & Medicaid Servs., Annual Report to Congress on the Open Payments Program 20 (2016). 91  George Loewenstein, et al., The Unintended Consequences of Conflict of Interest Disclosure, 307 JAMA 669, 670 (2012). 92  George Loewenstein, et al., The Limits of Transparency: Pitfalls and Potential of Disclosing Conflicts of Interest, 101 Am. Econ. Rev. 423, 424 (2011). 93  See, e.g., Licurse supra note 56.

530   Richard S. Saver Act disclosures.94 Also, the evidence is mixed as to whether financial conflicts disclosure programs significantly influence physician prescribing.95 Additionally, behavioral research suggests that, similar to patients, disclosure can have unintended, counterproductive effects on physicians. Psychological studies of advisors and their advisees, analogous to the doctor–patient relationship, indicate that when advisors have to reveal their financial conflicts they tend to offer more biased guidance than if the financial ties remain undisclosed. They may offer more extreme advice to offset the anticipated discounting of their recommendations, as well as feel more emboldened to give extreme advice because their advisees have been warned about the underlying financial conflict.96

3.4  Disclosure’s Value As with much disclosure regulation, the Sunshine Act’s real value rests on what it may accomplish in the longer term as the information disseminates to other audiences. There are signs that such promising applications are already occurring. One noteworthy investigation using Sunshine Act data found a strong correlation between a physician receiving a single meal from industry and higher rates of prescribing costlier brand-­name medications.97 Studies such as this better inform evidence-­based regulation of industry’s promotional activities. More immediately, Sunshine Act information helps regulators enforce the existing healthcare fraud and abuse laws by revealing unknown financial ties or outliers that warrant further scrutiny. A dramatic example is the ongoing saga of Insys Therapeutics, an Arizona pharmaceutical company. Various regulatory investigations into indiscriminate prescribing of Subsys, an opioid manufactured by Insys, gathered steam and widened considerably after release of the first batch of Sunshine Act data.98 The Open Payments Database revealed that Insys had been paying speaker fees, travel, and meals, and 94  Susan Chimonas, et al., Bringing Transparency to Medicine: Exploring Physicians’ Views and Experiences of the Sunshine Act, 17 Am. J. Bioethics 4, 6–7 (2017). 95  Compare Genevieve Pham-­Kanter, Effect of Physician Payment Disclosure Laws on Prescribing, 172 arch. intern. med. 819 (2012) with Tong Guo, et al., Let the Sun Shine in: The Impact of Industry Payment Disclosure on Physician Prescription Behavior (Apr. 18, 2017), https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2953399&download=yes. 96  Sunita Sah, Conflicts of Interest and Your Physician: Psychological Processes That Cause Unexpected Changes in Behavior, 40 J. L. Med. & Ethics 482, 485 (2012). 97  Colette Dejong, et al., Pharmaceutical Industry-­Sponsored Meals and Physician Prescribing Patterns for Medicare Beneficiaries, 176 JAMA Intern. Med. 1114 (2016). 98  Press Release, Or. Dep’t of Just., AG Rosenblum Settles with Pharmaceutical Company Insys over Unlawful Promotion of the Powerful Opioid Subsys (Aug. 5, 2015), http://www.doj.state.or.us/releases/ Pages/2015/rel080515.aspx; Katie Thomas, Former Insys Officials Charged in Scheme to Push its Painkiller, N.Y. Times, Dec. 8, 2016; Dep’t of Justice, United States Intervenes in False Claims Act Lawsuits Accusing Insys Therapeutics of Paying Kickbacks and Engaging in Other Unlawful Practices to Promote Subsys, a Powerful Opioid Painkiller, May 15, 2018, https://www.justice.gov/opa/pr/ united-­states-­intervenes-­false-­claims-­act-­lawsuits-­accusing-­insys-­therapeutics-­paying.

Provider Conflicts of Interest in US Healthcare   531 ­ aking other payments to the nation’s top prescribers of Subsys, including several m physicians disciplined by state medical boards for problematic conduct.99 Thus, COI disclosure programs retain important value despite their many downsides and unintended effects. The greatest benefit arises from use of the information not by patients but secondary intermediaries, analogous to what has been seen with provider quality reporting laws.100 These secondary intermediaries can process the information, translate it into salient components responsive to patient needs, and present it in a format more easily understood by patients.

4  Why Optimal Regulation of COIs Remains So Challenging If disclosure is a necessary, but by itself insufficient, regulatory response for managing many COIs, the question remains: What else can be done? Unfortunately, the path to optimal regulation is fraught with many difficulties. Provider COIs remain a particularly thorny, unwieldy subject matter for regulation for several reasons.

4.1  Contested Causal Impact of Financial Conflicts A threshold problem is the limited evidence base. Despite the long-­standing concerns about COIs, especially financial conflicts, scant data exist concerning their effects on the US healthcare system, and the data remain subject to conflicting interpretation.101 Opponents of heightened regulation charge that proof is still lacking that financial ties sway physicians’ prescribing and referral decisions in ways that actually harm patients therapeutically.102 For example, while financial ties have been correlated with more expensive drug prescribing,103 this does not mean that patients are receiving clinically 99  Katie Thomas, Using Doctors with a Troubled Past to Market a Painkiller, N.Y. Times, Nov. 27, 2014; Roddy Boyd, The Black World of Insys Therapeutics, The Foundation for Financial Journalism (July 14, 2015), http://sirf-­online.org/2015/07/14/the-­darkening-­world-­of-­insys-­therapeutics/. 100  See, e.g., Kristin Madison, Health Care Quality Reporting: A Failed Form of Mandated Disclosure?, 13 Ind. Health L. Rev. 310, 330–333 (2016). 101  Compare Christopher Robertson, et al., Effects of Financial Relationships on the Behaviors of Health Care Professionals: A Review of the Evidence, 40 J. L. Med. & Ethics 452 (2012), with Teddy D. Warner & John P. Gluck, What Do We Really Know About Conflicts of Interest in Biomedical Research?, 171 Psychopharmacology 36 (2003). See also Ronald Bailey, Scrutinizing Industry Funded Science: The Crusade Against Conflicts of Interest (2008). 102  See, e.g., D. Barton, et al., After 20 Years, Industry Critics Bury Skeptics, Despite Empirical Vacuum, 68 Int’l J. Clinical Prac. 666, 666 (2014). 103  See, e.g., Roy H. Perlis, et al., Physician Payments from Industry Are Associated with Greater Medicare Part D Prescribing Costs, Plos One, May 16, 2016.

532   Richard S. Saver inappropriate treatments. Standards of care are rarely uniform; wide variations exist in the way clinicians practice medicine.104 Debate over the true impact of financial conflicts boils down to correlation versus causation. Many studies are observational and typically report an association between financial ties and problematic conduct, as opposed to causal evidence.105 Risks such as harming patients and increasing costs have generally not been linked to underlying financial conflicts through comprehensive studies employing gold standard proof, such as randomization and use of controls. Any number of confounding variables, including small sample sizes and the contaminating effect of peer influence, may be responsible for associations between financial ties and certain problematic conduct.106 Furthermore, some correlations may reflect bidirectional causation as physicians who prescribe in larger volume or more expensively may in turn willingly expose themselves to more promotional activities and industry influence.107 Evidence of correlation, even if not causation, may still be enough to warrant stronger regulatory responses. For one, financial relationships may reinforce problematic prescribing patterns through association, even if not initially causing those prescribing choices.108 Moreover, the risk of harm from certain financial relationships may be of such concern that, because of risk mitigation and the “precautionary principle,”109 it may be better to regulate prophylactically. Nonetheless, the lack of firm causation evidence remains a serious challenge. It calls into question the fairness and ultimate justification for much financial conflicts regulation. This makes it all the harder to obtain consensus, including all-­important buy-­in from more entrepreneurial physicians and other providers. Perceptions that conflicts regulation inequitably burdens physicians who collaborate with industry can encourage cynicism, disregard, and noncompliance.

4.2  Underdeveloped Regulatory Alternatives Related to the limited evidence base about COIs’ causal impact is the limited data assessing effectiveness of regulatory responses. In particular, alternatives to disclosure have not been well evaluated. These approaches include supplying alternative sources of information to guide healthcare decision-­making, such as mandatory second opinions from independent physicians about recommended treatment and research. Similarly, 104  See, e.g., John E. Wennberg, Variation in Use of Medicare Services Among Regions and Selected Academic Medical Centers: Is More Better?, Commonwealth Fund Pub. No. 874 (2005). 105  See, e.g., Stanford C. Taylor et al., Physician-­Industry Interactions and Anti-­Vascular Growth Factor Use Among US Ophthalmologists, 134 JAMA Ophthalmology 897, 897 (2016). 106  See, e.g., Barton supra note 102, at 671. 107 Id. 108  See, e.g., Bernard Lo, et al., Payments to Physicians: Does the Amount of Money Make a Difference? 317 JAMA 1719, 1720 (2017). 109  Daniel Goldberg, On Physician-­Industry Relationships and Unreasonable Standards of Proof for Harm: A Population-­Level Bioethics Approach, 26 Kennedy Inst. Ethics J. 173, 175 (2016).

Provider Conflicts of Interest in US Healthcare   533 academic detailing programs can provide neutral expert physician advice to community physicians about particular treatment options.110 Regulation can also impose outright prohibitions, such as transactional bans on financial ties of certain dollar amounts or within certain payment categories (e.g., banning speaker fees). Unfortunately, none of these approaches has been implemented in a way to accrue significant data because, in the United States, much conflict of interest regulation, whether through law or institutional policies, has developed ad hoc in response to particular crises.111 Accordingly, questions arise whether particular regulatory options work well in the broad range of situations contemplated. Furthermore, like mandated disclosure, it is likely that some of these alternatives have unintended effects and introduce new problems. For example, requiring patients to seek a second medical opinion from a physician who has no COI seems, at first impression, a promising way to counter potential bias. But behavioral studies suggest that professional advisors with a conflict, aware that their clients will have access to a second opinion, are more likely to give more biased recommendations.112 Therefore, a required second opinion policy could backfire by increasing the degree of extreme advice that patients receive from their primary physicians with financial conflicts.

4.3  Undivided Physician Loyalty, Proportionality, and Accountability The mere existence of a COI is not necessarily a problem.113 Some secondary interests are inevitable and/or vitally necessary. For example, shared savings payments for ACOs in the Medicare program, while generating potential financial COIs for the participating providers, are a novel way to improve care delivery efficiency and counter long-­ standing concerns about overtreatment, waste, and fragmentation.114 This potential for concurrent upside along with more worried-­about risks significantly complicates effective COI regulation. US health law continually struggles to find optimal balance in addressing conflicts that pose real risk of bias while not overly deterring important secondary incentives that advance larger societal goals beyond the immediate patient’s welfare. Legal rules that require the provider to maintain strict fidelity to the patient, always above all other secondary interests, might be simpler to understand and offer a comforting picture of physicians’ zealous dedication to healing the

110  See generally Jerry Avorn, Academic Detailing: Marketing the Best Evidence to Clinicians, 317 JAMA 361 (2017). 111  Lisa Rosenbaum, Beyond Moral Outrage: Weighing the Trade-­Offs of COI Regulation, 372 New Eng. J. Med. 2064, 2065 (2015). 112  Sunita Sah & George Loewenstein, Conflicted Advice and Second Opinions: Benefits but Unintended Consequences, 130 Organizational Behav. & Hum. Decision Process 89, 89 (2015). 113  IOM Conflict of Interest Report, supra note 4, at 61. 114 Gatter, supra note 31, at 244.

534   Richard S. Saver patient. However, legally demanding singular loyalty to the patient is likely not only infeasible but may be ill-­advised.115 As for impracticality, physicians face so many secondary interests that avoidance may be impossible and zealous management of all extremely burdensome. For example, COIs associated with providing too much or too little care may simply be inevitable in any form of physician reimbursement.116 Moreover, a legal regime that strictly demands the physician’s singular loyalty to the patient could undermine the important obligations physicians have to other patients, to the community, and to larger societal goals. Conceiving of the physician’s role as solely to benefit the patient in front of her risks introducing “relational bias” into healthcare decision-­making, with negative effects for health policy. This discourages frank consideration of how the physician’s decisions to “do best” for an individual patient impact the overall healthcare system and larger societal interests.117 It also can divert attention away from examining whether delivering care differently, such as with innovative delivery structures or financial incentives for efficient care, might improve access overall within the system and ultimately benefit more patients. For example, physicians, in the role of gatekeeper and prescriber, “are in a unique position to affect healthcare spending.”118 Thus, the American Medical Association’s (AMA) Council on Ethical and Judicial Affairs advises that physicians have an obligation to act as prudent stewards of healthcare resources in both their treatment and advocacy roles. The AMA guidance specifically disputes arguments that physicians should never pursue considerations other than the welfare of the patient before them. According to the AMA, the singular patient welfare view does not reflect the realities of clinical practice and overlooks physicians’ distinct obligations to promote public health and access to care generally.119 Thus, it remains critically important to account properly for the costs of potential over-­deterrence in COI oversight. As previously discussed, even a seemingly straightforward and “light” approach to regulation such as mandatory disclosure of COIs may bring unintended downsides, including increased costs, informational fatigue, and skewed behavioral responses by patients and providers.120 Consideration of proportionality significantly complicates the regulatory choices. For example, transactional 115  See Sam F. Halabi & Anthony J. Phillips, Against Fiduciary Utopianism: A Study of Legislative and Judicial Management of Physician Conflicts of Interest and Standards of Care in US Law 11 U.C. Irvine L. Rev. 433 (2020). 116  See, e.g., Frances H. Miller, Doctors’ Conflicts of Interest (and Altruism) in the United States and Great Britain, 27 Ind. L. Rev. 687, 692–693 (1994). 117  See generally William M. Sage, Relational Duties, Regulatory Duties, and the Widening Gap Between Individual Health Law and Collective Health Policy 96 Geo. L. J. 497 (2008). 118  Amer. Med. Ass’n Opinion 9.0652-Physician Stewardship of Health Care Resources, reprinted in 17 AMA J. Ethics 1044 (2015). 119  Amer. Med. Ass’n, Reports of the Council on Ethical And Judicial Affairs: Physician Stewardship of Health Care Resources (2012), https://www.ama-­assn.org/sites/default/files/ media-­browser/public/about-­ama/councils/Council%20Reports/council-­on-­ethics-­and-­judicial-­affairs/ ceja-­1a12.pdf. 120  See notes 85 to 96 supra and accompanying text.

Provider Conflicts of Interest in US Healthcare   535 bans limiting certain industry–medicine financial relationships may seem cleaner and more direct regulation than mandatory disclosure. But such aggressive approaches could also lead to slower development of important therapies and the lack of needed expertise in development of clinical practice guidelines. These lost benefits often are less visible and salient than the risks of potential bias arising from COIs.121 In addition to proportionality, accountability considerations further bedevil regulation of COIs. Accountability is extremely difficult to ensure because of the complicated dynamics at play. COIs arise in a variety of ways (clinical trial support, consulting agreements, equity interests, professional rivalries, etc.), among many different players (physician peers, payers, manufacturers, teaching hospitals, investigators, etc.), and at various levels (sole practitioner, larger medical group, academic department, research institution, etc.), resulting in difficult identification problems and diffusion of responsibility. Also, the practical enforcement of COI rules, at the institutional level, remains often too dependent on provider self-­reporting in the first instance, and, ultimately, the internal conflicts oversight committees that, as previously discussed, likely vary dramatically in their expertise and ability to both detect and manage problematic conflicts.122 The failure to subject conflicts oversight to rigorous assessment means that COICs operate without a strong empirical basis for determining which conflicts pose greater risk of bias and which management plans prove effective. Meanwhile, the resulting wide discretion afforded to COICs leaves little room for holding the conflicts monitors themselves accountable.

5 Conclusion Provider COIs within US healthcare are prevalent and correlate with serious risks. However, robust, effective legal oversight has proved arduous. Provider COIs implicate a confusing patchwork of overlapping legal sources that are limited in application and can impose difficult procedural and proof burdens on aggrieved patients. The primary response under current legal rules, mandatory disclosure, is necessary but insufficient. While increased transparency about COIs through initiatives like the Sunshine Act holds promise for improving healthcare system operations in the long run, much more work remains to be done. Unfortunately, COIs, and financial conflicts in particular, are unwieldy subject matter for regulation for many reasons, including the complex transactions involved, the necessary importance of contextual and comparative information, the seeming low concern and disregard about financial conflicts among patients, and the difficulty in ensuring accountability because of the complicated dynamics at play. More fundamentally, US health law continually struggles to find optimal balance and proportionality. 121 Rosenbaum, supra note 111, at 2065. 122  See notes 71 to 75 supra and accompanying text.

536   Richard S. Saver Some COIs seem unavoidable and systemic. Indeed, providers performing the critical gatekeeper and learned intermediary roles remain embedded between their patient, other patients, and larger society. If the physician always privileges the welfare of the patient in front of her, difficult relational bias for healthcare operations results. Furthermore, a zero tolerance approach for COIs can also lead to important lost benefits, such as product innovation and harnessing provider self-­interest to improve the efficiency of healthcare in an era of limited resources. Thus, the unfailingly devoted physician may be appropriately seen as a relic of the past. At the same time, patients remain vulnerable and heavily dependent on their providers, and physicians still command a great deal of healthcare resources through discretionary treatment decisions. All of this underscores the need for continual, vigorous management of COIs. Finding the right balance for secondary interests to influence providers in ways beneficial to the overall healthcare system while maintaining sufficient patient loyalty remains one of the greatest challenges for US health law moving forward.

Acknowledgments Special thanks to MacKenzie Dickerman and Michael Sindoni for excellent research assistance.

Chapter 27

Prov ider Con flicts of I n ter est i n Eu rope a n Hea lthca r e Trustworthy Physicians Sunčana Roksandić

Until recently, conflicts of interest (COIs) resulting from interactions of physicians with the pharmaceutical industry were not among the most systematically researched and analyzed areas of health law in Europe.1,2 It was somehow the golden rule that physicians are aware of risks that different roles in their professional lives create and are able to manage those risks to the best of their abilities in a way that is not, at the end of the day, contrary to the interests of their individual patients. Physicians understood the scope of acting on the principle of good faith or applying the duty of care standard in general. But this viewpoint created a problem. A phenomenon called ‘bias bias’ contributes to conflicts of interest situations across Europe: ‘while most people are ready to accept the possibility of bias in others, few are ready to acknowledge that they themselves might be biased.’3 The medical community seemed especially susceptible to bias bias. According to Giovanni Fava, ‘from the 1990s an impressive body of fact pointed to the danger of 1  The title is in line with para 3 of the Council of Europe, Parliamentary Assembly Resolution 2071 (2015): Public Health and the Interests of the Pharmaceutical Industry: How to Guarantee the Primacy of Public Health Interests. 2  Up to 2010, as underlined by Lexchin and O’Donovan, there have been a few classic studies of pharmaceutical regulation across Europe, but a systematic review of conflict of interest policies have been neglected. Joel Lexchin and Orla O’Donovan, ‘Prohibiting or ‘Managing’ Conflict of Interest? A Review of Policies and Procedures in Three European Drug Regulation Agencies’ (2010) Soc Sci & Med 70, 643–647, at 643. 3  Z Sharek, RE Schoen, and G Lowenstein, ‘Bias in the Evaluation of Conflict of Interest Policies’ (2012) 40 J. L. Med. 6 Ethics 368, 368–382, at 368.

538   Sunčana Roksandić conflict of interest in medicine. . . . [S]uch disclosure often took place in the media, instead of coming from the scientific community.’ In addition, corporate-­driven special interests were important. As Fava continues, ‘as with many cultural and scientific phenomena, issues pointing to the developing dangers of specific situations are initially overlooked, if not deliberately denied. Major medical journals have played a major role in such massive denial.’4 In any case, the phenomenon of COI was explicitly acknowledged in 2010 by the Council of Europe Parliamentary Assembly (CoE PA) Resolution 1749 (2010) ‘Handling of the H1N1 Pandemic: More Transparency Needed,’ where the Parliamentary Assembly expressed its concerns about the risks of COIs among experts involved in sensitive health-­related decisions. In this Resolution, the Parliamentary Assembly repeated its concerns, dating back to 2005, about good governance in the healthcare field and dangers of unregulated or secret lobbying. This Resolution expressly calls for all persons subject to COIs to be excluded from sensitive decision-­ making processes, for responsibility to be taken for any errors made, and for re-­establishment of public confidence in public health decisions and advice. Therefore, today, in the public arena across Europe, more emphasis is given to the need to regulate COIs arising from interactions with the pharmaceutical industry, although it is only a part of the COI phenomenon. In the United States it is estimated that almost 94% of physicians enter into some kind of relationship entailing economic benefits with pharmaceutical industry.5 The latest research on such interactions in Europe6 included relationships that could be research-­ oriented or could refer to drug promotion practices through direct marketing, funding continuing medical education, or using indirect marketing such as health campaigns, ghost-­writing7 or astroturfing.8 Types of interactions also include giving gifts, drug samples, reimbursements, and payments for a range of ‘services’, such as consulting, serving as a speaker or on a speaker’s list, on an advisory board, or for enrolling patients in clinical trials.9 All of those implicated patients’ safety and public health and led to enhanced regulation at the EU level. However, the relationship between physicians and 4 Giovanni F. Fava, Conflict of Interest and the Credibility of Clinical Medicine. A Psychosomatic Antidote, International Congress Series 1241 (Elsevier Science B.V. 2002), 229–236, at 230, 232. 5  EG Campbell, ‘Doctors and Drug Companies: Scrutinizing Influential Relationships’ (2007) N Engl J Med, 357, 1796–1797, as cited in Mental Health Europe, Shedding Light on Transparent Cooperation in Healthcare: The Way Forward for Sunshine and Transparency Laws Across Europe (hereafter MHE Study) (2019) 9. 6  MHE Study (n5) from January 2019. 7  Scientific articles prepared by professional copywriters employed directly by pharmaceutical companies and then signed in the name of a well-­known expert in the medical field, with their consent, to give the report more prestige. (MHE Study [n5] p. 10; based on Kliner M., ‘The interaction between the pharmaceutical industry and healthcare professionals’ (2012) A PharmAware Resource. 8  ‘Astroturfing’ refers to cooperating with patients’ organisations (e.g., through the funding of health campaigns) or even establishing new health-­related, seemingly bottom-­up organisations. Astroturfing masks the sponsors of a message and gives the public the impression of widespread grassroots support for a product or policy. MHE Study (n7), p. 9/10. 9 Id.

Provider Conflicts of Interest in European Healthcare   539 the pharmaceutical industry is inevitable: for example, drug companies rely on medical expertise in designing their studies and on physicians to conduct their clinical trials. And these relationships involve other conflicts, as between a physician’s commitment to advancing medical understanding through research and his or her duty to care for their patients. Very often throughout Europe, COI could be found under explanations of the notion of “corruption” as indicated in the European Commission’s Updated Study from September 2017 on Corruption in the Healthcare Sector. In the study, “corruption” was defined as ‘the abuse of power for private gain—that also encompasses aspects that go beyond the criminal law aspects, including situations such as conflict of interest, favouritism, etc.’10 Estimates from the European Healthcare Fraud and Corruption Network calculate an approximate €56 billion annual loss to Europe as a result of corruption.11 The magnitude of corruption and social fraud ranges between 3% and 8% of national health expenditure. As underlined by Sommersguter-­Reichmann et al. in a resolution in the German Bundestag on ‘combating corruption in healthcare,’ up to 10% of public expenditure was quoted. The European Commission’s Updated Study from September 2017 on Corruption in the Healthcare Sector concluded that bribery in medical service delivery remains one of the main challenges, especially in many Eastern and Southern European Member States. According to the results, corruption related to granting privileged access to healthcare or potential risks involving double practice are not isolated only to EU Member States with a high perception of corruption (in healthcare). Furthermore, the Updated Study concluded that transparent procedures are key to addressing corruption in procurement processes, and attempts to address improper marketing should increase at both EU and national levels. These data pinpoint the existence of institutional corruption within healthcare institutions, referring to a situation where the institutional setting generates COI, which promotes a behavior in those who perform duties within the institution that systematically compromises the institution’s purpose. The problem of pervasive corruption in the health sector is not limited to a particular country or continent (see further discussion in Section 1 of this chapter). As noted in the joint introduction to these chapters, not all COIs are immediately negative. Especially if one also understands COI situations where competing considerations are presumed to be legitimate; for example, performing a role of treating physician but also conducting medical research (the duty to care for patients and the duty to engage in research), (not) keeping the medical condition of a patient who has hereditary illness secret from family members, saving the life of a pregnant woman rather than the life of her future child, recognizing the possibility of conscientious 10  Report from the European Commission: Updated Study on Corruption in the Healthcare Sector Final Report, September 2017 (hereinafter Updated Study), p. 149. 11  See more in Margit Sommersguter-­Reichmann, Claudia Wild, Adolf Stepan, Gerhard Reichmann, and Andrea Fried, ‘Individual and Institutional Corruption in European and US Healthcare: Overview and Link of Various Corruption Typologies’ (2018) Appl Health Econ Health Pol, 16, 289–302, at 289. Also see Updated Study September 2017.

540   Sunčana Roksandić objection, etc. Some, to avoid confusion, refer to these last described situations as conflicting interests, not as COIs. As in the United States, in Europe ‘there is no consensus about which sources of influence should be treated as conflicts of interest, not all conflicts are easily avoidable, and for certain conflicts appropriate management techniques remain elusive.’12 Hence, to avoid any doubt and to underline that not all COIs are negative, COI should be primarily understood as a risk factor that increases the probability of occurrence of misconduct or acting contrary to the principle of good faith13 (see later discussion). This chapter, after providing definitions, examines the special nature of the physician– patient relationship in Europe and the main principles that a physician needs to follow in rendering medical care, including informed consent rules. The chapter also examines the role of law and self-­regulation in addressing provider COIs within European healthcare, including medical malpractice and anti-­bribery law, both civil and criminal. In spite of common fundamental principles as enshrined in the Council of Europe Convention on Human Rights and Biomedicine, current rules regulating COIs are limited in scope and diverse in many countries across Europe (see more in Section 2). In addition, when one analyzes the relationship between the pharmaceutical industry and healthcare professionals, it can be concluded that the European legal order regulates this relationship and its COIs through full legislation,14 partial legislation,15 or anti-­ corruption and transparency regulation16 (as detailed in Section  3). Therefore, this chapter also explores whether more stringent solutions, such as application of anti-­ corruption statutes, could be more suitable for preventing financial and nonfinancial COIs in health contexts, especially the most severe cases of COI resulting in serious consequences for patients and/or national health systems. As in the chapter on COI in the United States, this chapter does not consider in detail potential application of the healthcare fraud and abuse laws to COI as fraud and abuse regulation is addressed in a separate chapter of this volume.17

12  See Chapter 14 in this volume. 13  MHE Study (n5), p. 11. 14  Such legislation refers to a comprehensive effort by states to address a legal problem. Thus, for example, a state might enforce its disclosure obligations on the entire healthcare sector, including industry, healthcare professionals, and healthcare organisations, and also assume responsibility for monitoring the disclosure, as with the US Sunshine Law. For countries with full legislation, transparency is required when it comes to various transfers of value from the industry to healthcare professionals. 15  Legal obligations are included in other more general legal frameworks, or legal frameworks only include certain obligations, or only apply to certain stakeholders of the healthcare sector without being as comprehensive as what we consider full legislation. Such provisions cover, for example, obligations to disclose information about payments to a regulatory agency but not to the public, or requirements for the pharmaceutical industry, which do not apply to other industries (MHE Study [n 5] 2019, p. 23. 16  MHE Study (n5) 2019. 17  See the Chapter by Tracy Elliott in this volume.

Provider Conflicts of Interest in European Healthcare   541

1  Definitional Issues According to a leading European study of the issue, conflict of interest appears when an individual has personal, financial, professional or political interests that are likely to undermine his or her ability to meet or fulfil his or her primary professional, ethical or legal obligations. A conflict of interest is considered to be a risk factor, which increases the probability of occurrence of various types of misconduct. The problem is therefore not primarily the conflict of interest itself, but the biased or distorted judgment or action of those who have such conflict.  (MHE study [n5] p. 11)

Similarly, according to a widely used EU definition, a COI ‘refers to a situation where the impartiality and objectivity of a decision, opinion, or recommendation . . . is or might be perceived as being compromised by a personal interest held or entrusted to a given individual.’18 The European Study stressed that a COI provokes a combination of stimuli and motivators that can have harmful effects through their influence on judgments and actions.19 Those stimuli could be of financial or nonfinancial nature. Nonfinancial COIs may be political, academic, personal, ideological, or religious. The joint introduction to this chapter discusses some examples. In general throughout Europe, COI gives a broad definition of a situation in which a person is or appears to be at risk of acting in a biased way because of other interests. As indicated earlier, not all COIs have a negative effect or reflect physician self-­interest. Sometimes the physician faces competing legitimate interests, as with the conflict between the duty to care for patients and the duty to engage in research. However, regardless of the motives, the mere existence of a conflicting duty creates concerns because ‘the confidence of others in [provider] judgment is of paramount importance . . . when trust in [provider] judgment is impaired, the role of [providers] is diminished.’20 Accordingly, as Dennis Thompson has written, it is necessary to prevent conflicting interests from dominating or appearing to dominate the relevant primary interest in professional decision-­making.21 And conflicting interests can be surprisingly influential. According to one study, even small gifts and meals may alter physician decision-­making. 18  This definition refers to the acts of the EU Decentralized Agencies as provided in the Guideline for Conflict-­of-­Interest Policies in EU Decentralized Agencies, 10 December 2013, p. 5. Compare to TL Beauchamp and JF Childress, Principles of Biomedical Ethics (Oxford University Press 2013) 328. 19  MHE Study (n5), here citing DB Resnik and E Shamoo, Responsible Conduct of Research, 2nd ed. (Oxford University Press 2009). 20  See Introduction to this chapter; see also Harvey V. Fineberg, ‘Conflict of Interest: Why Does it Matter?’ (2017) 317 JAMA 1717. 21  Dennis F Thomson, ‘Understanding Financial Conflicts of Interest’ (1993, 19 August) 329(8) N Engl J Med, 573–576, at 573.

542   Sunčana Roksandić Hence, all definitions of COI in medicine used within Europe indicate a potential risk of not acting in good faith regardless of a motive. COIs might undermine the professional responsibility of physicians to put patients first. Throughout Europe, both financial and nonfinancial interests22 are considered risks to decision-­making. The legal responses to each are different in part because of their different characteristics (e.g., financial interests are easier to quantify). But one can easily fall into the trap of thinking that the obligation for physicians to disclose payments received from the pharmaceutical industry is a sufficient response to the problem of COIs. However, such an obligation will not prevent nonfinancial interests from influencing decision-­making. Further regulation is necessary.

2  Legal Background: Obligations to Avoid or Manage Conflict of Interest in Europe As stated in the “Report of the UN Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest Attainable Standard of Physical and Mental Health” in 2017, healthcare is among the most corrupt sectors in many countries. Health sector corruption, including, for example, the bribing of health officials and unofficial payments to healthcare providers, obstructs the ability of states to fulfil their duty to guarantee available, accessible, and good quality health services, goods, and facilities. Moreover, corruption has significant implications for equality and nondiscrimination since it has a particularly marked impact on the health of populations in situations of vulnerability and social exclusion, in particular children and those living in poverty. Furthermore, in the Report, it was emphasized that there is a need to apply a human rights-­based approach to health and related policies and avoid selective approaches to human rights. European countries follow Art. 12. of the United Nations International Covenant on Economic, Social and Cultural Rights (ICESCR) in their effort to guarantee the right to health across Europe. Therefore, there is always a possibility in Europe to invoke the violation of the right to health when serious issues of COI arise (e.g., vaccination was denied to the population because of the economic interests of physician members of the committee in charge of the approval of vaccination). 22  See, e.g., X Bosch, JM Pericas, C Hernández, et al. ‘Financial, Nonfinancial and Editor’s Conflict of Interest in High-­Impact Biomedical Journals’ (2013) 43(7) Eur J Clin Invest, 660–667. The finding of the research were as follows: author’s financial COI disclosures were required by about 90% of high-­impact clinical and basic journals publishing original research. Unlike recent studies showing a significantly lower prevalence of nonfinancial compared with financial disclosures, the former were required by about 70% of journals, suggesting that editors are increasingly concerned about nonfinancial competing interests. Only 40% of journals required disclosure of editor’s COI, in conflict with the recommendations of the most influential editor’ association.

Provider Conflicts of Interest in European Healthcare   543 These principles are reflected in European constitutional settlements. The right to healthcare across Europe is generally perceived as a basic human right, sometimes constitutionally embedded, and the expectation of having impartial physicians who act according to the highest ethical standards is a deeply rooted aspect of this human right across Europe.23 From the legal point of view, physicians across Europe are expected to act in ‘good faith’ when providing healthcare. That also follows from the Council of Europe’s Convention on Human Rights and Biomedicine that serves among the most important legally binding instruments in setting the boundaries of acting in good faith. It sets out fundamental principles applicable to daily medical practice and is directly applicable at the national level. The Convention on Human Rights and Biomedicine is the first legally binding international text designed to preserve human dignity, rights, and freedoms through a series of principles and prohibitions against the misuse of biological and medical advances. The Convention’s starting point is that the interests of human beings must come before the interests of science or society, and the whole Convention is inspired by the principle of the primacy of the human being; therefore, all its articles must be interpreted in this light.24 This is also a clear principle for risks coming from COIs. For example, as the individual human being has a priority, the desire for medical research could not breach the primary due to render medical care to a patient. In interpreting healthcare-­related rights, the European Court of Human Rights is particularly attentive to other legal and policy materials relating to health which have been adopted within the framework of the Council of Europe.25 In that sense, the Council of Europe’s European Social Charter guarantees the right to protection of health, and the European Code of Social Security regulates the right to medical care. Hence, the rights to health and healthcare represent the fundamental values of each European country. These rights are often additionally protected by the highest state act, usually the Constitution/Basic Law. Accordingly, agreed-­upon European-­level international human rights norms can be read as defining responsibilities for all stakeholders, including the principles upon which medical professions are regulated. In compliance with these broader European-­level international law obligations and (often) constitutional protections, across Europe, special national legislation and more detailed regulation put flesh on the bones of a ‘right to health’ by providing concrete guarantees of various aspects of a right to health and healthcare for individuals, including protections from COIs. National legislation also sets out in detail the scope of 23  For EU health law and policy, see Tamara K Hervey, Calum Alasdair Young, Louise E Bishop, and Edward Elgar, Research Handbook on EU Health Law and Policy (Edward Elgar Publishing 2017); Tamara K Hervey and Jean V McHale, Health Law and the European Union (Cambridge University Press 2004); and André den Exter (ed.), European Health Law (Maklu 2017). 24  Explanatory Report to the Convention for the protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine, para 30: https://rm.coe.int/16800ccde5 25  See European Court of Human Rights, Thematic report, Health-­Related Issues in the Case-­Law of the European Court of Human Rights, Council of Europe, 2015, p. 4.

544   Sunčana Roksandić the duty of care of physicians. In countries with a common law tradition, court precedents play an equally important role.26

2.1  Physician–Patient Relationship European countries guarantee patients’ rights to be protected from harmful COIs through professional codes of ethics (drawn up by the state or the profession), codes of medical conduct, general obligations in civil codes, and/or health legislation. The content of professional standards, obligations, and rules of conduct is not identical in all countries.27 Specific medical duties may vary slightly from one society to another. However, the right to healthcare and the fundamental principles of the practice of medicine that flow from it apply in all countries. Doctors and, in general, all professionals who participate in a medical act are subject to legal and ethical imperatives. They must act with care and competence and pay careful attention to the needs of each patient.28 This refers to acting in good faith and fulfilling the duty of care. By this, medical ethics effectively becomes a binding source of law once adopted by national medical chamber.29 As underlined in the joint introduction to these chapters, the specific sources of obligations to deal with the different aspects of potential COIs are varied. They include acting in good faith, applying informed consent rules, and tortious liability, as well as detailed obligations provided in national civil codes and statutory/regulatory law. If one were to seek a unified European approach towards COIs, all of this specificity would have to be taken into account. However, a point of commonality, as underlined in the previous Section, is that all European countries follow internationally recognized human rights. For example, Croatia, following the right to be free from torture and other cruel, inhuman, or degrading treatment or punishment has, in 2015, explicitly 26  See, e.g., the Supreme Court of UK decision in the Montgomery case (Montgomery v Lanarkshire Health Board [2015] SC 11 [2015] 1 AC 1430). The decision requires physicians to inform patients of all reasonable treatment options, including nontreatment and treatment options that the physician may not personally provide. Thus in a private physician–patient relationship it may be against a physician’s own financial interest to advise of those other treatment options. The duty of candour also requires a physician to inform patients of material mistakes, which may be contrary to that physician’s own interest. The ruling overturned a previous decision by the House of Lords (Sidaway v Board of Governors of the Bethlem Royal Hospital and others [1985] 871 AC) which had been law since at least the mid-­1980s. See more in Chan Sarah, Tulloch E, et al., ‘Montgomery and Informed Consent: Where Are We Now? (2017) Br Med J 357, j2224. 27  See Sara Patuzzo and Elisabetta, ‘Towards a European Code of Medical Ethics: Ethical and legal issues’ (2017) J Med Ethics 43, 41–46, at 44: ‘In some countries the codes of medical ethics have a legal value forming a specific source of law. In other countries, the code of medical ethics is not a specific source of law, but formally an internal regulation of the medical profession, which may however gain relevance within the legal system, especially through case law and disciplinary responsibility.’ 28  Explanatory Report (n27), para 32. 29  While the specifics may vary from country to country, national medical chambers tend to combine the roles in the United States of professional associations, accrediting bodies, and licensing boards.

Provider Conflicts of Interest in European Healthcare   545 forbidden conducting biomedical research on people with mental disabilities if they are not able to provide an informed consent on such research. No surrogate decision is allowed as a substitution.30 Following the UN Convention on the Rights of Persons with Disabilities, it was concluded that the conflict between the primary duty of physicians to take care of the patient and the interest in conducting medical research posed too great a risk to patients with mental disabilities such that they cannot validly consent to the research. As stated, statutory obligations are among the sources of conflicts law. Those regulations are usually developed by professional organizations of healthcare providers. On the European level, there are various umbrella associations of physicians, national medical chambers, and students of medicine which are significant for providing guidelines and recommendations for national implementation of COI oversight. Examples include the European Council of Medical Orders, European Charter of Medical Ethics 2011,31 and the European Medical Students’ Association’s Policy Paper on Conflicts of Interest in Medical Education Settings 2019. Although the European Charter of Medical Ethics is not legally binding per se, it could become an important source of law if accepted by national medical professional associations as the basis of their codes of ethics and deontology. In addition, every European country has implemented criminal law provisions that serve as ultima ratio punishment for physicians who, either through negligence or intent, breach the duty of care, including failing to respect codes of ethics and, by that, creating undesired consequences for patients. This is particularly relevant for combating corruption as defined in the Updated Study on corruption in the health sector. All of the different sources of regulation must be taken into account when analyzing responses from European countries on COI in health settings.

3  Obligations to Avoid or Manage Conflicts of Interest Under European Regulation As in the United States, the legal obligation for healthcare providers to avoid or manage COIs may seem self-­evident. Such a rule arises from overlapping sources, including acting in accordance with the good faith principle and the common law duty of care. 30  See Roksandic Suncana, Galiot Vinko, and Roksandić Vidlička, ‘Eksperimenti na ljudima kao zločin protiv čovječnosti: Od Nürnberškog Medicinskog suđenja do predmeta Pfizer [Experiments on human beings as a crime against humanity: From Nürenberg Medical Trials to Pfeizer]’ (2016) 7(1) Godišnjak Akademije pravnih znanosti Hrvatske, 186–253; and Croatian Law on Protection of Persons with Mental Difficulties 2015, Art.12 (2). 31  Available online: http://www.ceom-­ecmo.eu/sites/default/files/documents/en-­european_medical_ ethics_charter-­adopted_in_kos.pdf. For the feasibility of a common European code of medical ethics, see Patuzzo and Pulice, supra 28, at 41–46.

546   Sunčana Roksandić There are additional special laws governing the provider–patient relationship (e.g., Physician’s Act or Law on protection of patients’ rights) and statutory or regulatory requirements in every European country. In countries with common law traditions, court decisions play an equally important role. The role of the European Convention on Human Rights is also indispensable for every European state, and many stakeholders within the health sector are invoking breaches of the Convention in rendering healthcare.32 A physician’s informed consent duty, as under US law, generally obligates disclosure of the proposed treatment, its likely benefits and material risks, and the medically viable or medically reasonable alternatives to the physician’s recommendation.33 This, if applied with good faith, should minimize the risk that a physician’s personal interests or a secondary duty, such as to research, will compromise the primary duty to patient health. As a prerequisite for informed consent, the advice from a physician must be given to a patient without ulterior motive (e.g., recommendation for medical treatment because it is necessary for the patient, not in order to test a new medical product produced by a colleague’s company). Throughout Europe, the most severe cases of ­malpractice and corruption are punished as criminal offences. In what follows, the emphasis is on how COI is regulated in addition to these just discussed sources. Following the US example, the EU is trying to additionally regulate potential COIs between the pharmaceutical sector and healthcare providers by imposing stricter obligations for the pharmaceutical industry on transparency and disclosure. These recent instruments prescribe more stringent regulation of the relationship between the pharmaceutical industry and healthcare providers to minimize situations in which a physician is or appears to be acting in a biased way because of personal interests. Because of diversity of thought in Europe and the existence of different traditions, the this Section is divided into the following subsections: 3.1, on relevant binding EU law; 3.2, on self-­regulation at the EU level; 3.3, on relevant binding national law, including voluntary codes at national level; and 3.4, on Council of Europe recommendations and European-­level ethical codes.

3.1  EU Law Directive 2001/83/EC, referred to as the EU Pharmaceutical Code, applies to all medicinal products for human use intended to be placed on the market in EU Member States either prepared industrially or manufactured by a method involving an industrial process.

32  E.g., as commented by Gilbar and Foster, ABC v ST George’s Healthcare NHS Trust is a ‘first case to reach the English courts regarding the duty of physicians to communicate genetic information to a patient’s relatives. In 2015, the High Court struck out a claim by a patient’s daughter against her father’s doctors, holding that they neither owed her a common law duty of care nor breached Article 8 of the European Convention on Human Rights.’ 33  See Chapter 14 of this volume, on COI in the United States.

Provider Conflicts of Interest in European Healthcare   547 As already underlined, the duties of informed consent and respect for the s­elf-­determination of patients require sufficient and accurate information so that patients can make an informed decision. This Directive regulates the information and the form in which the product has to be presented to the patient. It stipulates that the essential aim of any rules governing the production, distribution, and use of medical products must be to safeguard public health. Advertising significantly contributes to the information available to persons qualified to prescribe or supply drugs and, as such, remains desirable. However, it should be subject to stringent conditions and effective monitoring because health professionals must be able to perform their functions objectively, without being influenced by direct or indirect financial incentives from the industry. When medical products are promoted to persons qualified to prescribe or supply them, no gifts, pecuniary advantages, or benefits in kind may be supplied, offered, or promised unless they are inexpensive and relevant to the practice of medicine and pharmacy. The Directive also addresses the relationship between the pharmaceutical industry and its regulators. According to Article 126b, in order to guarantee independence and transparency, the EU Member States shall ensure that members of staff of the competent authority responsible for granting authorisations, and rapporteurs and experts concerned with the authorisation and surveillance of medicinal products, have no financial or other interests in the pharmaceutical industry which could affect their impartiality. These persons shall make an annual declaration of their financial interests. In addition, the Member States shall ensure that the competent authority makes publicly accessible its rules of procedure and those of its committees as well as agendas for its meetings and records of its meetings, accompanied by decisions taken, details of votes, and explanations of votes, including minority opinions. Again, as emphasized by Mental Health Europe’s Report from 2019, it is up to the EU Member States to take all measures necessary to this end in light of their own individual context. In general, EU laws regulating pharmaceutical promotion do not include provisions like those found in the US Sunshine Act, which requires public disclosure of financial relationships between industry and physicians, but some Member States have adopted stricter rules as national law, as discussed next.

3.2  Self-­Regulation on the EU Level Self-­ regulation has particularly targeted financial relationships between the pharmaceutical industry and healthcare providers. As emphasized in the Mental Health Europe’s Report from 2019, the European Commission set up the Process on Corporate Responsibility in the Field of Pharmaceuticals (2010–2013) to facilitate discussions on ethics and transparency (and also to address matters of access to medicines). Through one of the three platforms launched, the Platform on Ethics and Transparency in the pharmaceutical sector, the European Federation of Pharmaceutical Industries and Associations (EFPIA) has adopted a List of Guiding Principles Promoting Good

548   Sunčana Roksandić Governance in the Pharmaceutical Sector that aims to ensure that patients receive proper treatment and are provided with relevant, factual, and unbiased information. In June 2013, the EFPIA adopted a Code on ‘Disclosure of Transfers of Value from Pharmaceutical Companies to Healthcare Professionals and Healthcare Organisations.’ Transfers of value in the Code are understood as ‘direct or indirect, whether in cash, in kind or otherwise, made either for promotional purposes or otherwise in connection with the development and sale of prescription-­only medicines exclusively for human use.’ This Disclosure Code ‘although not being part of generally applicable law, is regarded as a turning point in European Sunshine policy.’34 Based on this Disclosure Code, national associations should introduce national codes of conduct if national sunshine regulation was not already implemented in their respective countries. For example, the French Industry Association stated that applying the French Sunshine Law fulfilled its obligations under the EFPIA Code. Stakeholders are obliged to document and disclose transfers of value, whether done directly or indirectly to or for the benefit of the recipient. Based on the EFPIA Code, then, pharmaceutical companies will disclose full data on payments made to physicians across Europe. The following items are required to be disclosed on companies’ own website or on a central platform, such as the one provided by the relevant government, regulatory, or professional authority or a member association: payments made to healthcare professionals (including sponsorships to attend meetings, speaker fees, and consultancy and advisory board fees), donations and grants paid to healthcare organisations, and payments made for research and development. The disclosure will include full names, addresses, and specialties of each healthcare professional. However, if healthcare professionals do not consent to their personal data being disclosed individually, their expenses shall be reported in the aggregate. For physicians who revoke consent after a report has been published, the company is not required to remove the data from the published report, but instead, for subsequent reports, must anonymize the healthcare providers’ data and move it to the aggregate section. However, unlike the US Sunshine Act, the EFPIA Code does not require disclosure of payments made for meals and drinks below a certain value. In addition, transfers of value related to research and development are reported in an aggregated form. In a similar vein, the EFPIA Code of practice on the relationship between the pharmaceutical industry and patient organizations35 lays down transparency obligations by which each company must make publicly available a list of patient organizations to which it provides financial support or significant indirect or nonfinancial support.

34  As underlined in the Mental Health Europe Report (MHE Study, n5) 2019, p. 31. 35  EFPIA Code (adopted on 22 March 2019, ratified 27 June 2019): https://www.efpia.eu/ media/413022/efpia-­code-­2019.pdf

Provider Conflicts of Interest in European Healthcare   549 Regarding sanctions for self-­regulation, as underlined by Alice Fabbri and colleagues,36 all the codes that are entirely voluntary treat nondisclosure as a breach of the code of conduct and treat this particular breach as similar to other breaches. Therefore, the national laws discussed in the next section must be invoked in sanctioning breaches by pharmaceutical companies. As this brief overview shows, transparency by disclosure seems to be taken as a recent key element for preventing COIs in EU law, following the approach of the US Sunshine Act. As discussed in the next section, the same seems to be a trend in national jurisdictions. These national disclosure laws can be seen as a complement to the well-­ accepted obligation to follow the principle of acting in good faith and fulfilling the duty of care, the duty to ensure informed consent, and the duty to observe ethical rules established by national medical councils or chambers (self-­regulation) and the potential for tort and criminal liability for the most severe forms of medical misconduct.

3.3  Relevant Binding National Laws and Voluntary Codes The relationship between the pharmaceutical industry and healthcare professionals concerning COIs throughout Europe is specifically regulated, generally, through so-­ called full legislation (e.g., Belgium, Denmark, France, Portugal, Romania, Slovakia),37 partial legislation (e.g., Greece, Latvia, Turkey), or just by following anti-­corruption and transparency regulation at the national level without particular requirements for additional reporting and public disclosure (e.g., Croatia, Germany, Hungary, Italy, the Netherlands, Slovenia, Spain, Sweden, UK).38 Full legislation exists when ‘states enforce disclosure obligations on the industry, healthcare professionals and healthcare organisations and are responsible for monitoring the disclosure. For countries with full legislation, transparency is required when it comes to various transfers of value from the industry to healthcare providers.’39 In general, pharmaceutical companies and/or healthcare providers ‘have to disclose information about their cooperation at the individual level, preferably in public databases. National legislation can set a threshold for disclosure and a range of fines for breaching the rules. Usually, there is no consent of [healthcare providers] required for disclosure.’40 Some European countries have ‘partial sunshine provisions’: legal obligations for pharmaceutical companies are included in more general legal frameworks or only certain obligations are regulated by certain stakeholders. Unlike in countries with full legislation, provisions in ‘partial legislation’ countries include for example, obligations 36  A Fabbri, A Santos, S Mezinska, S Mulinari, and B Mintzes, ‘Sunshine Policies and Murky Shadows in Europe: Disclosure of Pharmaceutical Industry Payments to Health Professionals in Nine European Countries’ (2018, 1 June) 7(6) Int J Health Pol Mgmt, 504–509. doi:10.15171/ijhpm.2018.20. 37  Resembling the US Sunshine Law (Physician Payments Sunshine Act of 2010). 38  For more details, see MHE study (n5) 2019. 39  MHE Study (n5), p. 23. 40 Id.

550   Sunčana Roksandić to disclose information about payments to a regulatory agency but not to the public, or requirements for the pharmaceutical industry, which do not apply to other industries.41 In other EU countries, there is a little requirement for reporting and self-­disclosure, and COIs with the pharmaceutical sector are primarily managed through anti-­ corruption law and regulations regarding advertising and promotional activities ‘which might apply to healthcare.’42 Anti-­corruption legislation includes special offences, either specifically targeting physicians (e.g., Germany) or all public and private actors (e.g., Croatia). In addition, disclosure could be managed by prescribing criminal sanctions for breaching the duty of care when injury occurred to a primary patient (e.g., medical malpractice offences). France, a country having full legislation, adopted a new health law in December 2011 (as amended in 2017)43 with specific provisions to increase transparency between the pharmaceutical industry and healthcare providers following the public revelation of potential deaths after the use of the diabetes medication benfluorex (Mediator) for weight loss. Sevier was the producer of benfluorex, and the drug allegedly led to the death of between 500 and 2,000 people in France before it was taken off the market in 2009. Together with Sevier, the French drug regulator is accused of not acting to prevent death and injuries. The content of the French Act is similar to the US Open Payments Program. The Act requires the publication of information on any agreements with healthcare providers that provide benefits in cash or in kind that exceed €10.44 For a good example of a country invoking criminal law in combating an especially egregious COI, consider the following Croatian case involving financial incentives to prescribe drugs. In the X affair, 364 physicians and a pharmaceutical company were indicted for having committed several criminal offences—associating to commit criminal offences, taking and giving bribes, abuse of power and authority, and instigating the abuse of power and authority. The 1st accused person was the Management Board president of a pharmaceutical company, the 2nd accused person was the executive director of corporate affairs and director of controlling of the company sales, whereas the 3rd accused, 4th accused and 5th accused were regional directors of certain sectors. They were indicted for the criminal offence of bribing a large number of physicians and pharmacists between 2009 and October 2012 to achieve the highest possible 41  Id., pp. 26–29. 42  Id., p. 28. 43  Libération (2011, 1 August). Après le Mediator, la loi remède de Bertrand. http://www.liberation. fr/societe/2011/08/01/apres-­le-­mediator-­la-­loi-­remede-­de-­%20bertrand_752578 44  For further details of the French law, see the MHE study at https://mhe-­sme.org/wp-­content/ uploads/2019/01/MHE-­SHEDDING-­LIGHT-­REPORT-­Final.pdf. The penalty that can be imposed on the basis of the national regulation reaches the amount of €225,000 and may be accompanied by other sanctions, such as the suspension of business activities. Disclosed information can be accessed by the public through the website https://www.transparence.sante.gouv.fr (p. 25). For criticism, see, e.g., Quinn Grundy, Roojin Habibi, Adrienne Shnier, Christopher Mayes, and Wendy Lipworth, ‘Decoding Disclosure: Comparing Conflict of Interest Policy Among the United States, France and Australia’ (2018) Health Pol 122, 509–518.

Provider Conflicts of Interest in European Healthcare   551 sales rates for certain drugs. They established and managed a continuous connection of large number of physicians and pharmacists in the Republic of Croatia who received bribes from the pharmaceutical company. Bribes were provided in the form of value tickets [coupons or vouchers], money, paid travel arrangements on behalf of the pharmaceutical company, as well as other gifts in the value of five to ten per cent of the value of prescribed drugs. Final convicting judgments have been rendered for 313 persons (of which 297 are physicians and 16 are ‘other’ persons - veterinarians, pharmacists etc.). Most persons pled guilty to avoid lengthy procedures. [Professional disciplinary sanctions were also imposed.]45

The United Kingdom is another country that relies primarily on self-­regulation and existing rules without particular requirements for additional reporting and public disclosure. For the UK, it could be stated in general that one of the best examples of self-­ regulation is the provision and guidance on financial and commercial arrangements and conflicts of interest from the UK’s General Medical Council that came into effect on April 22, 2013. The guidance explains what is meant by the obligation to be honest in financial and commercial dealings with patients, employers, insurers, and other organizations and individuals. Provision 78 provides that physicians must not allow any personal interest to affect the way services to patients are prescribed for, treated, referred, or commissioned. Provisions 79 and 80 provide that, when physicians are faced with a COI, they must be open about the conflict, declaring any interest formally, and should be prepared to exclude themselves from decision-­making. Physicians must not ask for or accept—from patients, colleagues, or others—any inducement, gift, or hospitality that may affect or be seen to affect the way they prescribe for, treat, or refer patients or commission services for patients. Physicians must not offer these inducements. Provision 79 clearly recognizes that nonfinancial interests could also be a great risk for making a biased judgment: ‘Trust between you and your patients is essential to maintaining effective professional relationships, and your conduct must justify your patients’ trust in you and the public’s trust in the profession. Trust may be damaged if your interests affect, or are seen to affect, your professional judgement. Conflicts of interest may arise in a range of situations. They are not confined to financial interests, and may also include other personal interests.’46 As in countries such as Croatia, for breaching the guidance, the most serious sanction could be striking off physicians from the professional register. The UK guidelines rely on the judgment of physicians to act properly as opposed to stating what kinds of conduct are acceptable and what kinds are unacceptable. 45  EU, EC Updated Study on Corruption 2017 (n10), p. 49: https://ec.europa.eu/home-­affairs/sites/ homeaffairs/files/20170928_study_on_healthcare_corruption_en.pdf 46  General Medical Council, Financial and Commercial Arrangements and Conflicts of Interest (published 25 March 2013, comes into effect 22 April 2013): https://www.gmc-­uk.org/-­/media/ documents/financial-­and-­commercial-­arrangements-­and-­conflicts-­of-­interest_pdf-­58833167.pdf?la=en

552   Sunčana Roksandić The experience with gifts guidelines in the United States suggests that relying on physician judgment does not have much effect on behavior, but drawing clear lines between acceptable and unacceptable gifts can have a meaningful effect (see more in the Conclusion).

3.4  Council of Europe and Its Recommendations Concerning Conflict of Interest and the Pharmaceutical Sector and European-­Level Ethical Codes In addition to adopting Conventions, the Council of Europe, consisting of all European countries, also issues policy recommendations. Each country must then decide whether to enact laws implementing these recommendations. However, there is a general obligation to consider recommendations. Two Council of Europe Resolutions are particularly relevant concerning the relationship between the pharmaceutical industry and healthcare providers. These are Resolution 2071(2015) and Resolution 1749 (2010). Resolution 1749/2010 was passed after scandals connected to H1N1 vaccines. As explained by Flynn, the Rapporteur for ‘The Handling of the NIH1 Pandemic,’47 various European countries signed so-­ called sleeping contracts with large pharmaceutical groups which were supposed to take effect on the declaration of a pandemic by the World Health Organisation (WHO) (para 46). However, the declaration of this pandemic by the WHO was only made possible by changing the definition of a pandemic and by lowering the threshold for its declaration in May 2009 (para 24). For the Rapporteur, the possibility that representatives of the pharmaceutical industry may have directly influenced public decisions and recommendations made about the H1N1 influenza remains one of the central issues of the ongoing debate (para 30). Amongst the factors leading to the suspicion of undue influence were the early measures taken on contractual arrangements for vaccine delivery between member states and pharmaceutical companies, as well as the enormous profits that companies were able to make as a result of the pandemic. The main suspicion, however, arises with regard to the issue of whether members of WHO advisory bodies have professional links to pharmaceutical groups, bringing into question the neutrality of their advice.

In the Resolution, it was underlined (para 5) that ‘notwithstanding the willingness of WHO and the European health institutions concerned to enter into a dialogue and conduct a review of the handling of the pandemic, the Assembly seriously regrets that they have not been willing to share some essential information, in particular to publish the names and declarations of interest of the members of the Emergency Committee of WHO and relevant European advisory bodies directly involved in recommendations 47 https://assembly.coe.int/CommitteeDocs/2010/20100604_H1N1pandemic_e.pdf

Provider Conflicts of Interest in European Healthcare   553 concerning the handling of the pandemic.’48 Also, the Parliamentary Assembly urged Member States to develop systems of safeguards against undue influence by vested interests if they have not yet done so (para 8.3) and to consider establishing a public fund to support independent research, trials, and expert advice, possibly financed by an obligatory contribution by the pharmaceutical industry (para 8.5). The EU’s European Centre for Disease Control and Prevention also relies on internal advisory bodies. The names and declarations of interest of experts on these bodies have just recently been released.49 In any case, the events with H1N1 influenza caused a loss of confidence among Europeans, both members of the public and decision-­makers. Resolution 2071(2015), ‘Public health and the interests of the pharmaceutical industry: how to guarantee the primacy of public health interests?,’ emphasizes the importance of education: academic programs should incorporate into their curriculum for healthcare professionals specific, mandatory training to foster awareness of the influence of pharmaceutical promotion and how to respond (para 6.1.1.). In addition, recommendations were given to place an obligation on pharmaceutical companies to declare their linked interests with all health sector players, to make these declarations accessible to the public, and to establish an independent authority responsible for monitoring this matter (para 6.1.3). All of these policies are designed to ensure that health-­ related decisions, including decisions on criteria for defining illness and thresholds for treatment, are taken on the basis of individual and public health considerations and are not profit-­driven (para 6.1.5). Also in this Resolution, the Parliamentary Assembly called on Member States to impose dissuasive penalties for any illegal practices carried out by pharmaceutical companies, where appropriate by imposing fines of a given percentage. Organizations of European professionals, such as the European Medical Students’ Association (EMSA), also play an important role here. EMSA’s Policy on Conflicts of Interest in Medical Education Settings50 recognizes that if medical students lack awareness of COIs in their academic surroundings, they risk compromising their independent decision-­making ability.51 Therefore, EMSA in its Policy of April 2019 calls on EU Member States and European Institutions to raise public awareness of transparency of collaborations in healthcare; to follow the call of the Council of Europe’s Parliamentary Assembly Resolution 2071 (2015) to incorporate into the curriculum for healthcare 48  According to the WHO, transparency is ensured by declarations of interest in which external experts present all their professional and financial interests, including funding received from pharmaceutical companies, consultancies, or other forms of involvement in relevant commercial activities (para. 31). Biased recommendations are prevented by only allowing those experts who have no perceived or real COIs to make recommendations. Finally, the relative weight of those who declare a COI is also an element to be taken into account in the WHO’s view: if a COI appears regarding a person who could otherwise give valuable input, the person concerned is only allowed to participate in the general exchange of views and communication (para 32). 49  See https://isirv.org/site/images/conferences/EPI_Stockholm/ECDC_practical%20info_for%20 visitors538163.pdf 50  Conflicts of Interest in Medical Education Settings (adopted 26 April 2019). 51  Id., p. 3.

554   Sunčana Roksandić professionals specific, mandatory training to foster awareness of the influence of  pharmaceutical promotion and how to respond; and to support an assessment of COI policies at European medical faculties. Furthermore, it is necessary, according to ESMA’s Policy on COI, to invest in the vocational training of healthcare professionals, foster greater awareness of the influence of pharmaceutical promotion and how to respond to it, and invest in research on actions in relation to COI and the implementation of COI policies at European universities.52 This policy is yet be fully implemented. In addition, the provisions on COI in the European Charter of Medical Ethics (2011), adopted by the Conseil Européen Ordres Médecines,53 are important. According to Principle 11, the physician acting as a simple practitioner towards a patient or as an expert or member of an institution must ensure the greatest transparency in what might appear to be a COI and act in full moral and technical independence. Respecting the provisions of this Charter might become binding for national physicians if a national professional organization adopts them as its principles alongside its national Codes of Ethics and Deontology.

4  Conclusion: Why Optimal Regulation of Conflicts of Interest Remains So Challenging Slow on the pick-­up and reactive more than proactive, the European response to COIs in healthcare remains a work in progress. Relying on fundamental duties of physicians to their patients, regulators and other groups have invoked a range of strategies from education and self-­regulation to professional discipline, statutes targeted at financial incentives from pharmaceutical companies, and general tort and criminal liability. In all of these approaches, the emphasis is placed on building and preserving the personal integrity of physicians. The difficulty for regulators is drawing the right balance among the different strategies. While education and self-­regulation are important ways to cultivate a desirable professional ethic, draw on professional expertise, and encourage professional buy-­in, they are not sufficient. This is generally the case with professional regulation, and especially so when very high financial and nonfinancial secondary interests come into play. To truly protect patient welfare, education and self-­regulation must be supplemented by well-­conceived and effectively enforced law.

52  Id., p. 5. 53  European Charter of Medical Ethics (adopted in Kos, 10 June 2011): http://www.ceom-­ecmo.eu/ en/european-­charter-­medical-­ethics-­67

Provider Conflicts of Interest in European Healthcare   555 Unfortunately, the experience to date indicates that current legal approaches need reworking. For example, while regulations often emphasize transparency and therefore disclosure of physician–industry relationships, this strategy has had dubious success and has invited criticism by scholars.54 A 2019 nationally conducted policy evaluation (survey) of the effect of transparency of industry payments on patients was based on the US experience with its Open Payments policy. Researchers studied the effect of public disclosure of industry payments on patients’ awareness of the payments and knowledge about whether their physicians had accepted industry payments. The results were that public disclosure of industry payments information did not significantly increase the proportion of respondents who knew whether their physicians had received industry payments. It also did not change the proportion of respondents who became aware of the issue of industry payments, but it did increase the proportion who knew that payments information was publicly available. The authors of this study came to the following conclusion: ‘Two years after the public disclosure of industry payments information, Open Payments does not appear to have achieved its goal of increasing patient knowledge of whether their physicians have received money from pharmaceutical and medical device firms. Additional efforts will be required to improve the use and effectiveness of Open Payments for consumers.’55 As this study indicates, policymakers need to develop a more appropriate perspective on the general balance of regulatory strategies and in particular on the role of transparency. While undoubtedly regulators can refine the content of disclosures and the method for making them, too much focus on disclosure diverts attention from other approaches that would be much more effective. In particular, it is important to recognize that some kinds of conflicts are so problematic that disclosing them is not enough. Such conflicts should be prohibited. As the Croatian experience illustrates, pharmaceutical companies must not directly reward physicians for prescribing their drugs; it also is important that industry not employ indirect rewards that compromise patient welfare.

Acknowledgments The author wishes to express gratitude to the editors for providing guidelines on how to structure this chapter. The same goes to the co-­author of the joint introduction, Richard Saver, especially for providing valuable insight on the US approach to conflict of interest. A particular debt of gratitude goes to Sarah Clarke QC, who did the same for the UK approach. Also, a deep thank you is owed to Kristijan Grđen, a board member of Mental Health Europe (MHE), for updating me on the latest research on COI done by MHE. Last, but not least, thanks to my student assistant Doris Skaramuca who arranged the footnotes. 54  See, e.g., Grundy et al., supra note 45. Also, for the critique of the US Sunshine Policy, see Chapter 14 on COI in the United States, this volume. 55  GP Kanter, D Carpenter, L Lehmann, et al., ‘Effect of the Public Disclosure of Industry Payments Information on Patients: Results from a Population-­Based Natural Experiment’ (2019) BMJ Open: 9:e024020. doi:10.1136/bmjopen-­2018-­024020

556   Sunčana Roksandić

Bibliography Health Action International, WHO, The Project on Medicine Prices and Availability, Resolution WHA 54.11 (2011) p. 3. Christopher C Muth, MD, JAMA Patient Page, Conflict of Interest in Medicine, May 2, 2017 JAMA. 2017;317(17):1812. doi:10.1001/jama.2017.4044. S Roksandić Vidlička, A Maršavelski, and I Goldner Lang, Regulating breaches of professional ethics and conducting disciplinary proceedings against physicians in Europe//Compendium on European Health Law/den Exter, Andre (ur.)., Antwerpen: Maklu, 2017. 669–700. Thomas Sullivan, ‘EFPIA Disclosure ‘Sunshine Act – Lite,’ last updated 5 May, 2018, https:// www.policymed.com/2016/04/efpia-disclosure-sunshine-act-lite.html UN General Assembly, Report of the Special Rapporteur on the right of everyone to the enjoyment of the highest attainable standard of physical and mental health, A/72/137 (14 July 2017), p. 3, para. 3.

MEDICAL M A L PR AC T IC E

chapter 28

M edica l Li a bilit y Comparing Compensation Systems Karl Harald Søvig and Barry Furrow

1  Joint Introduction Comparing liability/compensation systems across countries is both challenging and fascinating: challenging because of substantial differences in tort doctrine, judicial systems, and administrative compensation mechanisms; fascinating because those differences also suggest useful starting points for reform across national systems. In this chapter two of us—a Norwegian law professor (and Dean), Karl Harald Sovig, and an American health law/torts professor in the United States, Barry Furrow—address this question of comparative system analysis of medical liability. We approach this problem of patient injury in a healthcare system from the dual (and dueling) perspectives of system responsibility and provider culpability.

1.1  The US System As will be shown in this chapter, the US system remains solidly anchored in the tort liability system, with its emphasis on negligence-­based culpability in medical malpractice cases—but with timid moves toward enterprise liability and statutory communication and resolution programs. This system is grounded in state common law, where legal doctrine advances one case at a time under common law rules in the 50 American states in the federal system. Medical malpractice tort law has always been defined at the state level. Under the US Constitution, the federal government has only those powers explicitly granted in it. General police powers—the right to protect the health, safety, and welfare of citizens— are not listed and belong exclusively to the states. Malpractice cases, except under very

560   Karl Harald Søvig and Barry Furrow limited circumstances, can be filed only in state court.1 And even when cases can be filed in federal court, those courts apply state malpractice law. The US Congress can change only federal law; it would be unconstitutional for Congress to try to alter state medical malpractice law. Medical liability litigation can serve several larger social functions. Supporters argue that such litigation • compensates some severely injured patients, thus reducing the need for state-­ financed compensation systems; • reinforces good medical practice by forcing providers to consider how to improve their practices to reduce their risks of error; • articulates new duties of care to which providers are subject over time; and • gives voice to a subset of mistreated patients and offers them an explanation for their harms. These goals are admirable but are too often ill-­served by the US malpractice system, which undercompensates small claims and is dependent on hospitals and doctors to inform injured patients that they have a claim in less severe injury cases. In addition, special interest lobbying has in most states displaced compensation through layers of tort reform barriers and disincentives. It has become harder for plaintiffs to recover by reducing the economic potential of malpractice judgments. And the biggest flaw, by contrast to state compensation systems, is profound access problems. Small claims are rarely brought successfully, and healthcare providers are often motivated to conceal adverse events when they are not obvious. Litigation thus has virtues as well as flaws as a primary compensation system. While public compensation programs might fill in the gaps, the United States lacks public programs of any kind except for state and federal workers compensation for industrial accidents, the federal Vaccine Injury Compensation Program for the side effects of vaccines, and two state programs (in Virginia and Florida) that offer compensation for brain-­ damaged newborns.

1.2  The Scandinavian Systems By contrast, the Scandinavian systems have publicly funded patient compensation funds. The Scandinavian systems, such as the Danish and Norwegian, have turned to state responsibility for such injuries through public funds that provide a source of compensation to injured patients without regard to proof of culpability. The tort system in these countries was judged to be too slow and costly while also limiting access for patients with lower value injuries. The goal of these “non-­fault:” 1  Filing in federal court is an option when a question of federal law is raised, a federal preemption claim is asserted, or the suit involves two American states.

Medical Liability   561 systems is to compensate patients for medical injuries. The prioritization of injury compensation objectives also shifted preferences away from negligence as the compensation stand­ard. The Nordic countries have supplanted negligence with the concept of “avoidability.” Their schemes seek to compensate patients who have experienced injuries that could have been avoided under optimal circumstances. The choice has been to give patients access to a compensation fund but leave medical discipline and patient safety incentives to other systems. Claiming is free, and patients often receive assistance from healthcare providers in making their claims. Handlers then consult with medical experts about the patient claims and make a recommendation. The claims are evaluated using “avoidability” criteria rather than negligence stand­ ards. The “alternative treatment” rule can be applied if there is medical data to support an alternative approach that would be superior for the patient. A “hindsight” test may be applied as well, asking whether the provider could have found more information to inform her clinical judgment of an alternative procedure. The positive benefits of a non-­fault system, compared with a culpability system, are apparent. Non-­fault gives the patient a rather easy way to seek compensation. A complaint can be filed without the assistance of a lawyer; decisions about the award are made by an independent body whose members have expertise in the medical specialty. These non-­fault systems require legislation, and this promotes a political discussion regarding which situations are to be covered. And these non-­fault systems can focus the spotlight away from the health personnel, making them more eager to report and learn from their mistakes instead of hiding them. On the other hand, a non-­fault system can drive healthcare costs as award costs rise. The figures from the Scandinavian countries do not indicate this, but these are small countries and it should be kept in mind that the average level of compensation is in general low compared to the United States. A non-­fault system does not abolish the need for litigation. There will be borderline cases also here, regarding the questions of what is covered, causation, and the level of compensation. No-­fault compensation schemes share certain features that any comparative analysis must consider. These include the administrative structure of the system, whether state-­ run and funded, or private and driven by private attorneys who trigger the judicial awarding process; the threshold criteria to test whether the patient should qualify for compensation; the scope of damages and limitations on them; and, finally, whether a patient can seek private litigation if dissatisfied with her award or denial of any award.

1.3 Conclusion The basic question we ask in this chapter is this: Who should bear the economic costs of medical adverse events? We all agree: we need both a high level of patient safety and a sustainable healthcare system. The US system strikes a balance between patient safety and system costs by only allowing compensation when the healthcare provider has acted

562   Karl Harald Søvig and Barry Furrow negligently; this assumes that such a proof requirement will promote safety through deterrence of provider negligence and thus avoid excess costs that might occur in a more broadly based compensation system. A public system’s goals are slightly different, but the expectation is still that the patient should have a “sound service.” It is also fundamental to have a sustainable healthcare system whose costs are manageable. All may agree that obvious wrongdoings should be compensated, while risks that are likely as a consequence of the disease (and that are informed about) should not be compensated. The issue is all the gray zones “in between,” and here there is a lot of room for discussion. The developing trends have been the same in both the United States and Europe. More and more diseases can be cured or alleviated, medicine has become increasingly complex, medical specialization has increased, and so has the use of medical devices and drugs in treatment. Rarely is only one doctor involved, and coordination of care is often a problem within healthcare services (within the hospital or between the primary and specialized levels). Sorting out cause and effect when patients suffer harm is not a simple matter. The legal issues in all compensation systems start with the same foundational questions. How should the patient’s claim of a right to compensation be evaluated—by negligence/culpability standards in judging provider failures or “avoidability” tests that are easier to ascertain? Is there a causal link between provider conduct and patient injury that is morally and legally compelling? And should compensation be capped at a certain level to control system costs, or allowed to include costs such as patient suffering over time out of concerns for fairness to the patient?

2  Case Studies of System Differences We begin with two rather different case studies of patient injuries in the healthcare system. The case studies allow a first look at how the US and Scandinavian systems are likely to approach the process of compensation and what the outcomes are likely to be. The first case study, Karl’s Bowels, looks at a delayed diagnosis that delays necessary surgery, with severe damages to the patient. The second case study, Anetta’s Tumors, involves a missed therapeutic opportunity, a lost chance to improve patient survival, because of a system failure in a machine learning tool. It looks at the emerging problems of artificial intelligence (AI) applications that are rapidly being applied in healthcare diagnosis and treatment.

2.1  Karl’s Bowels Karl was a 58-­year-­old man who worked as a clerk at the ticket window at a local railroad station.2 He was having leg pains, so went to see Dr. Kamm, a general surgeon at Advent Hospital. Kamm ordered an arteriogram to determine the location of arterial blockages 2  This case study is based on Snelson v. Kamm, 787 N.E. 2d 796 (Ill. 2003).

Medical Liability   563 in Karl’s legs caused by arteriosclerosis (“hardening of the arteries”). Dr. Capati, a radiologist practicing at Advent, performed a translumbar aortogram on Karl but had trouble navigating the guide wire into Karl’s thoracic aorta and the wire instead entered the superior mesenteric artery (which supplies blood to the intestine). Capati withdrew the needle and guide wire and tried to reinsert the guide wire into the aorta but Karl’s blood pressure dropped; he had abdominal and back pain and an urge to have a bowel movement. His stools showed no discoloration but Capati terminated the procedure, unable to complete the test. Karl began to experience back and abdominal pain. Symptoms, Round 1. Karl was brought back to his room in pain and was regularly monitored by the nurses on staff, given blood tests, and had his vital signs taken every 15 minutes for 2 hours and then hourly thereafter. Karl had continued pain and a bowel movement with blood-­tinged mucous. Dr. Kamm was called, and he decided the bloody stool was caused by a mild hemorrhoid or fissure. Symptoms, Round 2. Karl had another movement with blood-­tinged mucous a short time later, and the nurses again paged Dr. Kamm, who was in surgery and called back half an hour later. According to the nurses’ description, Kamm continued to think that Karl’s bloody stool was caused by a hemorrhoid. He arrived at the hospital an hour and a half later and examined the nurses’ notes and records. He now thought that Karl was bleeding into his retroperitoneal area from the puncture sites because his abdomen was swollen and distended, with diminished bowel sounds. Symptoms, Round 3. Kamm returned to the hospital the next day in the early morning. Karl’s white blood cell count was high, and, over the next 4 hours, a computed tomography (CT) scan and x-­rays were taken showing abnormalities in Karl’s intestine, including air in his small intestine. Dr. Capati, who interpreted the CT scan and x-­rays, testified that the results were consistent with “small and large bowel infarction,” which meant that parts of Karl’s small and large bowel loops were gangrenous or dead. Capati further testified that the most likely cause of that condition was “acute embolism and thrombosis involving the superior mesenteric artery,” meaning that a plaque or clot moving within the blood vessel, or a preexisting plaque or clot, had blocked the superior mesenteric artery. Capati thought that the unsuccessful translumbar aortogram caused the death of portions of Karl’s intestine. Adverse Outcome. Dr. Kamm performed emergency exploratory surgery on Karl later that morning and found that almost all of his small intestine and half of his large intestine were dead due to lack of blood circulation to the area. He removed 95% of Karl’s small intestine and the right half of his large intestine. During Karl’s unsuccessful arteriogram, the guide wire had injured the lining of the superior mesenteric artery, which caused a blood clot to form and gradually caused the death of Karl’s intestines from loss of circulation.

2.2  Anetta’s Tumors Anetta was experiencing rectal bleeding, fatigue, and weight loss. Her physician, Dr. Nilsen, ordered a colonoscopy and took tissue samples for analysis. He concluded

564   Karl Harald Søvig and Barry Furrow that Anetta had colorectal cancer of an advanced type.3 The treatment choices at this point in the diagnosis included surgery to remove the primary tumor, if obstructed or bleeding; chemotherapy to shrink a large tumor and after surgery to destroy remaining cancer cells; radiation therapy using x-­rays or protons to kill the cancer cells or shrink a large cancer to reduce symptoms; or targeted drug therapy to block abnormalities in cancer cells and kill the cells (targeted drug treatments are for advanced colon cancer). Finally, immunotherapy is a drug treatment that uses the patient’s immune system to fight cancer. It is usually reserved for advanced colon cancer, and the cancer cells are usually tested to see if they respond to this treatment. Dr. Nilsen ordered a magnetic resonance imaging (MRI) study to better predict the development of Anetta’s cancer and selection of treatment. The MRI images were analyzed by a machine learning method developed by the University of Oslo Medical School, a method that claimed higher accuracy of diagnosis than the ordinary oncologist. A clinically useful prognostic marker had been developed using deep learning allied to digital scanning of stained tumor tissue sections, and it was tested in large patient populations. This approach outperformed established molecular and morphological prognostic markers and gave consistent results across both tumor and nodal stages. It grouped patients into distinct prognostic groups to guide selection of patients who would benefit from more intensive treatment regimes.4 Based on the screening and the analysis by the AI method, Dr. Nilsen decided that Anetta should undergo chemotherapy but not radiation based on the machine learning predictions. The outcome of the machine learning was cross-­checked by Dr. Engen, a physician with specialization in oncology who came to the same conclusion as Dr. Nilsen regarding proposed treatment. Anetta received chemotherapy and was routinely followed up a year later by Dr. Nilsen. He then discovered that she had a rare variation which required a combined treatment approach of both radiation and chemotherapy; the machine learning method was supposed to identify this variation. The problem was that due to an error in the transfer of data from the machine learning project to the hospital, this variation was overlooked. The machine learning program had not correctly identified the variation, and by identifying it, it would have signaled to Dr. Nilsen that Anetta was in that class of patients who would have benefitted from the more intensive treatment regime. Even for a trained physician, it would be difficult to diagnose the rare variation. Anetta now has stage 4 colorectal cancer. Her treatment options are limited, and her life expectancy has been reduced from 10 years with a proper correct diagnosis and treatment to less than 1 year.

3  E. Van Cutsem et al., Advanced colorectal cancer: ESMO Clinical Practice Guidelines for Treatment, Ann Oncol. 21 (Supplement 5): v93–v97 (2010). 4  Ole-­Johan Skrede et al., Deep Learning for Prediction of Colorectal Cancer Outcome: A Discovery and Validation Study, Lancet 395: 350–360 (2020).

Medical Liability   565

3  Medical Liability in the United States 3.1  Introduction This chapter examines the processes of malpractice litigation, changes in tort doctrine and judicial responses to medical injuries, and tort reform, in order to give a snapshot of US malpractice law. The terrain is hotly contested, with disagreements over the role of expert testimony, the bases for assessing damages, the exposure of hospitals to liability, and the reach of informed consent doctrine. Many malpractice claims are uneconomical to bring, given (1) the evidentiary and other costs of preparing these complex cases and (2) state reforms that cap damage awards and otherwise reduce plaintiff incentives to sue. On the other hand, insurers’ settlement decisions may lead them to pay for smaller claims that would be unsuccessful at trial, to reduce their exposure to actuarial risk. The shadow of tort doctrine and the risk of jury trials alter insurer behavior, and tort reforms alter lawyer incentives. And the role of liability insurers is a large one, since most medical malpractice cases will settle before trial.

3.2  Physician Liability Medical liability law requires that the plaintiff satisfy four elements of proof to establish “negligence”—the failure of a physician to exercise due care. These elements are: (1) duty, (2) breach, (3) causation—including cause-­in-­fact and proximate cause, and (4) damages. Duty means that a relationship between patient and provider exists, creating an obligation by the provider to exercise due care in the treatment of the patient. Breach means that the provider failed to exercise the care expected of a reasonably skilled physician. Establishing breach requires expert testimony, since liability is proved by a deviation from the prevailing medical standard of practice. Cause means that the physician’s breach of the standard of care is the cause in fact of compensable injury. Proximate cause excludes patient injuries that are too remotely related to the defendant’s breach; one example is when a doctor fails to identify a lump in a patient’s breast, and, a year after this initial opportunity is missed to identify the lump, the cancer is discovered. The patient, due to family history and other factors, has a mastectomy. If the treatment for the lump, given the family history, would have been a mastectomy even if found a year earlier by doctor, the failure to diagnose in a timely fashion is arguably not the proximate cause of the injury (the mastectomy). Causation often presents a gnarly problem in malpractice cases. In the American system, all states place the burden of proof in liability cases on the plaintiff, who must meet this burden by a preponderance of the evidence.

566   Karl Harald Søvig and Barry Furrow Compensatory damages take two forms: economic (e.g., lost earnings and medical bills) and non-­economic (e.g., pain and suffering). Economic loss damages are usually provable through expert testimony. Non-­economic losses are harder to prove, since the concept of pain and suffering is an elastic one, but such damages are valuable: they capture a dimension of harm beyond the provable lost earnings and medical costs, and they provide a fund for the lawyer’s fees and other litigation costs (which ordinarily are paid for by patients only when they recover damages, as a percentage of their recoveries—the lawyer’s “contingency fee”).

3.2.1  Physician Standard of Care The liability of healthcare providers is governed by special negligence principles. Malpractice is usually defined as unskillful practice resulting in injury to the patient, a failure to exercise the “required degree of care, skill and diligence: under the ­circumstances.5 In an American medical malpractice trial, expert testimony is critical to determining the standard of care. Each side in a malpractice must present expert witnesses, with the burden of proof on the plaintiff to prove that the defendant’s conduct was negligent and the defendant’s witnesses arguing to the contrary. Changes in tort law have raised the bar for expert qualifications. For example, the testifying physician might have to be in active practice in the specialty of the defendant. In Karl’s case, the expert medical witnesses made the following statements. Witness 1, testifying for the plaintiff patient, stated that the surgeon Kamm should have immediately operated to restore circulation, which would have saved a large portion of Karl’s intestine. The following signs and symptoms should have alerted Kamm to the mesenteric ischemia: (1) the radiologist Capati’s indication that, during the unsuccessful arteriogram, the guide line went into the superior mesenteric artery; (2) Karl’s drop in blood pressure and abdominal pain during the procedure; (3) Karl’s need to have an immediate bowel movement during the procedure; (4) the bloody bowel movements following the procedure; (5) abdominal pain that was severe enough for Kamm to increase the pain medication (Demerol); and (6) the distention and tenderness of Karl’s lower abdomen during Kamm’s 6 pm examination. Witness 2, testifying for the physician, stated that generally acute mesenteric is­che­ mia is very difficult to diagnose because the typical patient has an onset of abdominal pain with no clear history of causation. Kamm met the standard of care since there weren’t enough findings or symptoms to justify surgery. Revascularization surgery has a mortality rate often in excess of 90% and as low as 50%. Witness 1 noted that Karl was unlike the typical patient who is admitted to the hospital several hours after the onset of pain: (1) he was in the hospital at the time the ischemia began; (2) the problems that developed during the unsuccessful arteriogram involved the superior mesenteric artery; (3) he developed signs and symptoms quickly and did not just arrive at the hospital with “some obscure things going on”; (4) Karl was not the typical patient, who is elderly and has trouble communicating; and (6) while the length 5  Bardessono v. Michels, 91 Cal.Rptr. 760, 764, 478 P.2d 480, 484 (Cal.1970).

Medical Liability   567 of time that intestines will remain viable once blood supply is lost is variable from patient to patient and cannot be determined with 100% certainty, based on the clinical data contained in Karl’s medical records, the latest time his intestine could have been saved was around midnight. Pain medication should not have been given to Karl since such medication masks medical findings so that a doctor can’t know if the patient is getting better or worse. This case presents an example of a “missed diagnosis,” delayed corrective action, and severe damages that might shorten the plaintiff ’s remaining life. Accordingly, the jury returned a verdict in favor of Karl. Expert physicians not only testify to the standard of care. The medical profession itself has traditionally set its standards of practice, and the courts have enforced these standards in tort suits. Professional medical standards are typically given conclusive weight by most state courts6; however, several states have begun to reject customary practice when they find the practice dangerous or out of date.7 Medical practice guidelines may at times be a solution to the endlessly complicated proof problems of customary medical practice since such guidelines can establish the breach of a medical standard of care. The Institute of Medicine (IOM) defines clinical guidelines as “statements that include recommendations intended to optimize patient care that is informed by a systematic review of evidence and an assessment of the benefit and harms of alternative care options.”8 Guidelines are standardized specifications for using a procedure or managing a particular clinical problem. The development of clinical practice guidelines can be a transforming force in current US medical practice. Such guidelines provide decision rules based on current medical consensus on how to treat a certain illness or condition. In one case, for example, a physician defended his assessment of a patient’s chest pain by relying on a consensus statement on the interpretation of exercise treadmill tests that had been developed by the American College of Cardiology and the American Heart Association.9 The best guidelines have affected medical practice and settlement patterns, according to surveys of malpractice lawyers. Practice guidelines represent the blossoming of evidence-­based medical practice in all modern countries, not just the United States. Such practice guidelines are also essential to proof in a European expert’s assessment of a patient’s claim for malpractice. As scientific research proves which treatments work and which older treatments do not, the standard of medical practice will change. Courts can then allow the introduction of good evidence-­based guidelines as a source of the medical standard of care. 6  See, e.g., Holt v. Godsil, 447 So.2d 191 (Ala.1984). 7  See, e.g., Chiero v. Chicago Osteopathic Hospital, 392 N.E.2d 203, 209 (Ill. App. Ct. 1979). 8  Institute of Medicine. Graham R, Mancher M, Wolman DM, Greenfield S, Steinberg E, editor(s). Clinical Practice Guidelines We Can Trust 2 (2011), http://www.iom.edu/Reports/2011/ Clinical-­Practice-­Guidelines-­We-­Can-­Trust/Standards.aspx 9  Frakes v. Cardiology Consultants, PC, 1997 Tenn App 597 (Tenn Ct App 1997). See also Titchnell v. US, 681 F.2d 165 (1982)(considering public health guidelines for administration of swine flu immunizations).

568   Karl Harald Søvig and Barry Furrow

3.2.2  Causation Distinguishing a physician-­caused bad outcome from patient deterioration is a ­complicated proof problem in many tort cases, with experts often battling over the issue. The plaintiff must present evidence from an expert who will testify “to a reasonable degree of medical certainty, that the acts of the physician deviated from good and acceptable medical standards, and that such deviation was the proximate cause of the harm suffered.”10 The physician’s negligent act must be a substantial factor in causing the harm and without which the harm would not have occurred.11 Missed diagnoses are the source of many medical adverse events. The physician’s differential diagnosis may be complicated by ambiguous patient presentations and overlapping medical problems. Or sometimes providers just rush patients along, missing critical signs.12 In Karl’s case, Witness 1 testified that a brief window of opportunity existed to prevent the permanent loss of Karl’s intestine. At 6 pm, following his early afternoon arteriogram, Karl was stable enough to have surgery, and, if revascularization surgery had been performed in a timely fashion that evening, a large portion of Karl’s intestine could have been salvaged and he would not now be dependent on intravenous supplemental nutrition. The latest point in time that Karl’s intestines could have been saved was midnight, and “after that it was too late.” 3.2.2.1  Damages US common law requires the plaintiff to show by a preponderance of the evidence that because of the defendant’s negligence, the plaintiff suffered a specific injury, as in the case of Karl’s bowel infarction. The normal elements of damages, including lost wages, medical expenses past and future, impaired earning capacity, and pain and suffering, are allowed where the negligence of the provider is proved by a preponderance of the ­evidence. This test means proof by more than 50% in the mind of the trier of fact, usually the jury. Karl had substantial damages from his bowel removal. He has short bowel syndrome, a condition that creates diarrhea and a lack of ability to absorb sufficient nutrition and calories so that he has to rely on hyperalimentation, the intravenous infusion of a solution containing sufficient nutrients to sustain him, through a catheter into a vein in his upper chest, attached to a small machine that injects the nutrition. He has had repeated infections from his catheter requiring hospitalization, almost 1½ years since June 1994. He has had his catheter changed 20 times already, which can be painful, as are the intravenous antibiotics he needs to fight these infections. He will suffer from diarrhea and require hyperalimentation for the rest of his life. Karl will be unable to work regularly because of repeated infections and hospitalization, the effects of repeated infections.

10  Checchio By and Through Checchio v. Frankford Hospital-­Torresdale Div., 717 A. 2d 1058, 1060 (Sup. Ct. Pa. 1998). 11  Arredondo v. Rodriguez, 198 S.W. 3d 236 (C.A. Texas 2006). 12  Richmond v. Hunt,—S.W.3d—2019 WL 1496951 (2019).

Medical Liability   569 Karl will have continued pain and suffering from the catheter problems and the hyperalimentation for the rest of his life. The final verdict in favor of Karl, affirmed on appeal to the state supreme court, was $7 million. The court noted that the patient’s loss of his small intestine “has radically affected his life and subjected him to phenomenal suffering.”13 Here we see how US damage rules work to allow pain and suffering damages and generous recovery when the jury finds that the evidence justifies it.

3.2.3  Damage Innovations: Loss of a Chance Doctrine Liability also can result when a patient suffers harm from a condition that the physician didn’t cause if a missed diagnosis worsens the patient’s prognosis. Failures to diagnose medical conditions are the most common medical mistakes that physicians make. When that happens, the patient loses the opportunity for earlier, more effective treatment. This “loss of a chance” has now been adopted by about half of American jurisdictions, and the patient’s recovery is based on the proportion by which the physician’s error worsened the prognosis. For example, if the patient’s chance of survival from cancer dropped from 70% to 40% because of physician error, then the physician would be liable for 30% of the patient’s damages.14 Anetta’s case illustrates the loss of a chance doctrine. Because of the incorrect initial diagnosis, Anetta’s tumor variation was missed, and the delayed correct diagnosis meant that Anetta lost a chance for effective treatment that might have extended her life or cured her.

3.2.4  Informed Consent The risks of a medical treatment, and alternatives to it, are perhaps the most important pieces of information for a patient to consider. Patients might not consent to a drug or surgery if they understand the level of risks, risks that are material to a patient’s decision-­ making. Material risks must therefore be disclosed in order to give the patient a choice in her treatment.15 Courts limit the physician burden of disclosure by holding that remote risks can be omitted,16 as well as those commonly known by patients.17 Risks of drug side effects, however, are singled out for disclosure by some courts even if the risk of side effect is small.18 The actualities of tort-­driven informed consent are not impressive: patients don’t understand information or forget it, providers manipulate the risks to get the patients to undergo therapy the physicians think necessary, and the process is often little more than 13  Arredondo v. Rodriguez, 198 S.W. 3d 236 (Tex. App. 2006). 14  Richmond v. Hunt, 596 S.W.3d 103 (Ky. App. 2019). 15  See Harbeson v. Parke Davis, 746 F.2d 517 (9th Cir.1984). 16  See, e.g., Koapke v. Herfendal, 660 N.W.2d 206, 212 (N.D. 2003). 17  See generally Kissinger v. Lofgren, 836 F.2d 678, 681 (1st Cir.1988) (risks of infection from operations should be known to every patient). 18  Cunningham v. Charles Pfizer & Co., 532 P.2d 1377, 1381 (Okl.1974) (duty to disclose “even though the chances of the adverse reaction occurring are statistically small”).

570   Karl Harald Søvig and Barry Furrow a formal and mechanical exercise in futility. Few lawsuits based on informed consent theories are successful.

3.3  Hospital Liability 3.3.1  Introduction The US healthcare delivery system has a growing range of institutional forms of delivery. Hospitals are the classic provider institution. The United States has more than 6,000 hospitals—almost 3,000 are nonprofit, 1,000 for-­profit, and 1,200 are local, state, and federal government owned, with the rest psychiatric and long-­term care hospitals. Many of these hospitals are in one of more than 600 healthcare systems, which provide 88% of the US hospital beds. Hospitals are the major providers of emergency care and highly complicated surgical and other procedures in the United States. They are therefore the largest sources of patient harms in the US system. Hospitals provide acute care in severe health crises, and, given the possibility of errors and serious adverse events, we also think of institutional liability for those injuries.

3.3.2  Agency Doctrine: Stretching to Reach Hospitals Hospitals are responsible for the torts of their employees, including employee-­doctors, under the principle of vicarious liability. The hospital is legally an enterprise liable for the acts of its employees,19 so the question arises whether physicians count as a hospital’s employees or instead as independent contractors. The test of employee status is based on the extent to which the employer exercises “control” over the worker’s activities.20 Physicians need considerable autonomy in practice given the complexity of their decisions and their relationship to particular patients. Determining the degree of control necessary to create an employment relationship in a medical malpractice claim poses a unique set of difficulties.21 The liability complication in most US hospitals is that most physicians are still legally described as independent contractors, part of a hospital’s independent medical staff and bound by its bylaws, not under the “control” of the hospital administration and therefore not an employee for legal purposes. As independent contractors, the hospital is free of any agency-­based liability for their negligent acts. Many state courts have allowed cases to proceed on agency-­based theories of ostensible agency or apparent authority.22 19  41 C.J.S. Hospitals § 43—Theories of Liability. 20  The general definition of the term “servant” in the Restatement (Second) of Agency § 2(2) (1957) refers to a person whose work is “controlled or is subject to the right to control by the master.” The relevant factor for analyzing the hospital–physician relationship by agency tests is § 220(2)(a), which looks to “the extent of control which, by the agreement, the master may exercise over the details of the work.” This becomes a fact-­intensive analysis for the trier of fact. 21  Lilly v. Fieldstone, 876 F.2d 857 (C.A. 10 Kan. 1989). 22  Sword v. NKC Hospitals, Inc., 714 N.E.2d 142 (Ind.1999) (and cases cited).

Medical Liability   571 Courts are impatient with the law’s role in shielding modern hospitals from liability, noting that modern marketing promotional campaigns deliberately foster patient reliance on the hospital’s doctors.23

3.3.3  Corporate Negligence: Enterprise Liability One of the biggest developments in hospital responsibility for patient injuries is the doctrine of hospital corporate negligence. A healthcare institution has always been liable to its patients for negligence in maintaining its facilities or in providing and maintaining medical equipment.24 Individual hospitals are generally held to a national standard of care for hospitals of their size and treatment category. In addition, courts have expanded the doctrine of corporate negligence since the 1960s, recognizing the role of the modern hospital in providing care, support, and serv­ ice, not just a shell for the physicians to use.25 The corporate negligence doctrine has been adopted in 24 US states.26 The doctrine is summarized by the Pennsylvania Supreme Court in Thompson v. Nason Hospital.27 Corporate negligence, as the Nason court outlines it, involves four distinct duties for institutional providers: (1) a duty to use reasonable care in the maintenance of safe and adequate facilities and equipment, (2) a duty to select and retain only competent physicians, (3) a duty to oversee all persons who practice medicine within its walls as to patient care, and (4) a duty to formulate, adopt, and enforce adequate rules and policies to ensure quality care for patients. A hospital is thus directly liable for the failure of administrators and staff to properly monitor and supervise the delivery of healthcare within the hospital. A hospital has a non-­delegable duty directly to its patients and is liable for its action or inaction regarding its policies, rather than specific negligent acts of one of its employees.28 Thus, for example, if a hospital fails to monitor a staff physician’s surgical mortality rate or infection rate, it could be held liable for a physician’s malpractice.29 Courts have held that accreditation standards of the Joint Commission can establish the standard of care for a hospital, along with state hospital licensure laws and the hospital’s own by-­laws.30

23  Clark v. Southview Hospital & Family Health Center, 628 N.E.2d 46 (Ohio 1994) (promotional and marketing campaign stressed the emergency departments); Gragg v. Calandra, 696 N.E.2d 1282 (Ill.App.1998) (unless patient is put on notice of the independent status of the professionals in a hospital, he or see will reasonably assume they are employees). 24  Lamb v. Candler General Hospital, Inc., 413 S.E.2d 720 (Ga.1992) (hospital negligent in failing to use proper replacement parts in a medical instrument). 25  Alex Stein, Toward a Theory of Medical Malpractice, 97 Iowa L. Rev. 1201, 1229 (2012). 26  See Larson v. Wasemiller, 738 N.W.2d 300 (Minnesota, 2007) (adopting corporate negligence for Minnesota, the court noted that more than half of the state courts have adopted the tort, and it has support in Restatement (Second) Tort sections such as sections 320 and 411). 27  591 A.2d 703 (Pa.1991). 28  Moser v. Heistand, 681 A.2d 1322 (Pa.1996). 29  Strubhart v. Perry Mem’l Hosp. Tr. Auth., 903 P.2d 263, 273, 276 (Okla. 1995). 30  Decker v. St. Mary’s Hospital, 619 N.E.2d 537 (Ill.App.1993); Pedroza v. Bryant, 766 P.2d 166, 171 (1984). But see Kadlec Med. Ctr. v. Lakeview Anesthesia Assocs., 2006 WL 1328872 (E.D. La.) (holding that the Joint Commission guidelines do not create a cause of action).

572   Karl Harald Søvig and Barry Furrow These duties capture the complexity of the modern healthcare institution and are a first step toward enterprise liability. Proof of negligence is still required: any breach of one of the four duties just described must be proved by evidence of deviation from a standard of care, sometimes established by a hospital accrediting agency such as the Joint Commission. Anetta’s case provides a useful example of institutional liability, with the focus of liability moving from the physician, Dr. Nilsen, to the process by which data were transferred to the hospital. An error in the data communication and transmission process caused Anetta’s variation to be missed, delaying effective treatment. In this kind of data storage and transmission reliability problem, evidence is required as to the hospital’s errors in implementing the transfer process. It is a complex system negligence problem beyond the scope of most medical malpractice cases. The cause of misclassification was due to data transfer and not the machine learning systems as such. And yet the error clearly caused the missed diagnosis and the lost opportunity for proper treatment of Anetta’s cancer—it is a clear error that seems to fit a no-­fault system model perfectly. This case study presents a problem of legal uncertainty. Should it be viewed as a corporate negligence problem or a hospital system failure? Was the error a one-­time glitch, or a system failure to test the IT system of transmission for possible errors? Or is it also a products liability problem based on the error in the transmission process, based on promises by an IT vendor providing the machine learning tool? A different problem is raised if the machine learning approach was negligently developed, for example based on a flawed research foundation. Another variant would be a situation where an expert oncologist could have recognized the variant from the existing clinical data available to her but had become overly reliant on the new machine learning tools to dictate the treatment approach. The use of AI tools in healthcare creates risks that justify a no-­fault system that simplifies the search for avoidability, or perhaps the search for the causes of AI errors will task even the no-­fault systems built on avoidability tests. These new technologies applied to healthcare may mean that a pure no-­ fault system is needed for hard-­to-­establish patient adverse events.

3.3.4  Enterprise Compensation Possibilities: Emergence of Data Analytics to Detect Adverse Event Hotspots Big Data is coming to healthcare. Modern electronic health records allow huge quantities of data to be generated and stored in data warehouses for use. The benefits of data analytics are potentially groundbreaking for the healthcare industry. Big Data may identify ways for healthcare institutions to reduce adverse events and medical errors within their facilities. With that promise comes an enterprise responsibility that sets a new standard of practice. The federal Medicare program has linked hospital deficiencies to reimbursement penalties. Data analytic tools are used to search patient records to spot causes of patient harm. Such data mining has proved to be powerful in uncovering hospital acquired

Medical Liability   573 conditions (HACs). Data mining has been a key tool in reducing the causes of HACs since 2008. Data mining moves from tracking a particular patient, or physician, to an overview of the whole population of a hospital over time. The power of these new tools leads to detection of a much wider range of adverse events in healthcare systems. One example is the use by a hospital of IBM’s data mining program, which revealed that pneumonia patients who were not given medication immediately upon admittance suffered significantly worse outcomes than those who were given medication. At another facility, data mining showed that patients with cardiovascular disease were not always prescribed beta-­blockers because the discharge process did not include a crucial step to ensure the prescription was ordered. An easy solution—to change work processes—was implemented. This ease of detection points to the need for a Medicare-­based compensation system to better serve the elderly Medicare population subjected to adverse events, a form of enterprise responsibility for adverse events beyond the narrow culpability tests of ordinary tort cases.

4  Liability Reform in the United States: Modest Evolution Toward a Patient Compensation System 4.1 Introduction Hospital adverse events continue to occur at high levels in US healthcare institutions. The Office of the Inspector General found that 13.5% of Medicare hospital admissions suffered an adverse event, with an equal percentage experiencing temporary harm.31 Another study concluded that patients suffered adverse events in one-­third of all ­admissions.32 A study by John James concluded that as many as 440,000 patients suffer from preventable adverse events that lead to their deaths each year.33 Complexity in modern medicine produces more adverse events and errors than ever before. Errors in drug prescribing, physicians who practice medicine contrary to best practice, and adverse events resulting from complex system interactions continue to be major sources of patient harm. Despite substantial evidence of patient injury, progress in reducing the volume of patient harms remains slow. Moreover, patients who suffer adverse events, 31  See US Dep’t of Health & Human Servs., Office of Inspector Gen., Adverse Events in Hospitals: National Incidence Among Medicare Beneficiaries, at i–ii (2010). 32  David C. Classen et al., “Global Trigger Tool” Shows that Adverse Events in Hospitals May Be Ten Times Greater than Previously Measured, 30 Health Aff. 581, 581 (2011). 33  John T. James 9 J. Patient Saf. 222 (2013).

574   Karl Harald Søvig and Barry Furrow even severe ones, often do not realize what has happened, are rarely told about the adverse event, and seldom file a claim for compensation. The American tort system does a poor job of capturing most cases of actual malpractice while it also fails to weed out cases involving bad outcomes not the result of medical error. For every seven to eight incidents of verified medical negligence, only one lawsuit for medical negligence is filed. And for every bona fide claim of medical negligence filed, five malpractice claims were filed that did not involve medical negligence. When claims that did not involve negligence are eliminated, only 1 claim of medical negligence is filed for every 50 verified incidents of medical negligence.

4.2  Medical Malpractice System Reforms 4.2.1  Improving Insurance Availability Several strategies have been adopted over the decades to reform the tort system. One of the first approaches was to improve insurance availability since healthcare providers must carry malpractice insurance and the premiums are often high. Tort reform efforts starting in the 1970s led to changes in the insurance industry, such as joint underwriting associations, reinsurance exchanges, hospital self-­insurance programs, state funds, and provider-­owned insurance companies. Hospitals began to self-­insure.

4.2.2  Altering the Litigation Process A second strategy was to alter the litigation process, making it harder for patients to sue. The goal was to reduce either the frequency of malpractice litigation or the size of the settlement or judgment. The most powerful reform in actually reducing the size of malpractice awards has been a dollar limit, or cap, on awards. Caps may take the form of a limit on the amount of recovery of general damages, typically pain and suffering, or a maximum recoverable per case, including all damages.

4.2.3  Creating Alternatives to Tort Litigation A third strategy was to provide alternatives to tort litigation. The most common such choice was mediation, for example offered as elective mediation when a patient enters a hospital, permitting an alternative track to resolve patient injury if an adverse event occurs. Such elective mediation includes the patient’s right to sue for malpractice if she is unhappy with the mediation result. These mediation models have led to more elaborate schemes called communication and resolution programs (CRPs), discussed later. Alternative dispute resolution (ADR) was much discussed at the peaks of several malpractice crises (in 1973, 1987, 2000). No state to date has required arbitration of medical malpractice litigation, although some states encourage patient–provider arbitration agreements. Many state constitutions have a provision allowing a plaintiff ’s right to a jury trial, which means that arbitration can only be a contract solution by agreement.

Medical Liability   575

4.2.4  Communication and Resolution Programs The idea is that hospitals should voluntarily agree to identify, notify, and promptly compensate patients for avoidable injuries. These early proposals from the 1970s have resurfaced in modified forms and are being tried by some hospital systems as a means of controlling damage exposure. The US Veteran’s Administration has implemented such a program. A hospital using this approach typically adopts several steps. First, a representative discloses adverse events to affected patients and their families. Second, she apologizes on behalf of the institution for causing the adverse event and harms suffered by the patient. Third, she offers compensation where appropriate. The approach combines “early offer” ideas with the strategy of full disclosure and apology. The idea is to induce settlements more quickly by co-­opting plaintiff lawyers with the promise of much quicker, less adversarial settlements that bring the injured plaintiffs into the conversation earlier. If a compensation system rewards many more claimants, particularly small ones, in an evenhanded and more rapid fashion than does the current tort system, it may well be an improvement. One American state, Oregon, has developed a more structured CRP form, an elective mediation system imbedded in a delicate communication and resolution incentive system that puts the patient in the center but without a lawyer at her side. Such an approach is aimed at coaxing hospitals to join in honest disclosure of patient adverse events, with the promise of mediation as a cost-­effective way to resolve a claim more swiftly and economically than the tort system would. It also joins physicians in the conversation while limiting the need for them to report any settlement to the National Practitioner Data Bank. The Healthcare Quality Improvement Act of 1986 (HCQIA) required institutional providers to collect information on physicians in conjunction with granting physician staff privileges. The HCQIA established the National Practitioner Data Bank (Data Bank) to collect information about physicians all over the country and to monitor the credentialing of physicians by hospitals and states. The CRP approach, including the new Oregon program, is a far cry from a public patient compensation system such as the Scandinavian counties use, but such CRP programs may offer compensation for smaller claims. Such programs are constrained by the hospital’s assessment of what adverse events to disclose, and hospital incentives to remain silent on smaller adverse events are strong. The Oregon model perhaps is more likely to be transparent about the level of claims and compensation, as a legislatively created system, and has some promise depending on how the state manages it.

4.2.5  Apology Programs Apology laws have been adopted in 31 states and the District of Columbia based on case studies that suggest that apologies from physicians to patients will promote healing, understanding, and dispute resolution. These laws are thought to reduce litigation and medical malpractice liability by making patients grateful for the apologies and less likely to pursue malpractice claims and more likely to settle claims that are filed. Apology

576   Karl Harald Søvig and Barry Furrow strategies offer a strategic tool to buy off plaintiffs by showing them how sorry the provider is and to rush settlement by getting plaintiffs (and their lawyers) to buy into early settlement by the offer of money up front. A recent study concluded that once a patient has been made aware that the physician has committed a medical error, the patient’s incentive to pursue a claim may increase even though the apology itself cannot be introduced as evidence. Apology laws appear to have uncertain effects, with little evidence that they will lead to reducing litigation.

5   Pure No- ­Fault Compensation Systems 5.1  Birth Injury Cases No-­fault liability for medical error has proved far more popular with academics than with legislators in the United States. A few pockets of strict liability for medical malpractice exist in the United States. Two states—Virginia and Florida—have established administrative systems to handle birth injuries. Plaintiffs with strong cases prefer to litigate in the tort system since they can recover a greater amount, while plaintiffs with weak cases prefer the no-­fault system. Workers compensation is also an older no-­fault system that will compensate for worker injuries arising out of employment.

5.2  Vaccine Injury Cases Vaccine injuries provide the purest example of a patient compensation system in the United States. Such cases are handled by the US Court of Federal Claims. The 1986 National Vaccine Injury Act created a no-­fault system for handling such claims. Claimants must show that they experienced one of several listed adverse effects shortly after vaccination. The burden of proof is the same as in civil litigation: more likely than not. Compensation covers medical and legal expenses, loss of future earning capacity, and up to $250,000 for pain and suffering. There is also a death benefit of up to $250,000. Legal expenses may be compensated as well. The program is funded by an excise tax on every purchased dose of covered vaccine.

5.3  Retention of Adversarial Features The US examples all tend to move away from the model of an easily accessible program that compensates quickly for all levels of injuries. One study has found that programs such as the brain-­damaged infant state programs, along with workers compensation

Medical Liability   577 and no-­fault automobile insurance, have seen plaintiffs fleeing the program in order to gain more generous compensation. These programs have become quite adversarial, reducing their ability to compensate plaintiffs quickly and their no-­fault value.

6  Medical Liability in Europe 6.1 Introduction This part offers an overview of the systems of medical liability within Europe.34 It is impossible within the frame of this chapter to provide an in-­depth analysis of the legal questions concerning medical liability within the 47 states of Europe. The contribution will start with a presentation of the two main common legal frameworks of Europe: the European Union (EU; currently 27 member states) and the Council of Europe (CoE; currently 47 member states, covering all European countries). Subsequently, the contribution will provide examples from national jurisdictions, with an emphasis on the legal framework in the Scandinavian countries that have developed a Patient Compensation system that is of interest, especially in comparison with the US system, since the Scandinavian model is built on “non-­fault system.” The main three legal issues are basically the same as in the United States: 1. Foundation of the claim (culpability/non-­fault system) 2. Causation 3. The compensation (individual loss, non-­pecuniary damage, etc.) Even though the legal issues are basically the same, the legal solutions are indeed different, and this contribution will try to pinpoint the differences, with the two examples presented in the common introduction as a red thread.

6.2  The European Union and European Economic Area The EU currently covers 27 member states. In addition, the European Free Trade Association (EFTA) states, Iceland, Liechtenstein and Norway, are closely connected to the internal market by the agreement on the European Economic Area (EEA).35 The 34  The author wants to thanks research assistants Kristine Skorpen and Marit Tjelmeland for collecting material, the Norwegian National Office for Health Service Appeals for providing input to the estimation of damages for the case example, and to all the Scandinavian first-­instance bodies for providing figures to the statistics. 35  For an in-­depth analysis of this agreement, see Finn Arnesen and others (eds.), Agreement on the European Economic Area: A Commentary (Nomos/Hart 2018) and Carl Baudenbacher (ed.), The Handbook of EEA Law (Springer 2015).

578   Karl Harald Søvig and Barry Furrow relation between the United Kingdom and European Union is regulated by the EU-­UK Trade and Cooperation Agreement. The primary legislative framework of the EU clearly vests responsibility for healthcare with the domestic legislator.36 Union action “shall respect the responsibilities of the Member States for the definition of their health policy and for the organisation and delivery of health services and medical care.”37 The responsibilities of the Member States “shall include the management of health services and medical care and the allocation of the resources assigned to them.” The main reasoning for this approach, as expressed in the primary legislation, is the resources allocated to the healthcare system. Expenses related to the healthcare sector are funded by the Member States, which also entitles them to legislate in these matters. The average expenditure on healthcare services in EU 27 is 10% of gross national product.38 This approach is followed up in the secondary legislation. Directive 2011/24/EU of the European Parliament on the application of patients’ rights in cross-­border healthcare is of particular relevance.39 The directive is partly a codification of case law from the European Court of Justice and aims to guarantee patient mobility within the union. According to Art. 4(2)(c), the Member State of treatment shall ensure that there are “transparent complaints procedures and mechanisms in place for patients, in order for them to seek remedies in accordance with the legislation of the Member State of treatment if they suffer harm arising from the healthcare they receive.” Furthermore, the Member State of treatment is responsible for ensuring that “systems of professional liability insurance, or a guarantee or similar arrangement” are in place for treatment provided on its territory, cf. Art. 4(2)(d).40 The duty stipulated by the directive is to have a system that already will be in place in all Member States. The provision does not specify the content of the liability system, whether concerning the foundation for responsibility (culpability or non-­fault) nor concerning the quantification of damages. It should also be added that the directive is restricted to address civil liability and does not require a system of criminal liability. In sum, the requirements in the directive concerning redress must be said to be rather modest, and it does not intend to level the differences between national legislation in this field. It should be added that the approach of the cross-­border health directive is in line with similar secondary legislation within health law. The overall approach is that if

36  Cf. article 35 CFR (Charter of Fundamental Rights of the European Union) and Article 168 TFEU (Treaty of Functioning of the European Union). 37  Article 168(7) TFEU. 38  https://ec.europa.eu/eurostat/en/web/products-eurostat-news/-/ddn-20190904-1 (last retrieved 1 August 2020). The table also shows statistics and figures from UK and the EFTA states. 39  For an overview of the directive, see inter alia, André den Exter (ed.), Cross-­border health care and European Union Law (Erasmus University Press 2017). 40  Cf. Art. 5(3) regarding the responsibilities of the Member State of affiliation and the obligation for the national contact points to provide information about the complaints procedures and mechanisms for seeking remedies according to the legislation of the Member State of treatment.

Medical Liability   579 liability is regulated, the competence is vested with the Member States,41 with the appendix that the national legislation shall be effective, proportionate, and dissuasive.42 The tension in the cross-­border health directive, as in other secondary legislation concerning health, is between those wanting to maintain national control and those wanting to strengthen patient mobility and free choice.43 Healthcare is to some extent a service. A pillar within the EU is the establishment of the European single or internal market which seeks to guarantee the free movement of goods, capital, labour, and serv­ ices—the “four freedoms”—within the Member States. Most European countries have a combination of private and public healthcare providers, although it varies significantly which role each of them plays in the different countries and how they interrelate.44 The organization of healthcare is also closely related to financing. In many European countries the financing is to a large extent public, either directly or based on some sort of (compulsory) insurance.45 Hence, healthcare is so closely connected to the national resources that a harmonization of systems for medical liability seems unlikely in the foreseeable future.

6.3  Council of Europe 6.3.1  Biomedicine Convention Turning to the CoE, all 47 European countries are members. The Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine (Biomedical or Oviedo Convention, signed 4 April 1997 in Oviedo) addresses compensation for undue damage in Article 24. According to this provision, a person who has suffered “undue damage” resulting from an intervention 41  Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices article 10 no. 16; Regulation (EC) 726/2004 of the European Parliament and of the Council of 31 March 2004 laying down Community procedures for the authorisation and supervision of medicinal products for human and veterinary use and establishing a European Medicines Agency, article 15. 42  Directive 2002/98/EC of the European Parliament and of the Council of 27 January 2003 setting standards of quality and safety for the collection, testing, processing, storage, and distribution of human blood and blood components and amending Directive 2001/83/EC, article 27 (concerning penalties), cf. also art. 11 concerning a notification system; Directive 2010/45/EU of the European Parliament and of the Council of 7 July 2010 on standards of quality and safety of human organs intended for transplantation, article 23 (concerning penalties), cf. also art. 11 concerning a reporting system; Directive 2004/23/EC of the European Parliament and of the Council of 31 March 2004 on setting standards of quality and safety for the donation, procurement, testing, processing, preservation, storage, and distribution of human tissues and cells, article 27 (concerning penalties), cf. also art. 11 concerning a notification system. 43  See Martinsen, in den Exter (ed.), p. 5. 44  For an overview of 13 European countries (both primary and specialist care services), see Bernhard A Koch (ed.), Medical Liability in Europe: A Comparison of Selected Jurisdictions (De Gruyter 2011) 617. 45  For an overview, see W. Sauter et al. (eds.), The Law and Policy of Healthcare Financing: An International Comparison of Models and Outcomes (Edward Elgar 2019).

580   Karl Harald Søvig and Barry Furrow is entitled to “fair compensation” according to “the conditions and procedures prescribed by law.” The Explanatory report to the convention is rather brief on this matter. The due or undue nature of the damage must be determined “in the light of the circumstances of each case” and compensation “conditions and procedures” are “prescribed by national law.”46 This leaves to the domestic legislator whether to establish a system of individual liability based either on fault or on the notion of risk or strict liability, or to provide for a collective system of compensation irrespective of individual liability.

6.3.2  European Convention on Human Rights All member states of the CoE have ratified the European Convention on Human Rights (ECHR) for which the main supervision body is the European Court of Human Rights (ECtHR). There are several decisions by the ECHR concerning adverse outcomes of medical treatment. ECHR Article 2 that protects the right to life has been the main legal fundament in many cases. The ECtHR held in Calvelli and Ciglio v. Italy that the positive obligations under Article 2 “require States to make regulations compelling hospitals . . . to adopt appropriate measures for the protection of their patients’ lives” and “an effective independent judicial system to be set up so that the cause of death of patients in the care of the medical profession, whether in the public or the private sector, can be determined and those responsible made accountable.”47 This approach has been followed up in later cases. The decisive issue has normally been whether there has been a sufficient protection of the right to life (substantive part) or whether the subsequent administrative, civil or criminal proceedings has provided an effective protection (procedural part).48 Article 2 will only come into play where the outcome has been death or the situation has been life-­ threatening. Less severe injuries that fall outside the scope of Article 2 may fall in the

46  Explanatory Report to the Convention for the protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine, paras. 144 and 145. 47  Appl. No. 32967/96, judgment (Grand Chamber) 17 January 2002, para. 49. 48  See inter alia Šilih v. Slovenia, 9 April 2009 (Grand Chamber) (lengthy criminal proceedings after death); Eugenia Lazăr v. Romania; 16 February 2010 (art. 2: shortcomings in the domestic court proceedings after death); Mehmet Şentürk and Bekir Şentürk v. Turkey, 9 April 2013 (death of a pregnant woman and violation both regarding the substantial protection of life and the subsequent criminal proceedings); Gray v. Germany, 22 May 2014 (no violation after the death of a patient in his home in the United Kingdom as a result of medical malpractice by a German doctor who had been recruited by a private agency to work for the British National Health Service); Asiye Genç v. Turkey, 27 January 2015 (death of a prematurely born baby and insufficient organization and functioning of the public hospital service and shortcomings in the subsequent judicial review); Altuğ and Others v. Turkey, 30 June 2015 (failure to ensure appropriate implementation of the relevant domestic legislative and statutory framework geared to protecting patients’ right to life); Ioniță v. Romania, 10 January 2017 (lack of a proper investigation into the death of the applicants’ son); Mehmet Günay and Güllü Günay v. Turkey, 20 February 2018 (art. 2 and 6: allegations of medical negligence in relation to death after a hospital stay; the application concerning the right to life was found manifestly ill-­founded, while violation was found regarding lack of a fair trial within a reasonable time).

Medical Liability   581 domain of Article 8, which protects inter alia the “private life.”49 As with Article 2, a procedural part is also developed by the ECtHR under Article 8. Issues purely regarding the litigation process may be handled under Article 6’s right to a fair trial, and Article 13 can be invoked if the applicant does not have an effective remedy.50 In Lopes de Sousa Fernandes v. Portugal, the ECtHR summarized the general principles (paras. 214–215) concerning the procedural requirements:51 The Court has interpreted the procedural obligation of Article 2 in the context of healthcare as requiring States to set up an effective and independent judicial system so that the cause of death of patients in the care of the medical profession, whether in the public or the private sector, can be determined and those responsible made accountable. . . . While, in some exceptional situations, where the fault attributable to the health-­care providers went beyond a mere error or medical negligence, the Court has considered that compliance with the procedural obligation must include recourse to criminal law . . . in all other cases where the infringement of the right to life or to personal integrity is not caused intentionally, the procedural obligation imposed by Article 2 to set up an effective and independent judicial system does not necessarily require the provision of a criminal-­law remedy.

Hence, normally, the obligation is to establish a civil system for medical liability with the exception for exceptional cases where a criminal reaction may be required. The choice of means for ensuring the positive obligations under Article 2 is in principle a matter that falls within the Contracting State’s margin of appreciation (para. 216). There are different avenues for ensuring Convention rights, and even if the state has failed to apply one particular measure provided by domestic law, it may still fulfil its positive duty by other means. Furthermore, the ECtHR also addressed the role of medical reports and experts, which is a main feature of liability cases (para. 217) and stressed the need of independent specialists who can support the domestic courts in their judicial review. Many cases concerning medical liability relate to the lengthiness of the administrative or judicial review, and the ECtHR underlined that prompt examination of such cases is therefore important for the safety of all users of healthcare services (para. 218). States are responsible for establishing a system for patient protection. The ECtHR will hold the states to account in certain ways (e.g., a duty not to discriminate).52 When it comes to damages, the ECtHR will be reluctant to review all the elements of the

49  See inter alia Codarcea v. Romania, App no 31675/04, 2 June 2009, and Vasileva v. Bulgaria, App no 23796/10, 17 March 2016. 50  Cf. Oyal v. Turkey, App no 4864/05, 23 March 2010. 51  App no 56080/13, judgment 19 December 2017. 52  Cf. G. N. and others v. Italy, App no 43134/05, 1 December 2009 (Article 2 in conjunction with Article 14 (thalassaemia sufferers or their heirs had been discriminated against compared with haemophilia sufferers, who had been able to take advantage of the out-­of-­court settlements offered by the state).

582   Karl Harald Søvig and Barry Furrow quantifications. Nevertheless, there are cases where the ECtHR has held that the awarded compensation has been insufficient.53

6.4  National Legislation in Europe All European countries have their separate legislative system regarding compensation for medical errors. It is impossible within the frames of this chapter to provide an in-­ depth analysis of each system, and even a brief overview would be an overwhelming challenge.54 The European Region embraces nearly 900 million people with a significant variation in living conditions.55 There are huge differences, both due to differences in legal systems and legal cultures and partly due to the variations in available resources (technology and medical professions). A core element in all cases concerning medical compensation—in Europe as well other parts of the globe—is the assessment of whether the adverse event was caused by the medical management rather than the underlying disease.56 In Lopes de Sousa Fernandes v. Portugal, the ECtHR provides a short overview of medical liability systems in a selection of the Member States (para. 122–126). Of the 31 Member States surveyed, all offer a civil remedy with the possibility to claim compensation for medical negligence in either the civil or administrative courts. In the majority of countries, liability can be both contractual and extra-­contractual (e.g., in Austria, Azerbaijan, Bosnia and Herzegovina, Bulgaria, Estonia, Georgia, Germany, Italy, Luxembourg, Monaco, Poland, Spain, and Switzerland) while tort liability is the exclusive 53  Cf. Otgon v. Republic of Moldova, App no 22743/07, 25 October 2016 (the applicant had spent 2 weeks in hospital with dysentery after drinking infected tap water and was awarded damages with an amount the equivalent of €648, which the ECtHR held was insufficient for the degree of harm that had been caused to the applicant’s health; even though the judgments concerned a form of product liability, the reasoning is relevant also for healthcare services); Oyal v. Turkey, App no 4864/05, 23 March 2010 (the ECtHR held that the most appropriate remedy in the circumstances would have been to have ordered the defendants, in addition to the payment in respect of non-­pecuniary damage, to pay for the first applicant’s treatment and medication expenses during his lifetime); Reynolds v. UK, App no 2694/08, 13 March 2012 (the UK Supreme Court had just recently, and after the applicant’s case had been decided by domestic courts, confirmed the duty to protect suicide-­risk patients who were voluntarily admitted to hospital but, prior to that date, the applicant had not had any remedy available in respect of her non-­pecuniary loss); Aydoğdu v. Turkey, App no 40448/06 30 August 2016 (due to many cases against Turkey, the ECtHR called upon the state to take measures to require independent and impartial administrative and disciplinary investigations to be carried out within its legal system, affording victims an effective opportunity to take part; to ensure that bodies and/or specialists that could be called upon to produce expert opinions had qualifications and skills corresponding fully to the particularities of each case; and to require forensic medical experts to give proper reasons in support of their scientific opinions). 54  For an overview, see Koch supra 44, and E. Deutch and H.-L. Schreiber, Medical Responsibility in Western Europe (Springer Verlag 1985). 55  See WHO, European Health Report: More Than Numbers—Evidence for All (2018). 56  Cf. T. A. Brennan et al., Incidence of Adverse Events and Negligence in Hospitalized Patients: Results of the Harvard Medical Practice Study I, 324 NEJM 370–376.

Medical Liability   583 or main form of liability in some countries (Lithuania, Malta, Moldova, Serbia, Russia, Ukraine, and the United Kingdom). The above-­mentioned survey indicates that the situation differs within European jurisdictions in which part of the legal system medical liability is regarded as being placed. Another distinction relates to the foundation of the claim. One may divide the legislative approach in the European countries into two main categories57:

a) A culpability system (normally based on contractual and/or tort liability law) b) A non-­fault system (normally based on administrative law)

This division is, of course, rough, and many nuances disappear, but it will still be used in the following text to give an overview. The contractual and/or tort liability system has probably been the common approach in all European legal systems from the early stages of legislative response to medical errors. It stems from the notion that it is a relationship between the patient and the medical doctor/hospital, although it differs whether this is based on a contractual relationship or not. Normally, the contractual origin will be more common in countries where the healthcare services are provided by private hospitals or doctors. For short, this approach will be labelled as “tort liability” and patient’s claims are regarded in line with other cases of tort where prevention or deterrence is an underlying steering value. The core element in tort liability is that financial compensation will be awarded if it is proved that “negligence” has caused the injury.58 The parties will typically be the patient and the medical doctor or the hospital. In many countries, health personnel or institutions are under a legal obligation to have an insurance policy covering the patient’s claim. Normally, the patient must pursue the claim through the ordinary court system. The non-­fault system has emerged in several European countries, often based on the experience that a tort liability system is providing patients with limited protection. It may be too difficult to prove that “negligence” has occurred, including in matters relating to causation, or the costs relating to litigation are too high. The core element of a non-­fault system is that compensation is not established by culpability but more objective criteria. Normally, the claim can be pursued outside the court system, with specialized bodies set up to deal with such cases. The first country to establish such a system was Sweden (1975), and it has been introduced in all of the Nordic countries (Finland 1987, Denmark 1988, Norway 1992, Iceland 2000) but also elsewhere (e.g., France and Belgium).59 A tort liability system may be regarded as more focused on doctor’s duties than a “no-­ fault system,” which may be regarded as placing more emphasis on patient’s rights.60 In general, a non-­fault system will have a wider coverage than a tort liability system. As in a tort liability system, an adverse event that is considered to be caused by a “culpable” act 57  This division is, inter alia, used by K. Watson and R. Kottenhagen, Medical Liability in Europe, in André den Exter (ed.), European Health Law (Maklu 2017) 483. 58  Supra 57, 483. 59  Id. 495–501. 60  Id., 492.

584   Karl Harald Søvig and Barry Furrow or omission will be covered by a non-­fault system. The additional dimension of the non-­ fault system is the coverage of adverse events that are not considered as culpable or where culpability is not proved. The cases of Karl and Anna would probably have been covered in all European countries regardless of the liability system, but the legal questions would have been formulated diversively between formulated and depending on the underlying system. Additionally, the handling of the claim would be different. In the following, an overview of the systems in the three Scandinavian countries will be provided with an emphasis on the material conditions for compensation. Some core elements of the provisions will be cited in English (the authentic texts are in, respectively, Swedish, Danish, and Norwegian, with translation made by the author).

6.5  The Scandinavian Patient Compensation System 6.5.1  Introduction The patient compensation systems in all the Scandinavian countries are established by a separate Act. In Sweden, by the Patient Injury Act (1996:799, abbreviated PIA-­SWE); in Denmark by the Act on the right of appeal and compensation within the national health service (Act no. 547 of 24 June 2005, abbreviated RACNHS-­DEN); in Norway by the Patient Injury Act (15 June 2001 No. 53, abbreviated PIA-­NOR). The regulations are detailed and complicated, and many aspects are not dealt with in the following, thus inaccuracy is unavoidable. While the main study object in the United States is case law, the statutory provisions are at the forefront when examining the Scandinavian legal system. Within the Nordic countries, there is a close legislative cooperation, especially within fields not covered by EU/EEA-­law. When drafting legislation, an integral part of the proc­ess is normally to compare the legal situation within the Scandinavian or Nordic countries. Even if Sweden was the founding mother in this field, its legislation has subsequently been amended and developed. Today, the three Scandinavian countries have been inspiring and been inspired by each other. This means that the legislation has many similarities but also differences, and some of these elements will be highlighted here. Which kind of services are included under the patient compensation system differs a bit between the Scandinavian countries, but they all cover primary (e.g., general practitioners) and secondary (e.g., hospitals) healthcare. If a liability issue arises for a service not included under the patient compensation system, the patient will have to pursue the claim according to general principles of tort law.

6.5.2  Foundation of the Claim 6.5.2.1  General Legislative Basis In all three Scandinavian countries, a non-­fault system is established by legislation. The core content of a non-­fault system is that the patient does not have to prove negligence or culpability of the healthcare provider to be awarded compensation. The historical

Medical Liability   585 ­ ackground is that such a threshold was found to be too high for patients and also b associated with excessive costs related to the judicial procedures. A non-­fault system does not mean that all adverse effects of medical treatment and care are compensated. It could perhaps be more adequate to describe “non-­fault” as holding elements of strict liability. A non-­fault system offers the patient better protection than does a system based on culpability. However, even within a non-­fault system, a division must be made between the negative effects that are a consequence of the disease and those attributed to the healthcare providers. It is perhaps a paradox, but also within a non-­fault system the crux of the matter is—to some extent—whether something “went wrong.” Could the patient reasonably expect to undergo the treatment without injuries, given the severity of the disease and the available expertise within the healthcare sector? The spotlight is somewhat different from a culpability assessment, where the focus is on the possible errors, while the non-­fault system may be more concentrated on the system. Still, a crucial point is normally whether some form of substandard treatment is provided. However, a non-­fault system may have more elements of strict liability, especially in some specific situations where the responsibility is stipulated by the legislation. Since non-­fault systems are introduced by the legislature, the starting point for examination will be legislative texts (i.e., statutory provisions). Additionally, non-­fault systems will normally also offer patients a separate procedure to obtain compensation in which a separate body is established to handle the claims. This separate body can be a unit within administrative law, as in Denmark and Norway, or an insurance entity, as in Sweden (see later discussion). The aim is to provide the patient with a procedure that is less costly and swifter than a court trial. In the forthcoming text, a brief overview of the Scandinavian system will be offered. All three jurisdictions cover the “traditional cases” in forms of failures relating to examination/treatment and diagnosis. In Sweden and Denmark, the crux of the matter is whether the injury could have been avoided. The approach may be formulated as a matter of “sound medical practice.”61 Both in Denmark and Sweden, the fictitious assessment of an experienced specialist is placed in the forefront.62 In Norway, the approach is somewhat different, with “failure” as the operational condition, emphasizing that also situations where no individual can be blamed are covered.63 Still, when assessing what is a “failure” according to the Norwegian assessment, the “sound medical practice” will indeed also be a relevant element. Additionally, the patient perspective is included by explicitly taking into consideration the “reasonable” expectations of the patient, while the interests of the society are explicitly included by referring to allocation of resources. Insufficient resources shall not lead to liability if the allocation of resources has been acceptable and the service is commonly holding a reasonable standard. An illustrative example can be if a hospital lacks infection control equipment during a pandemic due to limited supplies, and this cannot be traced back to 61  Cf. PIA-­SWE section 6 No. 1 and RACNHS-­DEN section 20 No. 1. 62  Cf. RACNHS-­DEN section 20 No. 1 and PIA-­SWE section 6 second paragraph. 63  Cf. PIA-­NOR section 2, a.

586   Karl Harald Søvig and Barry Furrow insufficient planning. However, even if a similar provision is not found in the neighboring countries, these dimensions may also be taken into account by the Swedish and Danish legislation as part of the ‘sound medical practice” assessment. Sufficient information—which will normally be providing the patient with the knowledge required to understand the risk prior to an intervention—is explicitly mentioned in the Norwegian provisions,64 while this element must be included in Sweden and Denmark as part of the assessment of whether the injury could have been avoided if the patient had received relevant information. A particularity of the Norwegian legislation is the extraordinary inclusion clause, covering injuries which are especially severe or unexpected and that cannot be considered as an outcome that the patient must bear. This dimension has no direct counterpart in Sweden and Denmark. The Norwegian approach may be illuminated by the judgment of the Supreme Court in HR-­2018-­1056-­A (the judgment also addressed the issue of causation; see later discussion). During hip surgery, the patient suffered nerve damage to his legs caused by pressure against both thigh nerves. The nerve damage did not result in motoric dysfunction but led to loss of feelings in the skin. The patient had to be “fixed” during surgery in order for the physician to have sufficient working conditions, and the nerve damage was, according to the court-­appointed medical expert, caused by a combination of special anatomic conditions of the patient and the long-­lasting position during surgery. The Supreme Court emphasized that the legislator did not intend to introduce a rule of absolute liability. Responsibility requires a “failure,” which implies an assessment depending on common good and established practice within the relevant medical field at the time of treatment. Thus, the patient was not awarded compensation. Regarding diagnosis, some elaborations are needed. In Sweden, incorrect diagnosis is a separate foundation for compensation.65 In Denmark, incorrect diagnosis is explicitly mentioned but is to be assessed according to the general provisions.66 In Norway, diagnosis is not mentioned in the provision and must be regarded as a matter of whether a “failure” has occurred. HR-­2017-­687-­A concerned information for the patient in order to avoid future risks after being discharged. After a consultation with a cardiac specialist, a patient was informed that his symptoms were not due to heart disease but a myalgia disorder. The patient had a slightly higher risk of future cardiovascular disease than the average for his age group. He was not given further information about this risk or about what to do if he experienced the symptoms of a heart attack. About 3 months later, the patient suffered a heart attack but did not go to the doctor until the following day. The Supreme Court concluded that the lack of information from the doctor did not provide a basis for patient injury compensation.67 It was pointed out that not every injury caused by medical help entitles a patient to patient injury compensation. There must be a “failure” in the medical treatment. Following a specific assessment, the Supreme Court concluded that there had been no failure. The obligation to inform about potential risk factors had to be 64  Cf. PIA-­NOR section 2, third paragraph. 65  Cf. PIA-­SWE section 6 no. 3. 66  Cf. RACNHS-­DEN section 21. 67  Cf. PIA-­NOR, section 2, subsection 1, letter a).

Medical Liability   587 considered in view of the fact that the risk of heart disease in the near future was very small, not in view of the fact that he actually suffered a heart attack a few months later. It was within the legally recognised scope of practice not to provide such information when the examination had revealed that the patient did not have any heart disease. Turning to the case studies, Karl’s bowels would likely fall under the coverage in all the Scandinavian countries. Given the facts, an error was made by the physicians who overlooked symptoms that indicated that the wire had caused injuries. An earlier detection and intervention in the form of surgery would have impeded or reduced the injury. Thus, the adverse outcome caused by was a “failure” (according to the Norwegian approach), or an experienced specialist would have intervened at an earlier stage (according to the Swedish and Danish approach). Anetta’s tumors will be dealt with next, under the coverage of specific situations. 6.5.2.2  Coverage of Specific Situations Moving to the coverage of different incidents, the use of technical devices is included in all Scandinavian jurisdictions.68 An illustrating example may be injuries caused by pacemakers or implantable cardioverter-­defibrillators (ICD) that malfunction. The wording is a bit different (failure or defect) in the statutory provisions, but, in all jurisdictions, it is a form of strict liability relating to the equipment itself. Improper use is covered in Sweden by the special provision concerning devices, while in Norway and Denmark this situation will fall under the ordinary provisions of examination and diagnosis. All three jurisdictions cover infections.69 In Sweden and Denmark, infections that must reasonably be tolerated are exempted,70 while in Norway the exception is somewhat differently formulated. The emphasis is here whether this is not mainly caused by the patient’s condition or disease. Still, the different approaches may not lead to diverse outcome in parallel cases. If a patient with an open fracture is undergoing surgery and the infection is caused by bacteria present in the wound, he or she will not be awarded compensation except if the treatment is substandard (e.g., if the patient is not provided antibiotics according to ordinary guidelines). Accidents (e.g., a fire in the hospital) are explicitly mentioned in both Sweden and Denmark, but with different approaches. In Sweden, the patient is covered if the accident takes place during care or transport.71 In Denmark, the responsibility is more limited (incidents related to transport will not be dealt with here). Compensation is only awarded if there would be liability under general rules of compensation law. In this respect, Denmark is more in line with Norway. Accidents are not explicitly mentioned in the Norwegian legislation but would be covered if the incidents would fall under general rules of compensation law.72 Hence, a hospital fire caused by improper storage 68  Cf. PIA-­SWE section 6, no. 2, RACNHS-­DEN section 20, no. 2, and PIA-­NOR section 2, b. 69  Cf. PIA-­SWE section 6 no. 4, RACNHS-­DEN section 20 no. 4, and PIA-­NOR section 2, c. 70  Cf. PIA-­SWE section 6 no. 4, RACNHS-­DEN section 20, second paragraph. 71  Cf. PIA-­SWE section 6 no. 5. 72  Cf. PIA-­NOR section 2, e (see also below).

588   Karl Harald Søvig and Barry Furrow of inflammable substances harming patients would lead to compensation in all jurisdictions, while a hospital fire caused by a lightning strike would only lead to compensation in Sweden (assuming that necessary precautions were made to prevent fire after lightning). In Denmark and Sweden, vaccination is not explicitly covered, as it is in Norway.73 Still, side effects after vaccination can be compensated in Denmark and Sweden under the rules regulating pharmaceuticals (see later discussion).74 The purpose of including side effects after vaccination in a separate provision under the Norwegian system is the societal interest in vaccination. Those exposing themselves for this (often the decision is taken by parents on behalf of their children) should be covered if the minimal risk that is attached to vaccination occurs.75 The use of pharmaceuticals is paid special attention in all the Scandinavian countries, although with a somewhat different approach. The common issue in all countries is how to define the line between drug-­related injuries that are to be covered under the ordinary patient compensation regime and those to be covered under the distinct system of pharmaceutical liability. In Sweden, the coverage under the patient liability system is limited to prescription or provision of pharmaceuticals in contravention of regulations or instructions.76 If drugs are administered in violation of regulations or instructions, there may often be some sort of culpability by the healthcare providers. In Denmark, compensation is not awarded for injuries due to properties of the medicines used in examination, treatment, etc.77 In Norway, the use of pharmaceuticals is not regulated by the patient compensation regime, except those aspects of the regime relating to vaccines. All three Scandinavian countries have separate regimes for pharmaceutical liability. Denmark’s regime is an independent part of the patient liability act.78 In Norway, separate legislation governs this issue,79 while in Sweden the general law on damages applies.80 In all countries, pharmaceutical liability is covered by Directive 85/374/EEC concerning liability for defective products. In this field, the Directive provides for a full harmonization of European rules concerning the basis for liability. A question whether the emotional harm to third parties is compensable at all was recently decided by the Norwegian Supreme Court (cf. HR-­2018-­2080-­A). The case concerned a 15-­year-­old boy who was already severely disabled, including an inability to communicate. He was admitted to the hospital because of stomach pain and died during the stay due to healthcare failure. He was found without breath in the morning by his mother who stayed at the hospital by his bedside, and resuscitation was unsuccessful. 73  Cf. PIA-­NOR section PIA-­NOR section 2, d. 74  Cf. RACNHS-­DEN section 38. The Danish Supreme Court has decided two cases relating to vaccination, U.2005.2151H and U.2006.2143H (cases dismissed to due lack of proof of causation). 75  Regarding causation in this respect, see the judgment by the Norwegian Supreme Court in Rt. 2015 p. 1246. 76  Cf. PIA-­SWE section 6 no. 6 and section 7 no. 2. 77  Cf. RACNHS-­DEN section 21, third paragraph. 78 Id. 79  Cf. the product liability act (23 December 1988 No. 104) chapter 3. 80  Act 1972:207.

Medical Liability   589 The Supreme Court held that the boy’s mother was entitled to compensation. The parties agreed that there existed a basis for liability81 and a causal relationship between the healthcare failure and the mother’s mental injury, and the Supreme Court also found that the injury was not so unforeseeable and remote as to exclude liability. The crux of the matter was whether such shock-­related injuries by third parties were compensated, and the Supreme Court gave an affirmative answer. Turning to the case studies, Anetta’s tumors would likely qualify for compensation in all the Scandinavian countries. The trigger factor was the error in the transfer of data from the machine learning project to the hospital. This could be regarded as a matter under technical devices, but would probably be considered as a form of ‘system failure” where the personnel trusted the machine learning system without human cross-­checking.

6.5.3  Causation In both Sweden and Denmark, questions relating to causation are addressed as an integrated part of the provisions stating liability. The words used in the provisions can be translated into “preponderant probability.”82 This leaves the risk of doubt to the patient. However, it is sufficient that the probability is just above 50% in order to achieve compensation (if the other conditions are fulfilled).83 In Denmark, the preparatory works states that if it is a “clear failure” by the healthcare provider, it will be reasonable to let any doubt regarding the causation to be in favour of the patient.84 In Norway, questions regarding causation are regulated in a separate provision.85 If the cause of a patient’s injury cannot be established, and the injury is likely to be due to external influences on a patient during treatment, it should normally be assumed that the injury is due to malfunction or failure in the performance of healthcare. In HR-­2018-­1056-­A (regarding nerve damage, cf. above), the Supreme Court held that the conditions for compensation pursuant to the separate provision had been met. The provision is, however, an evidence rule and not a provision on strict liability. It does not permit exemptions from the basic condition (“failure”). Standard procedure had been followed, and it was assumed that no error had been committed. The appeal against the court of appeal’s judgment in favour of the respondent was dismissed. The Norwegian legislation also regulates causation relating to vaccines. In case of compensation as a result of damage caused by recommended or mandatory vaccinations, the right to compensation pursuant to PIA-­NOR Section 2(d), applies, unless the person responsible proves that one or more other causes are more probable. This gives the vaccinated person a stronger position regarding causation than other persons, and again, the object is to underpin the vaccination programmes. In Rt. 2015 p. 1246, a 12-­year-­old boy had been diagnosed with multiple sclerosis just a couple of months after being vaccinated for the second time (the ordinary mumps-­measles-­ rubella [MMR] vaccine programme provided to all children). The court-­appointed 81  Cf. PIA-­NOR section 2, a. 82  Cf. PIA-­SWE section 6 and RACNHS-­DEN section 20. 84 Id. 83  Cf. Mette Hartlev, Sundhed og Jura, 2013, p. 316. 85  Cf. PIA-­NOR section 3.

590   Karl Harald Søvig and Barry Furrow expert had separate opinions, and there was also disagreement between the experts engaged by the parties. The Supreme Court held that it was a practical possibility that the vaccine could cause the disease. The boy would not in the foreseeable future have developed multiple sclerosis without the vaccination. Since no other cause was more likely than the vaccination, the Supreme Court held the state liable.

6.5.4  The Level of Compensation In all the Scandinavian countries, the level of compensation is a starting point to be ­settled according to the general principles and legislation of tort law.86 As a general principle of tort law in all three countries, the injured shall have compensated their individual economic loss. Compared with the United States, the level of compensation is modest. Partly this is due to the welfare model in the Scandinavian countries. Expenses that are covered by the social security system, either in cash (e.g., disability pension) or in-­kind (e.g., assistance at home) will not be compensated. Additionally, the compensation for non-­economic loss is in general lower than in the United States. A practical example of the estimated compensation in the case study of Karl’s bowels will be provided here. The legislatures in all Scandinavian countries have made some exceptions from the general tort law, of which one is common. A fixed amount shall be deducted from the compensation. The objective is to avoid disputes about “minor” injuries. If the loss is below the fixed amount, it is not compensable. The fixed amount to be deducted from the awarded compensation is stipulated directly in the act in Denmark and Norway, with the difference that the Danish legislation has introduced an inflation clause, adjusting it on a yearly basis. The deduction amount is currently DKK7,792 (approximately €1,199 or US$1,200) and NOK10,000 (approximately €960 or US$1,030). In Sweden, the act refers to the social security act and the “basic amount” that is made administratively according to a fixed formula integrating the consumer index.87 The deduction amount is currently SEK2365 (approximately €245 or US$265). The living costs in the Scandinavian countries are rather similar, so the Swedish deduction amount is significantly lower than the two neighbouring countries. When it comes to the other exceptions, legislation differs in the Scandinavian countries. In Sweden, a cap is stipulated by law referring to a certain factor of the basic amount. The maximum level of compensation for each patient is currently SEK9, 46 million (approximately €979,000 or US$1,060,000).88 If several patients are injured in the same incident, the maximum amount is currently SEK47.3 million (approximately €4,900,000 or US$5,300,000). If the maximum cap has as consequence that not all patients can have full compensation, a proportionate division shall be made.89 In Denmark, the parents are entitled to a fixed amount in the event of a death of a child, defined as a person under the age of 18 years. Since there will be normally no 86  Cf. PIA-­SWE section 8, RACNHS-­DEN section 24 and PIA-­NOR section 4. 87  Cf. PIA-­SWE section 9. 88  Cf. PIA-­SWE section 10. 89  Cf. PIA-­SWE section 11.

Medical Liability   591 economic loss for the parents when losing a child, a lack of special provision would otherwise leave the parents without compensation. The amount is currently DKK162,000 (approximately €23,500 or US$25,500).90 In Norway, the injured person is entitled to satisfactory compensation if the offender has acted willfully or with gross negligence. However, satisfactory compensation is not covered by the patients’ injury compensation scheme.91 In such cases, the patient must pursue the claim for satisfactory compensation in a case against the perpetrator according to tort law, while the patients’ injury compensation is responsible for the ordinary claim. Satisfactory compensation is rare, and the awarded amounts are rather low. Turning to the example of Karl’s bowels, the general starting point would be to stipulate the individual economic loss. Here, the Norwegian regulations are used as a basis, but the approach would be rather similar in the other Scandinavian countries. The main components of the compensation would be - Loss of income due to inability to work (until the estimated retirement age which often is 67 years). Disability pension according to the social security system shall be deducted, as well as collective disability pensions that he would be entitled to. Private pensions will normally not be deducted. If the patient will be entitled to a lower retirement pension after the age of 67, this loss will also be compensated. Loss of income until settlement (or court decision) is taxed as ordinary income, while loss of future income is exempted from income taxation but will be taxed as assets. The loss of income could be estimated to around NOK850,000 (approximately €80,000 or US$94,000). - Expenses, including the need for assistance at home, beyond those provided by the social security system. The lost ability to carry out domestic work would also be compensated, normally to the age of 75 since most people after this age will have assistance with domestic work. The expenses, including assistance for domestic work, could be estimated to around NOK200,000 (approximately €19,000 or US$22,000). - Non-­economic loss for significant and long-­lasting medical injury. This is made on the basis of the awarded compensation under the social insurance scheme for work-­related injuries, with a surplus. It depends on the invalid level and age. For Karl’s case, with a shortened intestine and the need for intravenous nutrition, the invalid grade would be between 25% and 44%, and the non-­economic loss could be estimated to around NOK450,000 (approximately €42,000 or US$50,000). In total, Karl’s example would lead to a compensation of NOK1,500,000 (approximately €140,000 or US$165,000). This is indeed an estimation of the awarded damages but gives an indication of the differences between the US and Scandinavian approaches.

90  Cf. PIA-­NOR section 24a.

91  Cf. PIA-­NOR section 4.

592   Karl Harald Søvig and Barry Furrow An interesting case is pending before the Norwegian court system, one that can illustrate the different levels of compensation in Norway and the United States.92 Nora was born at a hospital in Bergen, Norway, after a caesarean section that was made too late. She suffered from severe brain damage and the parties agreed that the injury was compensable under the patient’s injury scheme. She was awarded NOK10.3 million (approximately €980,000 or US$1,060,000) in compensation. The parents moved to the United States and claimed that the compensation should be calculated on the basis of the expenses in their new country of residence with its limited social security system. The compensation claim was NOK55.5 million (approximately €5,200,000 or US$5,700,000). The city court held that the change of residence to the United States was not necessarily due to the medical error, although understandable. Thus, the increased expenses were not compensable. The judgment is under appeal, and the parents have applied for a direct appeal to the Supreme Court.

6.5.5  Interplay Between Ordinary Compensation Rules and the Patient Compensation Scheme The introduction of a separate regime for patient compensation raises a question regarding the interplay with the established compensation rules. In Sweden and Norway, this interplay is explicitly regulated but using different approaches. In Norway, the patient can pursue his or her claim under the patient liability system even if the basis is a claim covered by established compensation rules. In Sweden, such a claim can be pursued outside the patient liability system.93 In Denmark, if a patient is covered by the patient liability scheme, he or she cannot submit a claim according to established compensation rules.94 In practice, the differences are of little interest since patients prefer to seek compensation according to the patient liability systems.

6.5.6 Organisation In Sweden and Denmark, healthcare providers are obliged to have an insurance policy covering patient liability claims.95 In Denmark, the state, counties, and municipalities are exempted from the insurance obligations. In Norway, the funding is based on contributions from the healthcare providers.96 The ministry can by regulation stipulate that providers could or should take out an insurance policy to cover the responsibilities according to the act.97 In Sweden, the patient submits the claim to the insurance company. A nationwide Swedish insurance company is established whose main task is to insure publicly financed healthcare providers. If the patient is not satisfied with the outcome of the decision by the insurance company, the patient can bring the case before the patient claims

92  18-­124767TVI-­BERG/2 (the judgment is in Norwegian only). 93  Cf. PIA-­SWE section 18. 94  Cf. RACNHS-­DEN section 26. 95  Cf. PIA-­SWE section 12 and RCHNHS-­DEN section 30. 96  Cf. PIA-­NOR sections 7 and 8. 97  Cf. PIA-­NOR section 8a.

Medical Liability   593 panel.98 The panel issues advisory opinions on matters concerning compensation under the PIA-­SWE. However, the opinions of the panel will normally be followed. If the provider has not entered into an agreement with an insurance company, the claim can be submitted directly to the board. In both Denmark and Norway, there is a two-­level system. The claim can be submitted to a board.99 In Denmark, the decision-­making competence in the first instance can be delegated to the insurance companies or self-­insured providers. The decision by the board—or in Denmark in certain cases the decision on behalf of the board—can be appealed by the patient to the appeal board.100 It is therefore a one-­sided two-­level system in which the providers cannot appeal. The board and the appeal body are part of the administrative branch but with a certain autonomy. They cannot be instructed by the government regarding the assessment in individual cases.101 The decisions by the panel in Sweden can be brought before the courts, and the patient can also bring the decision by the insurance company directly before the courts. These cases are handled by the ordinary courts and not the administrative courts, which indicates that these disputes in Sweden are regarded as civil law. A decision by the appeal board in Denmark and Norway can be brought before the courts.102 It should be noted that in Denmark both the provider and the patient can instigate court proceedings (in practice, most writs of summons are from patients), while in Norway only persons claiming compensation can instigate court proceedings. Both Denmark and Norway have general courts with three levels of jurisdiction. The organization in the three countries is rather different. Sweden has an insurance system, organized with an insurance company with compulsory requirements for equity. The former patients are “customers” when applying for compensation, and the insurance company is under the same regulatory scheme as other insurance companies. In Denmark and Norway, the bodies are administrative entities. The former patients are “citizens” when applying for compensation, and the bodies are under a public law regulatory regime governed by administrative law. A noteworthy part of the legislation in all the Scandinavian countries is the duty for healthcare providers to inform patients about adverse events. The wording is a bit different, but in all three countries it is stipulated by law that the patient shall be informed of the possibilities to seek compensation after incidents that have led to injuries that may be compensable.103

98  Cf. PIA-­SWE section 17. 99  Cf. RACNHS-­DEN section 32–34 and PIA-­NOR section 9. 100  Cf. RCHNHS-­DEN section 58b and PIA-­NOR section 15. 101  Cf. RACNHS-­DEN section 11, second paragraph (for Norway’s part, see regulation 20 December 2002 No. 1625 section 2). 102  Cf. RACNHS-­DEN section 58d and PIA-­NOR section 18. 103  Cf. RACNHS-­DEN section 23, the Swedish Patient Safety Act (2010:659) section 8 and the Norwegian Patient’s Rights Act (2 July 1999 No. 63) section 3–2 fourth paragraph.

594   Karl Harald Søvig and Barry Furrow

6.5.7  Statistics All the Scandinavian patient insurance schemes publish annual statistics, and some key figures will be presented here. However, the indicators used in the countries are not exactly the same, and it should be remembered that the coverage may be different. Moreover, even if the same indicators are used, there could be differences in the definitions. Hence, any cross-­border comparison should be made with care. Annual decisions 20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 NORWAY

SWEDEN

DENMARK

Figure 28.1  Annual decisions. Source: Data from the Swedish Patient Insurance (Landstingens Ömsesidiga Försäkringsbolag), the Danish Patient Compensation Association (Patienterstatningen), and the Norwegian System of Patient Injury Compensation (Norsk pasientskadeerstatning).

Figure 28.1 includes the annual number of decisions in the Scandinavian countries. Figure 28.1 indicates in all countries an increase in number of cases. In Sweden, the nationwide insurance company expects the number of applications for damages to rise.104 The main factors are identified as more treatment given to a population living longer, more complex medical interventions, more active work on patient’s safety, and the strengthening of the obligation to inform patients of the possibilities to apply for compensation (a reform that came into force in 2011). The same factors are present in all the Scandinavian countries, as well in the rest of Europe (with some modifications concerning the rules regarding information obligations). Figure  28.2 includes the percentages of cases where claim is successful in the Scandinavian countries. 104  https://lof.se/patientsakerhet/skadestatistik/ (accessed 1 August 2020, only in Swedish).

Medical Liability   595 Successful claims 50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00%

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 NORWAY

SWEDEN

DENMARK

Figure 28.2  Successful claims. Source: Data from the Swedish Patient Insurance (Landstingens Ömsesidiga Försäkringsbolag), the Danish  Patient  Compensation Association (Patienterstatningen), and the Norwegian System of Patient Injury Compensation (Norsk pasientskadeerstatning).

Figure 28.2 indicates some differences in the Scandinavian countries although it is difficult to draw conclusions. Thus, it should be emphasized that the figures are either rather stable (Sweden) or slightly decreasing (Norway and Denmark), which indicates that more applications are not necessarily leading to more successful claims. The available statistics all provide information on the different categories leading to compensation, but they are rather differently organized, which makes comparison difficult. In Denmark, the public statistics are divided based on service provider (somatic/psychiatry, primary/secondary), while in Sweden and Norway they are also presented by the medical field although with rather different level of specification. Of the six categories used in the Swedish report (figures from 2017 and 2018), orthopaedics represented 26% of the awarded damages (28% of the cases), followed by cases of maternal care (including pregnancy, delivery, and postnatal care) representing 19% of the damages (8% of the cases) and surgery representing 10% of the damages (10% of the cases).105 Other specialties accounted for 35% of the damages (32% of the decided cases). Of the 30 medical fields used in the Norwegian report (figures from 2019), orthopaedics represented 24% of the awarded damages (31% of the decided cases), followed by tumors and cancer representing 20% of the damages (14% of the cases) and birth (injuries of

105  https://lof.se/wp-­content/uploads/Statistik-­2018-­Hela-­Sverige.pdf (accessed 1 August 2020, only in Swedish).

596   Karl Harald Søvig and Barry Furrow child) representing 7% of the damages (1% of the decided cases).106 Other specialties accounted for 49% of the damages (54% of the decided cases).

7  Joint Conclusion The developing trends within health law are rather similar in all European countries as well as in the United States. The population is growing elderly, and many are having health problems in their later life stages. The medical field is becoming able to cure and alleviate more conditions and offer more advanced and personalized treatment. Seldom is medical treatment a one-­to-­one relationship between patient and physician. New technology is part of this development, one in which AI will take a dominant position. In sum, this development produces high expectations for medical treatment from both society and patients. Still, not all patients will be cured, and things will go wrong. Patients will seek compensation for medical errors, and the ongoing developing trends may lead to more cases concerning medical liability and certainly rather more complex cases. The European countries have different legal responses to issues concerning medical liability, and harmonization is not foreseeable in the near future, driven either by the EU or the CoE. Questions regarding medical liability are too closely connected to the funding of the healthcare system and the ongoing debates of the sustainability of the welfare system. However, both legislation by the EU and judgments by the ECtHR will set some cornerstones, especially regarding the procedural framework for medical liability. At the same time, it will probably be a development in which the jurisdictions are inspired by each other, and perhaps we will see more countries adopting a non-­fault system. An advantage of non-­fault systems may be that they are able to include failures that cannot be attributed to individuals, such as anonymous and cumulative errors (e.g., incidents that involve both humans and machine learning). Movement toward any form of no-­fault compensation system is currently lacking in the United States. The current political climate—with one of the two major political parties averse to expansion of tort systems generally, and lack of popular pressure—creates powerful resistance to any innovation such as a no-­fault compensation system for medical injuries. Some counter pressures, however, can be seen. First, the level of adverse events in US hospitals seems to be increasing in spite of Medicare efforts to improve quality through Centers for Medicare and Medicate Services (CMS) reimbursement penalties, readmissions, and hospital-­acquired conditions. Medicare is the single largest regulator of quality in American hospitals; if Medicare were to impose some form of patient compensation system in hospitals for only Medicare patients, it is likely that such a compensation system would be adopted over time for all patients in hospitals, not just Medicare patients, due to inevitable hospital administrative pressures to treat all patients consistently. 106  https://www.npe.no/no/Om-­NPE/statistikk/statistikkoversikt/vedtak—medisinsk-­omrade/ (accessed 1 August 2020, only in Norwegian).

Medical Liability   597 Second, and connected to the previous observation, the growth in the sheer size of baby boomers retiring into the Medicare system provides energy for disruptive changes.107 This age cohort is more affluent and educated and is likely to be more demanding of higher quality care as they retire into Medicare coverage from their employment-­based private insurance and age into more chronic illnesses. Third, the consumer movement in healthcare—manifested in patient engagement models and ideals of transparency in hospital billing—is sharpening citizen awareness of system failures in the US healthcare system.108 The goal of transparency in hospital billing, pursued by the current administration, overlaps with other forms of transparency, including reporting on adverse events and bad outcomes. And fourth, the system effects of COVID-­19 on access to healthcare, cost, and patient harms are only now becoming apparent. The long-­term health effects of the pandemic on a large percentage of the population, the visibility of gaps in insurance coverage, and the failures of many institutions such as nursing homes to protect their populations—all may make healthcare reform again an agenda item for reform by a new administration. The Scandinavian no-­fault systems provide a useful model for such reform, and one that the United States would do well to learn from.

Bibliography Joint Introduction Bismark, Marie, & Ron Paterson, No-Fault Compensation in New Zealand: Harmonizing Injury Compensation, Provider Accountability, and Patient Safety, 25 Health Aff. 278 (2006). den Exter, André (ed.), European Health Law (Maklu, Antwerpen/Apeldoorn/ Portland 2017).

The US System Baker, Tom, The Medical Malpractice Myth (2005). Classen, David C. et al., “Global Trigger Tool” Shows that Adverse Events in Hospitals May Be Ten Times Greater than Previously Measured, 30 Health Aff. 581, 581 (2011). Dep’t. of Health & Human Servs. Adverse Events in Hospitals: National Incidence Among Medicare Beneficiaries 15, 22 (2010), available at http://oig.hhs.gov/oei/reports/oei-06-0900090.pdfX Hyman, David, Charles Silver, & Randall  R.  Bovbjerg, Legislation on Medical Malpractice: Further Developments and a Preliminary Report Card, 22 U.C. Davis L. Rev. 499 (1989). Hyman, David  A., & Charles Silver, Medical Malpractice and Compensation in Global Perspective: How Does the US Do It, 87 Chicago-Kent L. Rev. 162 (2011).

107 c2b Solutions, Understanding Boomers, a Disruptive Force in American Healthcare 2, 7 (2019). 108  See Julia James, Health Policy Brief: Patient Engagement, Health Aff., Feb. 14, 2013, at 1–2.

598   Karl Harald Søvig and Barry Furrow Hyman, David A., Double, Double, Toil and Trouble: Justice-Talk and the Future of Medical Malpractice Litigation, 63 DePaul L. Rev. 547 (2014).

The Scandinavian System Institute of Medicine. Graham R, Mancher M, Wolman DM, Greenfield S, Steinberg E, Eds.), Clinical Practice Guidelines We Can Trust 2 (2011), http://www.iom.edu/ Reports/2011/Clinical-Practice-Guidelines-We-Can-Trust/Standards.aspx James, John, A New, Evidence-based Estimate of Patient Harms Associated with Hospital Care, 9 J. Patient Saf. 222 (2013). Kachalia, Allen B. et al., Beyond Negligence: Avoidability and Medical Injury Compensation, 66 Soc. Sci. & Med, 387, 389 (2008) (comparing the patient compensation approaches of Sweden, Denmark and New Zealand). King, Joseph  H.  Jr., Causation, Valuation and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequences, 90 Yale L.J. 1353 (1981). Koch, Bernhard  A. ed., Medical Liability in Europe: A Comparison of Selected Jurisdictions (De Gruyter 2011). Mello, Michelle et al., Administrative Compensation for Medical Injuries: Lessons from Three Foreign Systems, 14 Commonwealth Fund Pub. 1 (2011). Oliphant, K., Medical Malpractice and Compensation: Comparative Observations, in K.  Oliphant et al. (eds.), Medical Malpractice and Compensation in Global Perspective (Berlin: De Gruyter 2013) 514. Sauter, Wolf et al. (eds.), The Law and Policy of Healthcare Financing: An International Comparison of Models and Outcomes (Edward Elgar 2019). Watson, Kenneth, & Rob Kottenhagen, Patients’ Rights, Medical Error and Harmonisation of Compensation Mechanisms in Europe, 25 Eur. J. Health L. 1 (2018).

Case Studies of System Differences Anetta’s Tumors Chou, Roger, Carlos Cuevas, Rongwei Fu, Beth Devine, Ngoc Wasson, Alexander Ginsburg, Bernadette Zakher, Miranda Pappas, Elaine Graham, & Sean Sullivan, Imaging Techniques for the Diagnosis and Staging of Hepatocellular Carcinoma, Comparative Effectiveness Review Number 143 (AHRQ). Froomkin, A. Michael, Ian Kerr, & Joelle Pineau, When AIs Outperform Doctors: Confronting the Challenges of a Tort-Induced Over-Reliance on Machine Learning, 61 Ariz. L. Rev. 33 (2019).

Medical Liability in the United States Standard of Care Avraham, Ronen, Overlooked and Underused: Clinical Practice Guidelines and Malpractice Liability for Independent Physicians, 20 Conn. Ins. L.J. 273 (2013–2014). Cooke, Brian K. et al., The Elusive Standard of Care, 45 J. Am. Acad. Psychiatry L. 358 (2017) Ferrara, Santo Davide, Eric Baccino, Thomas Bajanowski, Rafael Boscolo-Berto, Maria Castellano, Ricardo De Angel, Alvydas Pauliukevičius, Pietrantonio Ricci, Peter Vanezis,

Medical Liability   599 Duarte Nuno Vieira, Guido Viel, Enrique Villanueva, & The EALM Working Group on Medical Malpractice, Malpractice and Medical Liability: European Guidelines on Methods of Ascertainment and Criteria of Evaluation, 127 Int. J. Legal Med. 545, 552 (2013). Institute of Medicine, Graham R, Mancher M, Wolman DM, Greenfield S, Steinberg E, eds., Clinical Practice Guidelines We Can Trust 2 (2011), http://www.iom.edu/ Reports/2011/Clinical-Practice-Guidelines-We-Can-Trust/Standards.aspx Peters, Philip  G.  Jr., The Quiet Demise of Deference to Custom: Malpractice Law at the Millennium, 57 Wash. & Lee L. Rev. 163 (2000). Rosoff, Arnold  J., The Role of Clinical Practice Guidelines in Healthcare Reform, 5 Health Matrix 369 (1995). Taylor, Chris, The Use of Clinical Practice Guidelines in Determining Standard of Care, 35 J. Legal Med. 273 (2014).

Causation Newman-Toker, David et al., Serious Misdiagnosis-related Harms in Malpractice Claims: The “Big Three”—Vascular Events, Infections, and Cancers, 6 Diagnosis 228, 233 (2019).

Damage Innovations: The Loss of a Chance Doctrine Guest, Lauren, David Schap, and Thi Tran, The “Loss of Chance” Rule as a Special Category of Damages in Medical Malpractice: A State-by-State Analysis, 21 J Legal Econ. 53, 58–60 (2015).

Informed Consent Furrow, Barry R., Smashing into Windows: Limits of Consumer Sovereignty in Healthcare, in Transparency in Heath and Healthcare in the United States: Law and Ethics (I. Glenn Cohen, Holly Fernandez Lynch, Barbara  J.  Evans, and Carmel Shachar, eds.) (Cambridge University Press, 2019).

Hospital Liability Introduction Agency for Healthcare Research and Quality (AHRQ), 2016 Comparative Health System Performance: Snapshot of US Health Systems. American Hospital Association, Fast Facts on US Hospitals, 2018. https://www.aha.org/statistics/fast-facts-us-hospitals

Enterprise Compensation Possibilities: Emergence of Data Analytics to Detect Adverse Event Hotspots Furrow, Barry R., Searching for Adverse Events: Big Data and Beyond, 27 Ann. Health L. 1 (2018). Furrow, Barry  R., Adverse Events and Patient Injury: Coupling Detection, Disclosure, and Compensation, 46 N. Eng. L. Rev. 437 (2012).

600   Karl Harald Søvig and Barry Furrow Liability Reform in the United States Introduction Borden, William B. et al., Patterns and Intensity of Medical Therapy in Patients Undergoing Percutaneous Coronary Intervention, 305 JAMA 1882, 1886 (2011). Chassin, Mark R., & Jerod M. Loeb. The Ongoing Quality Improvement Journey: Next Stop, High Reliability, 30 Health Aff. 559, 563 (2011). Thomas  T.  Tsai et al., Contraindicated Medication Use in Dialysis Patients Undergoing Percutaneous Coronary Intervention, 302 JAMA 2458, 2463 (2009).

Improving the US Medical Malpractice System Abraham, Kenneth, Medical Liability Reform: A Conceptual Framework, 260 JAMA 68–72 (1988). Barringer, J., Paul David M. Studdert, Allen B. Kachalia, & Michelle M. Mello, Administrative Compensation of Medical Injuries: A Hardy Perennial Blooms Again, 33 J Health Politics, Policy L.275 (2008). Kachalia, Allen, Michelle M. Mello, Brahmajee K. Nallamothu, & David M. Studdert, Legal and Policy Interventions to Improve Patient Safety, 133 Circulation 661 (2016). Kinney, Eleanor, Learning from Experience, Malpractice Reforms in the 1990s: Past Disappointments, Future Success?, 20 J. Health Pol. Pol’y & L. 99 (1995). Mello, Michelle  M., & Allen Kachalia, Evaluation of Options for Medical Malpractice, Medicare Payment Advisory Commission (April 2010). For a full inventory of malpractice reform strategies.

Altering the Litigation Process David A. Hyman, Bernard Black, Charles Silver, & William M. Sage, Estimating the Effect of Damage Caps in Medical Malpractice Cases: Evidence From Texas, 1 J.  Legal Analysis 355 (2009). Kinney, Eleanor  D., Learning from Experience, Malpractice Reforms in the 1990s: Past Disappointments, Future Success? 20 J. Health Pol. Pol’y & L. 99 (1995).

Alternatives to Tort Litigation Nussbaum, Lydia, Trial and Error: Legislating ADR for Medical Malpractice Reform, 76 Maryland L. Rev. 247, 293 (2017).

Communication and Resolution Programs Disclosure of Adverse Events to Patients, VHA Handbook 1004.08 (October 2, 2012). Havighurst, Clark, & Laurence Tancredi, “Medical Adversity Insurance”—A No-Fault Approach to Medical Malpractice and Quality Assurance, 51 Milbank Mem. Fund Q. 125 (1973). Murtagh, Lindsey, Thomas  H.  Gallagher, Penny Andrew, & Michelle  M.  Mello, DisclosureAnd-Resolution Programs That Include Generous Compensation Offers May Prompt a Complex Patient Response, 31 Health Aff. 2681 (2012).

Medical Liability   601 Apology Programs Ayers, Ryan, 5 Ways the Healthcare Industry Could Use Big Data—and Why It’s Not, Dataconomy (Aug. 7, 2017), available at http://dataconomy.com/2017/08/5-ways-healthcare-big-data/. Conduent Business Servs. Healthcare Analytics: How Data Is Changing Everything 12 (2017) (eBook). Furrow, Barry  R., Searching for Adverse Events: Big Data and Beyond, 27 Ann Health L. 1 (2018). McMichael, Benjamin J., R. Lawrence Van Horn, & W. Kip Viscusi, “Sorry” Is Never Enough: How State Apology Laws Fail to Reduce Medical Malpractice Liability Risk, 71 Stan. L. Rev. 341 (2019).

Pure No-­Fault Compensation Systems Birth Injury Cases Bismark, Marie, & Ron Paterson, No-Fault Compensation in New Zealand: Harmonizing Injury Compensation, Provider Accountability, and Patient Safety, 25 Health Aff. 278 (2006). Mello, Michelle  M. et al., Administrative Compensation for Medical Injuries: Lessons from Three Foreign Systems, 14 Commonwealth Fund Pub. 1 (2011). Virginia Birth-Related Neurological Injury Compensation Program, https://www.vabirthinjury.com/; Florida Birth-Related Neurological Injury Compensation Association (NICA), https://www.nica.com/what-is-nica.html

Vaccine Injury Cases https://en.wikipedia.org/wiki/National_Vaccine_Injury_Compensation_Program; https://www. hrsa.gov/sites/default/files/hrsa/vaccine-compensation/data/monthly-stats-april-2019.pdf

Joint Conclusion Engstrom, Nora Freeman, Exit, Adversarialism, and the Stubborn Persistence of Tort, 6 J. Tort L. 75, 113 (2013).

Pa rt I V

R E GU L AT I NG T H E DE V E L OPM E N T A N D USE OF M E DIC A L T R E AT M E N T S

HUMAN E X PE R I M E N TAT ION A N D R E SE A RC H

chapter 29

I n troduction to R esea rch w ith H um a n Pa rticipa n ts Carl H. Coleman

Research with human participants is conducted for a variety of reasons, including developing drugs, medical devices, or other medical interventions; understanding human cognition and behavior; and evaluating the impact of public policy interventions. It can take place in a broad range of settings, ranging from universities and ­hospitals to biobanks and internet chatrooms. Funders, or “sponsors,” of research include governmental agencies, pharmaceutical and medical device companies, healthcare institutions, charitable foundations, and even wealthy individuals. Research with human participants can provide enormous social benefits, but it also raises significant ethical dilemmas. These dilemmas stem from a tension that is inherent in the nature of the activity: the goal is to generate knowledge for the potential benefit of persons in the future, but achieving this goal often requires exposing individuals in the present to the possibility of harm. This tension is particularly pronounced in clinical trials involving investigational drugs, devices or other medical interventions, where the risks of participation may be particularly significant. Although ordinary medical treatment can also involve risks, in the treatment context, those risks are offset by the potential for the individual patient to receive direct medical benefits. Indeed, the only purpose of medical treatment is to benefit the patient; if the risks to the patient are not outweighed by the potential direct benefits, providing the treatment would constitute medical malpractice. By contrast, with research, the purpose is not to provide direct benefits to the individuals enrolled in the study, but instead to generate knowledge that may eventually benefit other patients in the future.

608   Carl H. Coleman Research with human participants has taken place for centuries, but the modern regulatory apparatus did not emerge until after World War II, inspired in large part by the horrific medical experiments performed at the Nazi concentration camps. Following the war, doctors involved in these experiments were prosecuted for war crimes. The judges at these trials promulgated a series of ethical principles for research with human subjects, known as the Nuremberg Code, which were intended to prevent similar atrocities from occurring in the future. The Code’s first principle states, “the voluntary consent of the human subject is absolutely essential.” While the Nuremberg Code was an important affirmation of fundamental ethical principles, it did not address more practical questions about how research with human participants could ethically take place. Over the years, other, more detailed ethical guidelines have been promulgated to respond to this need. One of the earliest, and still most influential, was the World Medical Association’s Declaration of Helsinki, first issued in 1964, and most recently revised in 2013. The Declaration focuses specifically on medical research involving human participants, including research on identifiable human biospecimens and data. Among other issues, it addresses questions like the role of research ethics committees, the process of informed consent, and the appropriate use of placebo controls. Other important ethical guidelines have been issued by the Council for International Organizations of Medical Sciences,1 the World Health Organization,2 and numerous national and regional advisory groups. Although these guidance documents are not in themselves legally binding, they have been influential in the development of legislation, regulations, and other legal texts. The following two chapters provide a general overview of the legal framework surrounding research with human participants in two important centers of research, the United States and the European Union. In the United States, research with human participants is subject to a variety of federal and/or state legal regimes, depending on nature of the research and the source of the funding. At the federal level, the so-called Common Rule requires ethics committees, known as institutional review boards (IRBs), to review and approve most studies conducted or funded by one of 20 different federal agencies.3 Regulations promulgated by the Food and Drug Administration (FDA) require a similar process of IRB review for studies involving drugs, medical devices, or biological products.4 Other federal laws and policies related to research with human participants 1  The Council for International Organizations of Medical Sciences. “International Ethical Guidelines for Biomedical Research Involving Human Subjects. Accessed June 29, 2020. https://cioms.ch/publications/ product/international-ethical-guidelines-for-biomedical-research-involving-human-subjects-2/ 2  World Health Organization. “Standards and Operational Guidance for Ethics Review of HealthRelated Research with Human Participants,” 2011. http://www.who.int/ethics/ publications/9789241502948/en/ 3  US Department of Health & Human Services. “Federal Policy for the Protection of Human Subjects (‘Common Rule’).” Text, June 23, 2009. https://www.hhs.gov/ohrp/regulations-and-policy/ regulations/common-rule/index.html 4  Food and Drug Administration. “Regulations: Good Clinical Practice and Clinical Trials.” FDA, February 25, 2020. https://www.fda.gov/science-research/clinical-trials-and-human-subject-protection/ regulations-good-clinical-practice-and-clinical-trials

Introduction to Research with Human Participants   609 include regulations protecting the privacy of medical records,5 policies promoting the inclusion of women and racial minorities in research,6 and requirements related to clinical trials registration,7 reporting clinical trial results,8 and disclosing and managing research-related conflicts of interest.9 In addition to these federal laws, statutes or regulations in some states provide additional protection. Research with human participants can also implicate a variety of common law legal doctrines, including tort laws requiring compensation for negligently caused injuries. In the European Union, the oversight of research with human participants is similarly divided between EU-level law and the laws of individual Member States. As of this writing, clinical medical research is subject to the EU Clinical Trials Directive 2001/20/EC, which sets forth general standards for the oversight of clinical trials and requires Member States to implement those standards through national laws. The Directive will soon be replaced by the EU Clinical Trials Regulation 536/2014, which provides more specific rules that directly apply to all Member States. However, even after the Regulation comes into effect, some aspects of clinical trials oversight will continue to be subject to Member States’ discretion, including decisions about how to constitute the committees charged with conducting ethics reviews. In addition, neither the Clinical Trials Directive nor the Clinical Trial Regulation covers research with human participants that does not involve clinical trials, such as social science research or studies of human biospecimens. These studies are governed primarily by individual Member State laws, with the exception of issues related to the use of personal medical data, which are governed by the EU’s General Data Protection Regulation.10 Research in the European Union is also subject to a variety of human rights principles. Some of these principles derive from the EU’s own Charter of Fundamental Rights and are enforced by the European Court of Justice, which is part of the EU. Others stem from the European Convention on Human Rights, which is enforced by the Council of Europe’s European Court of Human Rights. As all EU Member States are also members 5  Office for Civil Rights, Department of Health & Human Services. “Privacy Rule General Overview.” November 5, 2015. https://www.hhs.gov/hipaa/for-professionals/privacy/guidance/ general-overview/index.html 6  Food and Drug Administration. Investigational New Drug Applications and New Drug Applications: Final Rule, Federal Register 63 (1998): 6854–6862; National Institutes of Health. “NIH Policy and Guidelines on The Inclusion of Women and Minorities as Subjects in Clinical Research.” 2001. https://grants.nih.gov/policy/inclusion/women-and-minorities/guidelines.htm 7  National Institutes of Health. “ClinicalTrials.gov Background–ClinicalTrials.gov.” Accessed June 23, 2020. https://clinicaltrials.gov/ct2/about-site/background 8  Clinical Trials Registration and Results Information Submission, 81 FR 64981 (2016). https://www. federalregister.gov/documents/2016/09/21/2016-22129/clinical-trials-registration-and-resultsinformation-submission 9  Food and Drug Administration. “Guidance for Clinical Investigators, Industry, and FDA Staff: Financial Disclosure by Clinical Investigators,” 2013. https://www.fda.gov/media/85293/download; “Responsibility of Applicants for Promoting Objectivity in Research for Which Public Health Service Funding Is Sought and Responsible Prospective Contractors: Final Rule,” Federal Register 76 (2011): 53256–53293; 42 C.F.R. Part 403. 10  Regulation (EU) 2016/679.

610   Carl H. Coleman of the Council of Europe, research in any EU country is subject to the jurisdiction of the Court of Human Rights. The legal regimes governing research with human participants in the United States and the European Union share many similarities. First, for the most part, they are grounded in the same substantive standards, including respect for autonomy, concern about protecting persons who are vulnerable, and attention to equity and justice. This is not surprising, as these are the core principles that underlie the major ethical guidelines on research that have been adopted at the international level. Second, under both US and EU law, most research with human participants is required to undergo prospective ethical oversight, by IRBs in the United States and by ethics committees in the European Union. Again, this requirement is consistent with international ethical guidelines. Third, both US and EU law subject clinical trials of drugs and devices to different regulations than other types of research. In the United States, the substantive standards in these regulations are largely similar to those found in regulations applicable to other types of federally funded research, but the drug and device regulations apply even if the studies are funded by private entities. In the European Union, there are more significant differences in the regulation of clinical trials and other types of research with human participants because only clinical trials are governed by overarching EU-wide law. Finally, in both the United States and European Union, some types of research with human participants are subject to general laws governing the privacy of personal data, which provide additional protections beyond those contained in research-specific laws. There are also important differences between the US and EU approaches to regulation. One difference is the scope of jurisdictional-wide regulations (i.e., federal regulations in the United States, and EU directives and regulations in the European Union). In both the United States and the European Union, clinical trials are subject to jurisdictional-wide ethics review requirements, but in the European Union, ethics review of non-clinical trials is primarily a matter of Member State law. In contrast, in the United States, most non-clinical trials involving human participants are also subject to federal ethics review requirements, either formally (because the research is conducted or supported by a Common Rule agency) or informally (because the research is conducted at an institution that has adopted the Common Rule as a model for its own institutional policies). For this reason, the European Union has a greater diversity of approaches to ethics review of research than exists in the United States. A second difference relates to the responsibilities entrusted to ethics review committees with respect to clinical trials. In the United States, IRBs are expected to make a comprehensive assessment of all aspects of a clinical trial, including the critical question of whether the risks to participants are reasonable in relation to the anticipated benefits. In contrast, under the soon-to-be-implemented EU Clinical Trials Regulation, the primary responsibility for reviewing risks and benefits is entrusted to the Competent Authority of the relevant Member State as part of its “scientific” review. Although ethics committees are permitted to consider risks and benefits if they choose to do so, their primary role is to address issues related to informed consent, compensation, recruitment, and data protection. Whether this division of authority will be workable in practice remains to be seen.

Introduction to Research with Human Participants   611 A related point has to do with the ability of ethics committees to insist on study-wide changes in multisite research. As of 2020, the United States has required single IRB review of most multisite studies, thereby eliminating the need for multiple reviews by IRBs at each research site. If the single IRB identifies ethical problems, it can require changes to be made to the study wherever it will be conducted. In contrast, once the EU Clinical Trials Regulation takes effect, the Competent Authority of a single Member State will be responsible for a centralized scientific review of multisite studies, but each involved Member State will be expected to conduct its own ethics review. If a Member State’s ethics review committee identifies ethical concerns, it can prevent the study from going forward in that particular Member State, but it has no authority over other study sites. Indeed, for ethical issues that do not also fall within the scope of the Competent Authority’s scientific review (e.g., the wording of informed consent forms), no entity will have the authority to insist on changes that will be effective at all research sites. Substantively, the most important difference between the US and EU regulatory frameworks relates to the availability of compensation for research-related injuries. In the United States, absent a specific contractual agreement or a finding of negligence, research sponsors have no obligation to provide compensation or medical treatment to participants who suffer injuries. By contrast, EU law requires all clinical trials to have systems to compensate participants for damages. The EU approach is consistent with international ethical guidelines; the US system is not. A final difference between the US and EU systems is the greater role of non–research specific bodies of law in the EU regulatory framework. One example is the EU’s General Data Protection Regulation, which comprehensively regulates all aspects of personal data privacy. In the United States, by contrast, the Health Insurance Portability and Accountability Act’s Privacy Rule covers only protected health information used or disclosed by specific “covered entities” (primarily, healthcare providers and insurers), leaving non-health personal data, or health data held by non-covered entities, with much less protection. A second example is the important role that human rights law plays in the European Union, both the EU’s own Charter of Fundamental Rights and the Council of Europe’s European Convention on Human Rights. In the United States, by contrast, courts rarely enforce international human rights principles. While some analogous principles exist in the US Constitution, the Constitution applies only to actions by federal or state governmental actors. It does not apply to privately conducted research, even if the research is supported by government funds. Overall, both the US and EU systems share the common goal of protecting the rights and well-being of research participants through requirements for prospective ethics oversight of studies. The systems are grounded in fundamentally similar ethical frameworks, with a few important differences, particularly with respect to the issue of compensation for injuries. The chapters that follow provide greater detail on how each system works.

chapter 30

R esea rch w ith H um a n Pa rticipa n ts i n the Eu ropea n U n ion David Shaw and David Townend

1 Introduction In the European Union, the driving force for legislation is the creation of the single ­market.1 This is qualified by the commitment to ensure public and individual health among the citizens of the EU, both through research and through ‘health in all policies’.2 These are not competing demands. Research with human participants is both a response to the urgent demand to advance the health of EU citizens through the efficiencies of the single market and to the commitment of the European Union to act in accordance with the shared social and cultural values of its Member States. Both give consumers and, perhaps more importantly for the practice of research, participants confidence in the credibility and integrity of the processes. Accordingly, the European Union is a major funder of research, including research with human participants,3 and the industries of research are a major part of the economies of the Member States that constitute the Union.4 One might expect, therefore, that EU legislation on research with human participants would be clear, creating a harmonised framework within the single market both for internal consistency and to enhance external competitiveness. As will be seen in this chapter, this is not yet the case.

1  Treaty on the Functioning of the European Union (TFEU), Article 114. 2  In particular, see TFEU, Article 168. 3  A Abbott and Q Schiermeier, ‘How European Scientists Will Spend €100 Billion’ (2019) Nature 569, 472–475. doi:10.1038/d41586-­019-­01566-­z 4  Directorate-­General for Research & Innovation (European Commission). ‘The Economic Rationale for Public R&I Funding and Its Impact’. Policy Brief Series. (2017) European Commission.

614   David Shaw and David Townend Each of the 27 Member States of the EU, as will be seen, operates with a large amount of independence in relation to the governance of research with human participants.5 When comparing this chapter with that of Coleman in this volume,6 many of the issues relating to key ethics and fundamental rights—for example, informed consent, diversity and inclusion—are not within the competence of the European Union. Some of those issues are the subject of Council of Europe7 discussions and conventions (e.g., the European Convention on Human Rights or the Oviedo Convention on Human Rights and Biomedicine). However, the granularity of these regulations is found in Member States’ law and practice. Setting out each of these approaches in detail is out with the scope of this chapter. Fortunately, however, there are many commonalities between these distinct approaches, and we will attempt to provide a detailed picture of the broad modalities by which medical and social sciences research is reviewed, approved, and conducted across the EU. Clinical medical research has a high level of direct EU regulation. Such clinical medical research involves patients, risks harm to them, and has, certainly since the late 1940s, been regarded as sensitive research. As will be seen, the European Union has created legislation to harmonise the governance of clinical trials across its Member States since the turn of the 21st century. However, as also will be seen, the necessity for interpreting legislative provisions that are susceptible to different understandings makes harmonisation difficult to achieve. Moreover, clinical trials represent only a small proportion of the medical research being conducted at any one time in the Union. Other research involving human participants, non-­clinical and social science research, is not covered by any specific legislation from the European Union, and its governance is more piecemeal at the Member State level. Still, there is some amount of consistency. All research with human participants is subject to human rights requirements, particularly relating to dignity and privacy.8 Medical research using personal data, be that data gathered for specific research or the

5  Norway, Iceland, and Lichtenstein, as members of the European Economic Area, are bound by EU law; Switzerland, as a former EEA member and a current member of the European Free Trade Association, has many bilateral treaties with the EU. 6  Coleman, ‘Research with Human Participants in the United States,’ this volume. 7  Whereas the European Union is the product of the economic harmonisation track to European recovery following World War II, the Council of Europe is the other track to that recovery, with its focus on human rights, democracy, and the rule of law. They are separate institutions. The Council of Europe exists in international law (compared with the Supranational State of the EU): i.e., the Members of the Council of Europe remain free to sign and ratify (or not to do so) any of the measures proposed through the Council of Europe. Hence, the Oviedo Convention has a relatively small number of signing and ratifying States. For the Council of Europe generally, see https://www.coe.int/en/web/ portal/home and for the Oviedo Convention see https://www.coe.int/t/dg3/healthbioethic/Activities/ Bioethics%20in%20CoE/ (both visited 15 November 2020). 8  See the Charter of Fundamental Rights of the European Union 2012/C 326/02 ELI: http://data. europa.eu/eli/treaty/char_2012/oj

Research with Human Participants in the European Union   615 secondary processing of already gathered personal data for medical purposes, falls under the purview of the EU General Data Protection Regulation (GDPR).9 This chapter concerns four key elements in the EU governance of research with human participants and is structured as follows. First, we consider the values that underpin the governance of research with human participants. Second, we describe and analyse the EU legislation on clinical trials. Third, we outline the regulation of ­non-­clinical research involving human participants in the European Union. Finally, we consider the protection of privacy through the GDPR.

2  Values Underpinning Research with Human Participants in the European Union The governance of research with human participants in the European Union, like all medical law, is driven primarily by fundamental values concerning the dignity of the individual, but also the individual within society. There is an international basis upon which the EU legislation is framed: the medical research–specific Declaration of Helsinki and the more general canon of human rights legislation. Both of these elements emerged after World War II and were a response to the experience of the atrocities witnessed during that period, in part in the name of ‘medical science’. During the Nuremberg Trials, the judges in the trial of the Nazi doctors produced a Code containing 10 safeguards for the protection of human participants in research.10 These 10 elements were modified and adopted by the World Medical Association in its Declaration of Helsinki.11 The broad themes of these standards are, inter alia, • respect for the autonomy of the participant (reflected in, for example, requirements for informed consent and a right to terminate participation in a study); • the particular protection of the vulnerable participant (children or those lacking capacity to act for themselves);

9  For completeness, it should be noted that there is legislation in Europe harmonising patents and copyrights protecting the outcomes of research. That is dealt with in the chapter ‘Regulating Medicines in the European Union’ by Mahalatchimy in this volume. Furthermore, related issues of privacy in relation to the operation of biobanks is also considered separately in the chapter by Laurie and Coleman. 10  United States of America v. Karl Brandt, et al. 21 November 1946–20 August 1947. 11  First adopted 1964, then subsequently revised eight times; current revision adopted at the 64th WMA General Assembly, Fortaleza, Brazil. https://www.wma.net/what-­we-­do/medical-­ethics/ declaration-­of-­helsinki/(last viewed 12 May 2020).

616   David Shaw and David Townend • the necessity for the study to have the capacity to yield a benefit, that benefit should outweigh the risk, and that a study should be terminated in the event of the risk of disability or death to a participant; • that researchers should be suitably qualified; • that the research is subject to ‘ethical standards that promote and ensure respect for all human subjects and protect their health and rights’,12 and • that the protocol and practice of the research should be scrutinised by an in­de­ pend­ent review/ethics board. The Declaration (2013) emphasises that the physician involved in medical research has the duty to ‘protect the life, health, dignity, integrity, right to self-­determination, privacy, and confidentiality of personal information of research subjects’ (paragraph 9) and to ‘consider the ethical, legal and regulatory norms and standards for research involving human subjects in their own countries as well as applicable international norms and standards’ (paragraph 10). Paragraph 9 of the Declaration echoes the broader Human Rights standards to which EU Member States are bound.13 As 27 of the 47 Member States of the Council of Europe, the EU Member States are bound by the European Convention on Human Rights and the judgments of the European Court of Human Rights. Furthermore, they are bound by the Charter of Fundamental Rights of the European Union and the judgments of the Court of Justice of the European Union. Taken together, these produce what might be described as a set of ‘headline’ values that apply to the European Union and its Member States and to individual citizens. However, arguably, they, of themselves, do not provide a sufficient degree of granularity to produce a harmonisation of values to guide researchers. This can be seen at two levels. First, at a conceptual level they are not clear. O’Neill argues that the range of ‘human rights’ is very much limited by the culture from which they were initially drafted and that they are conceived as ‘rights’ and not ‘duties’.14 There is a lack of clarity in the rights. The right to privacy, for example, bears a multitude of interpretations between disciplines.15 At one level, this can be seen in different interpretations of ‘anonymisation’ between different professions (and different researchers), whereby anonymisation, as a safeguard of privacy, can mean a range from complete removal of all individual-­identifying markers in data to coding where the individual can be identified by reintroducing the key to the dataset (what others might call ‘pseudonymisation’). At another level, one might compare the different legal definitions 12  Declaration of Helsinki (2013), Paragraph 7. 13  Although the Treaty of Lisbon (2009) committed the EU to accede to the European Convention on Human Rights, at the time of writing, negotiations and accession have still to be completed. Treaty of European Union, Article 6(2). 14  O O’Neill, A Question of Trust (Cambridge: Cambridge University Press 2010). 15  For a discussion of this problem see, for example, D Townend, ‘Privacy, Politeness and the Boundary Between Theory and Practice in Ethical Rationalism’ in P Capps and S Pattinson (eds) Ethical Rationalism and the Law. (Oxford: Hart Publishing 2017) 171–189.

Research with Human Participants in the European Union   617 of privacy with those in, for example, sociology.16 Laurie, inter alia, highlights two other levels that are useful to consider in our context.17 First, he points to the range of elements that are interpreted as privacy issues by the European Court of Human Rights, ranging from informational privacy issues, to the potential of privacy to include concern for well-­being. Second, he challenges the autonomy model of privacy, arguing that privacy is a matter of inclusion and ‘spatial’ privacy. This is not to suggest that clarity cannot be brought to bear on the concept,18 but the legislation and its interpretation in the Courts do not bring clarity to the concepts of the fundamental rights. Second, there is little consensus between Member States as to the granular, binding meaning of the rights. The European Convention’s ‘margin of appreciation’ for Member State authority prevails in this area. Indeed, cultural difference is enshrined in the Treaty of European Union; ­ethics are a matter of cultural diversity in the Union.19 Research ethics and research integrity are, however, extremely important to the functioning of the EU’s research funding programme.20 Accordingly, there are efforts to employ a de facto common regulatory approach to medical research funded by the European Union. Currently, Horizon 2020 grants are only made after protocols have been screened and assessed for compliance with ethical standards; funded projects can be subject to checks and audit. The basis and tenor of those evaluations are seen in the guidance given to applicants.21 The concerns of the review are perhaps more in practical aspects of research ethics and integrity than in any debate about the nature of ethical duties. The assessment concerns, inter alia, • conformity with EU prohibitions on research with human embryonic stem cells and more general legislation concerning the use of human cells or tissue; • ensuring that research with human participants will be subject to local ethics review and particularly in relation to safeguarding the vulnerable; 16  Warren and Brandeis give a starting point to privacy in the law (1890, Harv L Rev 4(5), 193–220). Allen gives a modern legal typology in ‘Genetic Privacy: Emerging Concepts and Values’ in MA Rothstein (ed.) Genetic Secrets: Protecting Privacy and Confidentiality in the Genetic Era (Yale University Press 1997) 31–60. Westin presents a perspective of privacy from a sociological background, most notably in Privacy and Freedom (London: Bodley Head 1970). For a general discussion of the range of interpretations, see J DeCew, ‘Privacy’ in Edward N Zalta (ed) The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), http://plato.stanford.edu/archives/fall2008/entries/privacy/ (last visited 15 November 2020). 17  G Laurie, Genetic Privacy: A Challenge to Medico-­Legal Norms (Cambridge: Cambridge University Press 2002). 18  See, e.g., the classification of four types of genetic privacy advanced by Anita Allen: informational privacy, decisional privacy, physical privacy, and proprietary privacy. AL Allen, ‘Genetic Privacy: Emerging Concepts and Values’ in MA Rothstein (ed) Genetic Secrets: Protecting Privacy and Confidentiality in the Genetic Era (Yale University Press 1997) 31–60 at page 33. 19  Treaty of European Union, Article 3. 20  https://ec.europa.eu/research/participants/docs/h2020-­funding-­guide/cross-­cutting-­issues/ ethics_en.htm (last viewed 12 May 2020). 21  European Commission (2019) Horizon 2020 Programme Guidance: How to complete your ethics self-­assessment. Version 6.1. https://ec.europa.eu/research/participants/data/ref/h2020/grants_manual/ hi/ethics/h2020_hi_ethics-­self-­assess_en.pdf (last viewed 12 May 2020).

618   David Shaw and David Townend • the protection of personal data in accordance with the GDPR; • the protection of animals used in the research;22 • ensuring conformity with EU standards on benefit-­sharing in developing and least developed economy countries; and • identifying any potential for dual use or other abuse of research results. These are all proper considerations; they either concern conformity with EU legal obligations or aim to ensure that local ethics evaluations are applied. This is in part because the assessment is undertaken before the grant is made, but as checks and audits become more widely used, it will be interesting to see how far substantive questions about the ethical choices that researchers have made are brought to the in- and post-­ grant reviews. Whereas the reviews outlined are significant in the EU framing of ethics in research, as are the positions and advice given by the European Group on Ethics in Science and New Technologies,23 the normative force of the governance framework struggles with the lack of ethical consensus. Principles such as those advanced by Beauchamp and Childress or the understanding of classical liberalism with the limit of autonomy being harm to others provide principles and opportunities for discussion. The different elements of the principles of biomedical ethics (respect for autonomy, justice, non-­ maleficence, and beneficence) are contested. For example, how far can this approach enable appeals to be made to the common good or the public interest in relation to the right to privacy? Such a limitation on the right is clearly available in Article 8(2) of the European Convention on Human Rights and in Article 52 of the Charter of Fundamental Rights of the EU, both of which allow for restrictions of the individual right when necessary to protect the public welfare. However, such an imposition on the autonomy of the individual is contested, such that the appeal is very difficult to make. Where, though, can an authoritative position be developed so that there is consistency in the rights of individual citizens across the European Union, particularly in respect of their participation in medical research? We return to this point later in the chapter when we discuss the secondary processing of personal data. The position of the European Union is not special in this regard; ethics is arguably about discourse and participation. However, the debate seems to operate more at the Member State level than at the EU level in relation to finding particular expressions of the broad values in research with human participants. Therefore, despite the ambition for both a single market and fundamental rights and freedoms across the EU, the first conclusion of this chapter must be that there is at least the potential for (or the presumption of) different interpretations of principles in different Member States.24 22  Through the application of Directive 2010/63/EU. 23  https://ec.europa.eu/info/research-­and-­innovation/strategy/support-­policy-­making/scientific-­ support-­eu-­policies/ege_en (last viewed 12 May 2020). 24  See, e.g., the range of sensitivities expressed in the Eurobarometers, https://ec.europa.eu/ COMMFrontOffice/publicopinion/index.cfm (last viewed 12 May 2020).

Research with Human Participants in the European Union   619

2.1  Regulation of Clinical Trials Completing this chapter in 2020, one might have thought that the section on the regulation of clinical trials would be an area of harmonisation—a success story for the EU. The twin drivers of the desire to create a more effective, harmonised governance structure for clinical trials of new pharmaceuticals and the wish to safeguard the rights and safety of the participants in those trials (and the integrity of the clinical research process for the broader community) brought clinical trials onto the EU legislative agenda at the end of the twentieth century. Prior to that, the regulation of clinical trials had been a matter for Member States. With the increase of multicentre, multijurisdictional trials, the need for a more streamlined, harmonised process to reduce time and economic costs in the development of research into new pharmaceuticals produced the legislative momentum to develop the Clinical Trials Directive 2001/20/EC. This responded to some of the needs for harmonisation. Over time, however, difficulties emerged about the differences in the Member States’ implementation of the Directive25 and the length of time the review process took under the Directive, and revision was required. The revision proc­ ess was difficult.26 The resulting Clinical Trials Regulation (536/2014) addressed the concerns over both harmonisation and timeliness, particularly with the streamlining effect of the Clinical Trials Information System (CTIS) (containing the EU Portal and Clinical Trials Database). Writing in 2020, six years after the Regulation was passed, the new system should be running. It is not. The CTIS is still under negotiation and development.27 Therefore, in this section of the chapter, it is necessary to discuss both the Directive and the Regulation.

2.1.1  Clinical Trials Directive 2001/20/EC The Directive remains in force until six months after the European Commission has issued a notice declaring the CTIS to be ‘fully functional’ following an independent review. At time of writing, the review is scheduled to start in December 2020. 25  Note that, in EU law, a Regulation has direct effect in Member States’ law. Directives ‘direct’ Member States to transpose the EU law into their domestic law. This appears a procedural point—the transposition must give effect to the Directive in Member States’ law—but in practice, it gives rise to differences in the interpretation of the law, thus reducing the harmonising effect of the EU measures. Within Regulations and Directives, there are Articles and Recitals. Articles contain the binding, ‘operative’ elements of the instrument. Recitals are not in themselves binding, but give context to the Articles. For example, in the GDPR, key issues for research are left ambiguous or unaddressed in the Articles—for example, the concept of ‘reasonableness’ in the remoteness of reidentification of individuals in de-­identified data is introduced in Recital 26; the possibility of including ‘broad’ consent within the meaning of ‘informed’ consent appears in Recital 33; context to the meaning of ‘compatible’ in relation to secondary processing of data is found in Recital 50. Unfortunately, as will be seen when studying those three examples, Recitals do not necessarily provide resolution of ambiguity in Articles. 26  For example, the first drafts of the proposed Regulation relegated ethics review to the discretion of the Member States. Ethics review was reintroduced into later versions of the proposals. 27  The development of the CTIS is the responsibility of the European Medicines Agency. The progress can be seen at https://www.ema.europa.eu/en/human-­regulatory/research-­development/ clinical-­trials/clinical-­trial-­regulation (last viewed 12 May 2020).

620   David Shaw and David Townend The Directive required Member States to implement in their national law a process for the evaluation and approval of clinical trials. The Directive required that the process should follow principles of human rights and dignity, citing as an example an expression of such rights the Declaration of Helsinki (1996).28 It enshrined the principle that those participants who could give their informed consent to participate in a clinical trial had the right to do so and that the vulnerable who could not (children and those without the capacity to act for themselves) should be protected by a ‘legal representative’.29 The Directive required the European Commission to adopt good clinical practice standards, which it did through Directive 2005/28/EC. The Directive required Member States to ensure that they had a review process, pri­ ma­rily through Ethics Committees, to assess the following under Article 6(3): (a) ‘the relevance of the clinical trial and the trial design; (b) ‘whether the evaluation of the anticipated benefits and risks as required under Article 3(2)(a) is satisfactory and whether the conclusions are justified; (c) ‘the protocol; (d) ‘the suitability of the investigator and supporting staff; (e) ‘the investigator’s brochure; (f) ‘the quality of the facilities; (g) ‘the adequacy and completeness of the written information to be given and the procedure to be followed for the purpose of obtaining informed consent and the justification for the research on persons incapable of giving informed consent as regards the specific restrictions laid down in Article 3; (h) ‘provision for indemnity or compensation in the event of injury or death attributable to a clinical trial; (i) ‘any insurance or indemnity to cover the liability of the investigator and sponsor; (j) ‘the amounts and, where appropriate, the arrangements for rewarding or compensating investigators and trial subjects and the relevant aspects of any agreement between the sponsor and the site; (k) ‘the arrangements for the recruitment of subjects.’ The protocol and investigator brochure are used by the Ethics Committee to inform the evaluation of the risks and benefits of the proposed study and to verify that the information provided to participants is accurate. The arrangements for recruitment are checked to ensure that they are appropriate and not potentially distressing or disruptive for potential participants. It can be argued that the harmonising effect of the Directive is limited in relation to the interpretation of the criteria just indicated and the constitution of the Ethics Committee. Consider, for example, the Article 3 protections required for a clinical 28  Directive 2001/20/EC, Recital 2. 29  Directive 2001/20/EC, Articles 3, 4, and 5, and Recitals 3, 4, and 5.

Research with Human Participants in the European Union   621 trial participant, some of which must be evaluated by the ethics committee under its  Article 6(3)(b) duty to assess the anticipated benefits and risks of a proposed study. Article 3(2) issues that must be considered and addressed before a trial can commence are (a) ‘the foreseeable risks and inconveniences have been weighed against the anticipated benefit for the individual trial subject and other present and future patients. A clinical trial may be initiated only if the Ethics Committee and/or the competent authority comes to the conclusion that the anticipated therapeutic and public health benefits justify the risks and may be continued only if compliance with this requirement is permanently monitored; . . . (c) ‘the rights of the subject to physical and mental integrity, to privacy and to the protection of the data concerning him in accordance with Directive 95/46/EC are safeguarded. . . .’ 

As can be seen, these are closely related to the Declaration of Helsinki requirements (as are the provisions for the reporting of adverse effects in the Directive). Equally, the underpinning concepts require interpretation, and different moral cultures can produce different decisions. This, of itself, is not problematic. However, in the case of multicentre, multijurisdiction trials, applications are made to the Ethics Committee in each ­jurisdiction, and each Committee considers the merits of the application independently, returning its own assessment without discussion with the other Committees c­ onsidering the same protocol. It is then for the applicant to reconcile the (perhaps fundamentally) different opinions of the Committees. This is a barrier to harmonisation of the Clinical Trials process in the Directive and a driver for reform. The Directive raised concerns in addition to those about harmonisation. For example, it is immediately apparent under Article 6(3)(a) and (b) that the Ethics Committee is charged with the evaluation of the science as part of the ethical assessment.30 On the one hand, this is a legitimate question for an ethics committee: bad science is necessarily unethical. However, it does require that the Committee must have competence to make an assessment of the science. A major criticism of Ethics Committees across the EU (and indeed, internationally) is the lack of common expectations about the required collective competencies of Committees. The Directive does not alleviate this criticism; there is no indication of the minimum requirements for the constitution of a competent Ethics Committee in the Directive. Furthermore, the qualification of the Committee to assess the societal benefit of a proposal might also be questioned: the Committee is not democratically appointed or representative of the broader community’s interests, and

30  The Member State’s Competent Authority can address issues in Article 6(3)(h), (i), and (j).

622   David Shaw and David Townend lay involvement through Committee membership is arguably problematic as a method of connecting Ethics Committees to their communities.31 Another important concern lies in the timeframes that the Directive creates for the processing of an application. Sixty days is the timeframe for processing standard applications (under Article 6(5)), with a stopping of the clock should the Committee make a (single) request for further information from the applicant (Article 6(6)). Those applications relating to ‘medicinal products for gene therapy or somatic cell therapy or medicinal products containing genetically modified organisms’ can take a further 30 days, with the possibility of a further 90 days extension of the time limit (Article 6(7)). These times were seen as too generous and were a driver for reform of the process. A final driver for reform is the lack of transparency in the information surrounding clinical trials that leads to duplication of failed trials; this poses unnecessary and unethical risks to future participants and constitutes a large waste of resources. At this point it is worth reiterating that, at present, the Directive is still in force and in operation.

2.1.2  Clinical Trials Regulation 536/2014 The Clinical Trials Regulation seeks to address the criticisms raised against the Directive. Immediately, as a Regulation, it goes some way to address divergence between Member States’ implementation of the law: the Clinical Trials Regulation has direct effect in Member States’ law and variation in transposition is to some extent limited. A Regulation does not automatically remove difference; Member States are given some explicit choices within the Regulation (notably, still how they constitute Ethics Committees), and the language of the Regulation has to be interpreted in the Member States. However, in many respects, the Regulation is a tighter set of procedures and expectations than the Directive. One immediate difference is that scientific efficacy questions and ethics process questions are separated (into Part I and Part II of the assessment report). The former are to be addressed primarily, although not exclusively, by the Competent Authority (the national body responsible for medicines authorization) and the latter by the Ethics Committee ‘in accordance with the law of the Member State concerned’.32 Parts I and II of the assessment have a different procedural structure. Part I, the scientific part of the review conducted by the Competent Authority, is led by a ‘reporting Member State’. In a trial involving only one country, this will be the Competent Authority in that country (thus designated the ‘reporting member state’); in a multicountry trial, one country’s Competent Authority will perform the scientific review. Thus in a multicountry 31  See, e.g., D Townend and ES Dove, ‘Approaching Ethics Review Equivalency Through Natural Justice and a “Sounding Board” Model for Research Ethics Committees’ (2017) Med Law 36(1), 61–86. For an authoritative academic study of research ethics committees, see ES Dove, Regulatory Stewardship of Health Research: Navigating Participant Protection and Research Promotion (Edward Elgar Publishing Ltd. 2020). doi:10.4337/9781788975353 (last visited 12 May 2020). For more practical issues concerning research ethics, research integrity, and ethics committees in the EU, see EUREC (a network of European Union Ethics Committees) at http://www.eurecnet.org/index.html (last visited 12 May 2020) and the EU-­funded research project ENERI at http://eneri.eu (last visited 12 May 2020). 32  Clinical Trials Regulation 536/2014, Article 4.

Research with Human Participants in the European Union   623 trial, in Part I, after the application is made, one committee will have responsibility to drive the review by checking the submitted documents, drafting a first opinion, gathering the opinions of the other committees, and presenting a final report. In contrast, all involved countries are expected to conduct an independent Part II ethics review of each proposal to ensure that the ethical standards in a given country are followed. Normally, the Part II review will be conducted by an ethics committee, but some countries (including Denmark) have elected to unite ethical review with scientific review in one Competent Authority. The implication of this overall approach is that science and scientific review are universal while ethics is a more local affair.33 The logic of this implication may be questioned, but the main problem with the Regulation is that scientific and ethical review are to a large extent interlinked, and the Regulation itself tacitly admits this by acknowledging that several aspects of applications can (but note, not ‘should’) be considered by the ethical review as well as the scientific review. To aid the reader, we now provide a systematic summary of what should be considered only by the Competent Authority in the Part I scientific review, what should be considered only by the ethics committee performing Part II of the review, and what aspects can be considered by both reviewing bodies. Article 6 of the Regulation sets out the elements that should be considered in Part I of the assessment; these are the elements that must be considered in the scientific review and can be considered in the ethical review. These include (among other, more technical requirements and elaborations of the following): ( i) – – (ii)

‘The anticipated therapeutic and public health benefit’; ‘the relevance of the clinical trial’; ‘the reliability and robustness of the data generated in the clinical trial’; and, ‘the risks and inconveniences for the subject’.34

In Article 7, the Regulation sets out the elements that must be considered in Part II, the ethical review, and should not normally be considered in the scientific review. The important aspects here are informed consent, compensation, recruitment, and data protection, or specifically, Article 7(1)(a)–(d): (a)  ‘compliance with the requirements for informed consent as set out in Chapter V; (b) ‘compliance of the arrangements for rewarding or compensating subjects with the requirements set out in Chapter V and investigators; (c) ‘compliance of the arrangements for recruitment of subjects with the requirements set out in Chapter V; (d) ‘compliance with Directive 95/46/EC’ [the Data Protection Directive, now superseded by the GDPR 2016/679]

33  This perception of ethics as has a ‘local’ variation is seen in paragraph 23 of the Declaration of Helsinki, although, again, how in practice an Ethics Committee connects to its local community (if at all) is not clear. 34  Article 6(1)(b).

624   David Shaw and David Townend While it is commendable that ethics committees are at least permitted to consider aspects mentioned in Article 6, this should not obscure the fact that (as stated earlier) contemplating the potential risks and inconveniences for the subject is the core role of the typical ethics committee. Compliance with the requirements for informed consent can only be checked by comparing what is stated on participant information sheets with the actual risks and burdens of the study; reviewing these aspects is thus essential to the role of the ethics committee. Similarly, informed consent materials normally make claims about the potential benefit of a trial to society, so the ethics committee must also review those aspects. Similarly, it is hardly helpful for Article 6 to stipulate that the designated competent authority must consider inconveniences to the subject but not the arrangements for recruitment of subjects. Furthermore, the division of labour set out in the Regulation strips ethics committees of the other central aspect of their normal role: considering the potential benefits of a trial to society and thus its relevance, and examining whether the design of a study will allow it to fulfil its aim. As we have argued elsewhere: ‘the experts on an ethics committee might realise that similar research has already been conducted, making a repeat study irrelevant; or that the risks posed are not outweighed by prospective benefit to society even if the study is novel; or that the study is statistically underpowered and thus cannot confirm its hypothesis.’35 Taken together, these serious limitations in the design of the Regulation suggest that, although compliance with it is necessary for the legal conduct of clinical trials, it is an ethically weak piece of legislation that may itself expose participants to increased risk of harm. There are further problems relating to the criticisms of the Directive. The Regulation reduces the timeframe for reviews to 45 days. This is possible for the Part I scientific review as science has a perceived universal language, removing the need for local assessment. Moreover, the scientific review has a structure where one Competent Authority leads the review, makes a preliminary assessment of the documentation provided by the applicants, and leads the drafting of the Part I report. Part II ethical reviews, on the other hand, remain separate and independent reviews and reports, as in the Directive. With the Regulation’s stricter time limits and fewer opportunities for delay and extension, it remains to be seen how feasible they will be in practice. The final problem relates in part to the criticism of the lack of transparency in the clinical trials process. The Regulation creates the CTIS to manage the Portal, where the applications are managed, and the database, where a record of the applications more  generally is built. The development of the CTIS has been problematic both ­technically and conceptually. The delay is perhaps not surprising: the very concept of transparency in such an intellectual property–dependent industry, and one that also deals with large amounts of personal data, is extremely difficult, as are the technical challenges in producing the right data science environment. 35  D Shaw and D Townend, ‘Division and Discord in the Clinical Trials Regulation’ (2016 22 July) J Med Ethics, 1–4. doi:10.1136/medethics-­2016-­103422.

Research with Human Participants in the European Union   625

3  Medical Research (Non-­c linical Trials) and Research Ethics Committees Non-­clinical trial medical research is often reviewed by the research ethics committees (RECs) of university medical faculties. Such committees can review social science–type research conducted by researchers in the medical faculty (see next section), but also certain bioscience and health research involving human participants. For example, biology researchers might conduct basic physiological tests on participants that do not pass the threshold to count as human subject research. One of the most important tasks for the chairs of university faculty RECs is to ensure that applications are reviewed by the right committee. Although the law does not tend to mandate review of such studies, it does stipulate that certain categories of study must be reviewed by a competent committee. These studies normally involve human tissue or clinical trials and must be reviewed by a ‘medical’ committee; for example, a National Health Service ethics committee in the United Kingdom or a Medical Ethics Committee accredited for Human Subjects Act type research in the Netherlands. Often coordination is required between university RECs and hospital RECs in order to reach agreement on who should review a particular study. There are normally rules that are supposed to determine which projects should be reviewed by which type of committee, but in practice they are often unclear. Certainly until recently, any study in the UK that recruited National Health Service (NHS) patients had to be reviewed by an NHS REC, regardless of whether it is a clinical study or an interview project. However, some studies involving physical or chemical interventions to improve fitness in healthy volunteers, although they might ‘sound’ medical, can be reviewed and approved by university RECs. Coordination between chairs of medical and non-­medical RECs is also important to avoid wasteful duplication of effort involved in dual review. It is also important to note that dual review is sometimes necessary when students or staff are travelling abroad to conduct research: here, it would be insensitive to only have review from the researchers’ home institution when there are also local ethics mechanisms, but it could also be unwise to rely entirely on those local mechanisms, which may not be so well-­developed (depending on the country).

4  Social Science Research Social science research is not specifically regulated by law in many EU states, though use of data in such research is subject to the GDPR (see later discussion). This label covers a wide variety of research, from psychology studies (which can involve deception) to questionnaire surveys of people’s health behaviours, to interviews with the public and stakeholders on a wide variety of academic subjects. While data must be handled in accordance with the law (see later discussion), regulation of such research was until

626   David Shaw and David Townend fairly recently conducted in a rather ad hoc manner, with some research institutions requiring ethical review of this type of research and others not requiring it. This has slowly begun to change, with some countries moving faster than others. The United Kingdom has made perhaps the greatest strides in this direction, with most universities now having research ethics review structures, for example through both faculty and central RECs tasked with reviewing social science research. The workload of these committees has grown substantially, not least because they are now often expected to review not only staff proposals, but also master’s degree projects and sometimes bachelor-­level research.36 The move towards performing ethical review of social science research is not pri­ma­ rily motivated by sudden recognition of the need to ensure that research is conducted ethically. In fact, it has had three main drivers. First, universities and other research institutions are risk-­averse, and having mechanisms in place for ethical approval increases accountability and reduces the risk of potential legal action by an aggrieved research participant. Second, many funders now require ethical approval for any project that they are supporting. Third, many journals will not publish research without ethical approval, or at least a letter from a committee saying that a project was exempt from ethical review. There have also been research misconduct cases in all jurisdictions, and these have also been a consideration in developing research ethics and research integrity agendas.37 These factors mean that REC review of social science research will soon become necessary even in countries where it has not traditionally been conducted. The scandals surrounding Facebook and Cambridge Analytica, where user data was shared with a private company without the users’ consent, have also led to increasing awareness of the need for thorough ethical review of social science research that involves use of online and social media data. Such studies raise a host of issues that are unfamiliar to university RECs and may not be covered by existing guidelines. For all these reasons, many university researchers now have to seek ethical approval for any human participant research they are conducting. Ethics review structures vary, but in many universities faculty ethics committees are responsible for research conducted in those faculties. Some institutions have meetings, at regular intervals, where applications are discussed; others conduct review remotely via email, with occasional meetings to discuss difficult applications. When applications raise issues that cannot be settled at the faculty level, they are normally referred to the central university ethics committee, which may have its own membership or be constituted of the chairs of the different faculty committees.

36  Research in the context of undergraduate education poses interesting questions, inter alia: the research is primarily for training purposes and not for publication and contribution to the broader scientific literature. It is often the first work that a student will undertake, and therefore, how far should such work involve risks to the participants? And what is the role of the supervisor in safeguarding the interests of participants in such research? 37  See, e.g., J Crocker, ‘The Road to Fraud Starts with a Single Step’ (2011) Nature, 479, 151. doi:10.1038/479151a

Research with Human Participants in the European Union   627

5  General Data Protection Regulation So far, this chapter has predominantly considered research with human participants where the involvement is direct. The potential harms concern physical interventions and psychological harms relating to the in-­person relationship of the participant to the research. Another aspect of research with human participants where EU legislation is well-­established is the processing of personal data. Before addressing that legislation, it is worth noting that whereas the human rights to privacy (e.g., in Article 8 of the European Convention on Human Rights) acknowledge that the right is held in balance with the interests of other rights holders (as mentioned earlier), the Declaration of Helsinki seems to place research using personal data on a level with direct intervention participation studies. There is only a limited acknowledgement in paragraph 32 that privacy is not an absolute right, and processing personal data in the public interest might be acceptable in ‘exceptional situations where consent would be impossible or impracticable to obtain for such research’. Clearly, a breach of privacy can cause an individual both consequential loss (which can be compensated) and, possibly, more fundamental harm (which is more difficult to compensate and repair). However, in relation to the secondary processing of already gathered data in a situation of confidentiality, one wonders if the interests in play are as high as would be the case in participation in a clinical trial for a new pharmaceutical product or an interview study dealing with deep emotions about child loss. Indeed, there could be an argument that where an individual has received treatment for a condition or is even simply a member of the public who has yet to become a patient, being part of society requires that their data are shared, with stringent safeguards, for the development of the treatments that are generally seen as public goods. At the least, some might argue, there is a need to set the default to requiring those who do not want to participate in such a system to opt out, as a matter of good, active citizenship. This is not to advocate these positions. However, it is again to ask where such debates can be had in society, as a presumption of an opt-­in paramountcy of autonomy might not be the desirable ethical starting point for much of the new medical research methodologies. International data protection dates back to the late 1970s and early 1980s. The first European legislation was advanced by the Council of Europe in 1981; the first EU legislation was Directive 95/46/EC. The current EU legislation is the GDPR 2016/679.38 It is possible to see the same heritage running through all these iterations. It should first be noted that the legislation is not specifically about the processing of personal data in research: it is legislation to cover all processing of personal data. Whereas the legislation rather subsumes or overtakes much national legislation on medical data, it does not

38  Hoffman and Herveg discuss the application of the GDPR to personal data in the clinical healthcare setting, in this volume.

628   David Shaw and David Townend generally replace medical (or other) confidentiality.39 Indeed, processing, under Article 5(1)(a) of the GDPR must be fair, transparent, and lawful, which leaves open conformity with Member States’ requirements for confidentiality. The other principles upon which data protection is built (all from Article 5) are that data should be processed for a defined purpose (or purposes) and not further processed for incompatible purposes. Data should be limited, in both amount and the length of time kept, to that required for the processing.40 Data should also be held securely. The GDPR creates a number of further elements that could be considered in the same vein as the Article 5 principles. Article 89 requires that ‘appropriate safeguards’, including pseudonymisation or anonymisation of the data, should be in place for processing for research purposes. These particularly apply where Member States choose to derogate from a data subject’s rights.41 Likewise, Article 25 introduces the concept of ‘data protection by design and by default’, requiring that any enterprise envisaging the proc­ essing of personal data should build in compatibility with the safeguards for and rights of data subjects found in the GDPR from the outset. As an aside, perhaps these requirements might encourage a similar concept of ‘ethics by design’ in research planning. The second element of the GDPR that should be understood is that data subjects, those to whom the data relate directly or indirectly (Article 4(1)), have a number of rights: • to know what data are held about them, • to ‘portability’ of the data they supplied (Article 20),42 39  For example, in Dutch Law, under the Wet Geneeskundige Behandelovereenkomst (WGBO)—the part of the Civil Code relating to medical treatment (patient rights)—Article 7: 457 concerns the right to confidentiality over information disclosed within the medical professional–patient relationship. It operates, for example, to allow other practitioners necessary for the care of the patient to be consulted. Article 7:459 requires the respect of the patient’s privacy. These provisions, and their development in precedent, remain in place alongside the requirements of the GDPR. These show congruence between the GDPR and confidentiality law. The more traditional medical interpretation of access to one’s health data is seen in Article 7: 448(3), where the practitioner (after consultation with another practitioner) might withhold data from the individual where there is a risk of causing serious harm to the patient by sharing the data. This is not reflected in, for example, the data subject’s right to access data under Article 15(3) of the GDPR. 40  This poses an interesting research requirement: how long are data required in an identifiable form to ensure that research integrity concerns can be met and so that incidental findings can be acted upon to the advantage of the participant? Data protection perhaps leans towards ‘data minimisation’, whereas a more sophisticated position might be that the research requires that more data be held in a secure way for longer to ensure the broader purposes of the processing. 41  Article 89 also applies where the presumption of compatibility operates in Article 5(1)(b) and the appeal to impossibility and disproportionate effort in Article 14(5)(b). 42  Article 20 “portability” gives the data subject certain rights to the return of their data from the data controller in a machine readable format that can then be passed to others for processing. The most interesting part is that the data that are covered are those that the data subject supplied to the data controller, and not any personal data that the data controller has about the data subject. Whilst the data controller has a right to be told what data the controller has relating to the individual, this portability right seems to be deliberately limited in the legislation. Whether it is so limited in practice is a different question. However, it raises very interesting questions of coverage in the medical research field.

Research with Human Participants in the European Union   629 • to correct errors in those data, • to have the data erased and ‘be forgotten’ (Article 17), and • to object to the processing of data that relate to them. Article 89(1) provides that Member States can formally derogate from these rights in relation to ‘archiving purposes in the public interest, scientific or historical research purposes or statistical purposes, [. . .] subject to appropriate safeguards, in ac­cord­ ance with this Regulation, for the rights and freedoms of the data subject.’ Article 89(1) indicates that the safeguards may include pseudonymisation or anonymisation of the data.43 A legal basis for processing is necessary to allow processing under the GDPR. Lawful processing is defined, for ‘general personal data,’ in Article 6. There are essentially four different bases: • informed consent or a contract that requires the processing; • the vital interests of the data subject; • the interests of the data controller (the person who determines the purpose and processing of the data) if they do not damage the interests of the data subject; and • processing in the public interest. Lawful processing of ‘sensitive personal data’ (which includes medical data) is defined in Article 9. These criteria are broader than the Article 6 criteria. Particularly for research with human participants, the bases are: • informed consent; • the vital interests of the data subject; • the public interest; • research, subject to Article 89. It is traditionally thought that the construction of Article 9 (that the processing of sensitive personal data is prohibited, but the prohibition can be lifted by an appeal to one of the conditions) meant that processing also needs a legal basis under Article 6. This is challenged by German lawyers in particular, suggesting that Article 9 creates a ius speciale (replacing the requirement for the general rule, i.e., Article 6). This is important, given the availability of processing for the purposes of research in Article 9(2)(j).44 However, this interpretation is contested. The final element required is the information provisions under Articles 13 and 14. Data Controllers must provide the Data Subject with, inter alia, their contact details, details about the purposes of processing, and the legal basis for the processing. Where 43  With the exception of the right to portability and the right to object to processing, unless it is undertaken on the legal basis of the public interest. 44  Note that, under the GDPR, ‘research’ includes applied research (Recital 159).

630   David Shaw and David Townend the data are gathered directly from the Data Subject, then this information must be supplied; where indirectly, it must be supplied unless it is impossible or requires a disproportionate effort. These provisions do not require that all processing be by informed consent. Rather, the information provisions enable the data subject to act upon their rights and, for example, ask the Data Controller which personal data are held about them (where the right is available, see earlier discussion). Having outlined the basic structure of the GDPR, there are a number of interesting areas that require consideration in relation to research with human participants. Where personal data are gathered directly from a data subject for stand-­alone research, the requirements are relatively straightforward. It is likely that the Data Controller will rely on informed consent and will inform the data subject directly about the processing. Difficulties arise where already gathered data—perhaps from a large research dataset, or from clinical or other medical records, or perhaps shopping data from loyalty cards— can be linked for an unforeseen research purpose. As modern medical research becomes more data-­driven and uses more artificial intelligence tools, these become more pressing issues. The first question is, what is the legal basis for the initial processing? If it is informed consent, what is within the scope of that original consent? Article 4(11) defines consent in a narrow way: it is interpreted to mean that a specific defined purpose and processing must be told to the data subject. There is an acknowledgement in Recital 33 that research requires a broader interpretation of consent, but how far this will be translated into broad consent will be a matter of practice; the interplay between RECs and data protection officers will be crucial in this practical regard. Informed consent is not the only legal basis. Indeed, researchers could argue that the work is in the public interest (although how far that will be available is also unclear; whether a Member State must first define the local meaning of public interest is not clear) or in the interests of the data controller where they do not damage the interests of the data subject (although that could be difficult to argue unless the supervisory authorities and RECs accept that safeguards [e.g., enforceable confidentiality contracts and data security and management measures] mitigate the danger to the data subject’s interests). Therefore, it could be that it will be difficult to set up the initial legal basis to enable further processing for unforeseen purposes. If this is the case, it is possible to argue that the processing is compatible with the original processing of the data. This is available under Articles 5(1)(b) and 6(4), and the case is strengthened by Recital 50. However, even where the conceptual basis for such a claim is accepted, compatibility for research is a rebuttable presumption, and the original consent (for example) has to be such that it does not limit the possibility for future research. If the proposed secondary processing is not compatible with the original processing, then arguably it is a new processing of the data. This would allow for a new legal basis to be identified for the proposed processing—perhaps the public interest, or the legitimate interests of the data controller, as discussed earlier, given the impracticality of seeking to re-­consent a large cohort (even if it is possible). However, there is some reluctance to

Research with Human Participants in the European Union   631 allow a change of legal basis during processing. It remains to be seen how far the idea that the processing is essentially new in this scenario is accepted by the supervisory authority or Ethics Committee. And none of these steps addresses the question of how a data controller who gathered data from a data subject directly has to act in relation to informing them of a further processing (given that there could be no appeal to impracticability or impossibility under Article 13). What this shows is that modern (medical) research methodologies pose considerable but not insurmountable difficulties for researchers navigating the GDPR. Under Article 40 of the GDPR, there is the possibility for sector-­specific Codes of Conduct to be approved by the European Commission. Clearly, there is a need for such a code in relation to medical research.45

6 Conclusion This chapter has been about the hard regulation that the European Union brings to research in its Member States. What we have not addressed is the influence that the EU institutions, particularly the European Commission, can exert on research through ‘soft’ power. That soft influence is considerable. The Union is a major funder of research, and health research features prominently in the priorities over the different periods of the Research Framework Programmes.46 Within the funding structure, there is, arguably, limited space for researcher-­led, blue-­skies research. The European Research Council (ERC) offers opportunities for such projects to nurture researchers at the early and established stages of their careers. The competition is adjudicated by a panel of in­de­ pend­ent experts and is highly competitive. Other grants within the long-­standing EU research funding programmes have been Commission-­led—with priorities in calls being generated by the relevant Directorates General. This is not surprising. The European Union, as most other governments, wishes to see returns on its investments in real technologies that answer identified social problems. How far this ‘soft’ influence on research, in the European Union or in any other jurisdiction, is useful for the development of science or democracy is a question for another book. Certainly, funding competition is a strong regulator of academic freedom, and successful science, as ever, requires skills in the lobby almost as much as it does in the lab. The European Union has aimed to harmonise the regulation of research across its Member States to some extent, but, as this chapter has illustrated, it has only been partially successful in doing so. Social science research is not governed by any specific legislation other than the GDPR, and this necessarily leaves a void of either individual 45  JE Litton, “We Must Urgently Clarify Data-­Sharing Rules’ (2017) Nature 541, 437 (26 January 2017) doi:10.1038/541437a. 46  The ‘Horizon Europe’ research budget is in the order of €80.9 billion for the period of 2021–2027. At the time of writing, the strategy and allocation of funds between different priorities is still being negotiated.

632   David Shaw and David Townend legislation and practice, or even no coverage at all; medical research other than clinical trials is also not covered by any specific EU laws other than the GDPR, but here the established operation of the Declaration of Helsinki is likely to produce some national legislation. In the domain of clinical trials there has been legislation in the form of the Clinical Trials Regulation, but its implementation has been delayed, thus requiring ongoing reliance on major aspects of the Clinical Trials Directive. Even once fully implemented, the Regulation will enable divergence due to performance of ‘Part II ethical review’ by individual Member States. The Regulation is further weakened by its removal from ethics committees of many of the duties traditionally associated with them. How can the EU do a better job of regulating research in the future? We suggest that it should do so by focusing on harmonisation not only at the practical level of implementation but also at the level of citizen engagement and empowerment. Legislation such as the Clinical Trials Regulation will not succeed if it is detached from the interests of stakeholders. Important ideas such as transparency in the design and reporting of clinical trials and the public interest in data protection have so far run aground on entrenched commercial interests and the anchor of individual autonomy. Only with engagement of a greater diversity of stakeholders in a wider societal discussion about what we want and expect from medical research—and are willing to accept in terms of tradeoffs to achieve that important goal—can EU research regulation function effectively.

chapter 31

R esea rch w ith H um a n Pa rticipa n ts i n the U n ited States Carl H. Coleman

1 Introduction Since the 1970s, most research with human participants in the United States has been subject to federal regulations requiring prospective ethical oversight by entities known as institutional review boards (IRBs). IRBs review studies to determine whether the risks have been minimized and are reasonable in relation to anticipated benefits; whether the selection of participants is equitable; whether requirements for obtaining and documenting participants’ informed consent have been satisfied; and, when appropriate, whether the research plan contains adequate provisions for data and safety monitoring and privacy protection. Research that is subject to the federal regulations may not begin until IRB approval has been obtained. This chapter begins by examining key aspects of the federal regulations governing IRB review of research with human participants, including the type of activities that fall under the IRB’s jurisdiction, how IRBs are organized, and some of the key substantive standards that IRBs apply. It then looks at additional regulatory standards that apply to studies involving particular populations, including pregnant women and fetuses, prisoners, and children. Finally, the chapter examines several other bodies of law related to research with human participants, including policies governing the inclusion of women and racial minorities in clinical trials; legal principles governing compensation for injuries to research participants; and requirements for registering clinical trials, reporting trial results, and disclosing research-­related conflicts of interest.

634   Carl H. Coleman

2  Overview of the Regulatory Framework The existing regulatory framework governing research with human participants emerged in the early 1970s, with the enactment of the National Research Act of 1974. The Act was passed in the wake of revelations about a variety of unethical research practices, the most prominent of which was the US Public Health Service’s Tuskegee Syphilis Study, in which poor African American men with syphilis were deliberately prevented from obtaining treatment for their disease so that scientists could study the “natural course” of the illness.1 The Act required an IRB for any entity receiving federal funds to conduct biomedical or behavioral research involving human participants. It also established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. In 1979, the Commission issued the Belmont Report, which identified three core ethical principles applicable to research with human participants: respect for persons, defined primarily as the protection of individual autonomy; beneficence, which involves the maximization or benefits and minimization of risks; and justice, which requires the equitable distribution of the benefits and burdens of research.2 In 1981, the Department of Health and Human Services (DHHS) and the Food and Drug Administration (FDA) updated their research regulations to incorporate the framework articulated in the Belmont Report. There are now two primary sets of federal regulations governing research with human participants. The first, known as the Common Rule, is a set of regulations jointly followed by 20 federal agencies, including DHHS, the home of the National Institutes of Health (NIH), the largest public funder of biomedical research in the world.3 The Common Rule applies to any research that is conducted or supported by any of the agencies that have adopted the Rule. For example, if the NIH awards a grant to a university to conduct a clinical trial, the trial must adhere to the Common Rule requirements. Until 2019, institutions receiving federal research funding could voluntarily commit to applying the Common Rule to all research conducted at the institution, even studies not supported by federal funds. If they did this, the federal Office for Human Research Protections (OHRP) would enforce the Common Rule at that institution regardless of a study’s funding source. However, following revisions to the Common Rule that took effect in 2019, OHRP now enforces the Common Rule only with respect to federally funded research.

1  Jones, James H, Bad Blood: The Tuskegee Syphilis Experiment (The Free Press 1981). 2  US Department of Health and Human Services. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research (US Department of Health and Human Services 1979). 3  The Common Rule appears in multiple places in the US regulations, as every agency that has adopted the Rule publishes its own version. The version published by DHHS, which is the most frequently cited, appears at 45 C.F.R. Part 46.

Research with Human Participants in the United States   635 In order for an activity to be covered by the Common Rule, it must satisfy the regulatory definition of “research” with “human subjects.” The term “research” is defined as “a systematic investigation . . . designed to develop or contribute to generalizable knowledge.”4 The term “generalizable” is not defined in the regulations, but it is generally understood to mean information that can be applied beyond the specific individuals or groups from which it was derived. For example, if an investigator gives one group of patients a new medication and a comparable group of patients an old one, it would be possible to derive statistically meaningful information about the impact of the different drugs on each group’s outcomes. This information could then be generalized to other patients with similar characteristics, which would bring the activity into the regulatory definition of research. The Common Rule gives several examples of activities that do not meet the definition of research. These include certain scholarly and journalistic activities, public health surveillance activities, criminal justice activities, and authorized operational activities in support of national security missions. Examples of these categories include oral histories or journalistic accounts of specific individuals, outbreak investigations conducted by a public health agency, the collection of biospecimens as part of a criminal investigation, and interviews conducted as part of an authorized intelligence operation. The common theme of these examples is that they generate information intended to apply only to the individuals or groups from whom the information was obtained. Even if a particular activity satisfies the regulatory definition of research, it will not be subject to the Common Rule unless it also implicates the definition of human subject, which refers to “a living individual about whom an investigator . . . obtains (1) data through intervention or interaction with the individual; or (2) identifiable private information.”5 Importantly, either intervention/interaction with an individual or the use of an individual’s identifiable information is sufficient to trigger the regulatory definition. Thus, research involving the review of medical records containing individually identifiable patient information would implicate “human subjects,” even if the researchers never interact with the patients at all. The Common Rule’s exclusive focus on activities that meet the definition of research with human subjects can be criticized as both over- and underinclusive. It is overinclusive in the sense that not all activities designed to produce generalizable knowledge raise the kind of ethical issues that warrant the administrative burdens associated with IRB review. To some extent, this problem is dealt with by the Rule’s “exemptions,” which identify types of human subject research that do not need to adhere to the Rule’s ­requirements.6 Examples include the use of anonymous surveys or questionnaires, the review of publicly available data or pathological specimens, and other similarly l­ ow-­risk, noncontroversial activities. However, not all low-­risk research activities can be fit into one of the exemption categories. Moreover, at some research institutions, the process of applying for an exemption can itself be fairly burdensome.

4  45 C.F.R. § 102(l).

5  45 C.F.R. § 102(e).

6  45 C.F.R. § 104.

636   Carl H. Coleman The Common Rule is also underinclusive to the extent that it does not require any ethics oversight of activities not designed to develop or contribute to generalizable knowledge, even if those activities, like research, will expose some individuals to risks or burdens for the potential benefit of others. For example, when public health officials engage in surveillance activities designed to identify the prevalence of infection in a community, persons who are already infected may not experience any personal benefits, but they are exposed to risks of stigmatization and discrimination if their identities are revealed. Nonetheless, because this activity does not meet the Common Rule’s definition of research, it may proceed without any ethics review at all. In addition to the Common Rule, the FDA has its own regulations governing “clinical investigations” involving FDA-­regulated products intended for human use (i.e., drugs, medical devices, and biologics).7 Unlike the Common Rule, the applicability of the FDA regulations does not depend on the source of funding for the research; any research involving an FDA-­regulated product is covered, even if the research is supported entirely by private funds. The FDA regulations are largely similar to the Common Rule, with a few relatively minor exceptions.8 Taken together, the Common Rule and the FDA human subject protection regulations cover the majority of research involving human participants in the United States, but some significant gaps remain. In particular, privately funded studies not involving FDA-­regulated products are not covered by either set of regulations. A few states have adopted laws that require IRB oversight of studies not otherwise covered by the federal research regulations, but most states have not. While most universities, as well as some private companies, have internal policies requiring ethics review of non-­covered research, in most states these review processes are entirely optional and not subject to any legally binding standards. The lack of mandatory ethics oversight of these activities remains a significant gap in the US regulatory framework.

3  The IRB’s Place in the Research Oversight System Oversight of research with human participants in the United States occurs primarily through IRBs. Under the both the Common Rule and the FDA human subject protection regulations, an IRB must have at least five members, at least one of whom is not affiliated with the institution. At least one member’s “primary concerns” must be in scientific areas, and at least one member’s primary concerns must be in nonscientific areas. 7  21 C.F.R. Parts 50 and 56. 8  Food and Drug Administration, “Comparison of FDA and HHS Human Subject Protection Regulations.” FDA, November 3, 2018. https://www.fda.gov/science-­research/good-­clinical-­practice-­ educational-­materials/comparison-­fda-­and-­hhs-­human-­subject-­protection-­regulations.

Research with Human Participants in the United States   637 An IRB may not consist entirely of members of one profession, and “every nondiscriminatory effort” must be made to ensure that the IRB is not comprised entirely of men or entirely of women.9 When IRBs were first created, the goal was to create a locally based system of ethics review, in which protocols would be subject to peer review by members of the researchers’ own institutions. Today, while many IRBs are still based within research institutions, an increasing number of free-­standing IRBs also exist, many of which are for-­profit companies. These IRBs contract with research sponsors to conduct independent ethics review of particular studies. Independent, commercial IRBs have both advantages and drawbacks. On one hand, they often have greater resources than traditional, institutionally based IRBs, including large professional staffs that can support the members’ deliberations. In addition, reviewers at commercial IRBs are not subject to the interpersonal pressures or conflicts of interest that can arise when colleagues review one another’s proposals. On the other hand, the fact that commercial IRBs are paid by the sponsors of the research creates an inherent conflict of interest, as boards fearful of losing future business may be reluctant to issue overly critical reviews. In recent years, there has been a movement toward greater use of central IRBs for review of studies involving multiple research sites. A central IRB may be an in­de­pend­ ent, commercial IRB; an IRB located at one of the participating research sites; or an IRB associated with a government funding agency, such as the National Cancer Institute. Since 2020, the Common Rule has required all US institutions engaged in cooperative research to rely on the approval of a single IRB for any portion of the research that is conducted in the United States.10 In addition to ensuring consistency across trial sites and avoiding unnecessary duplication of effort, central IRBs can provide greater protection to research participants because they can condition their approval on the researchers’ agreement to make study-­ wide changes to protocols. In contrast, local IRBs reviewing multisite studies typically have few options other than approving or rejecting the protocol as written because allowing site-­by-­site variations in protocols would undermine the scientific validity of the study. The primary challenge with central IRB review is the ability to take into account unique local conditions at particular trial sites, such as a larger populations of persons who do not speak English or who lack health insurance, or differences in applicable state law requirements. In some cases, supplemental reviews by local IRBs may be required to ensure that centrally approved studies adequately take into account a particular community’s needs.

9  45 C.F.R. § 46.107.

10  45 C.F.R. § 114(b).

638   Carl H. Coleman

4  Key Ethical Considerations for IRB Review The IRB’s responsibilities include assessing the risks and potential benefits of proposed studies and reviewing the researchers’ plans for obtaining informed consent, recruiting and paying participants, and ensuring that the confidentiality of research data is adequately protected. This section reviews the regulatory standards applicable to each of these tasks. In addition, it discusses other relevant sources of law, including anti-­fraud statutes and medical privacy rules.

4.1  Risk-­Benefit Assessment Under both the Common Rule and the FDA human subject protection regulations, the IRB must determine that the risks to participants have been “minimized” and that they “are reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected to result.”11 The regulations provide that IRBs “should consider only those risks and benefits that may result from the research (as distinguished from risks and benefits of therapies subjects would receive even if not participating in the research).” Thus, observational research ­involving the evaluation of interventions the participants are already receiving should normally be characterized as low risk, even if the interventions themselves involve a significant potential for harm. The types of risks and benefits the IRB considers will vary depending on the nature of the study. For example, in a clinical trial of a new drug, the IRB’s focus will primarily be potential physical consequences, whether positive or negative. By contrast, in a social science study designed to identify predictors of mental health issues among immigrant adolescents, relevant risks would include nonphysical considerations, such as legal or social repercussions that could result from the unauthorized disclosure of participants’ personal information. The concept of risk in the US regulations focuses entirely on the potential negative consequences of a study for the individual participants. This is unfortunate because it ignores the significant impacts that research can have on other individuals and communities. For example, interventions used in infectious disease research might alter the likelihood that participants will transmit the disease to third parties. Genetics studies that focus on discrete population groups may generate information that is stigmatizing to group members who did not consent to participate in the research. While IRBs are free to consider these third-­party risks, nothing in the regulations requires them to do so. 11  45 C.F.R. § 46.111(a)(1)–(2); 21 C.F.R. § 56.111(a)(1)–(2).

Research with Human Participants in the United States   639 One challenge in assessing the risks and potential benefits of research is that the concepts of risk and benefit are inherently subjective. The US regulations leave the identifying, quantifying, and weighing of risks and benefits to the discretion of individual IRBs, with no guidance on how they should go about making these judgments. It is therefore not surprising that studies show considerable variation in how IRBs quantify and weigh various risk-­benefit scenarios.

4.2  Informed Consent Another critical IRB responsibility is ensuring that the researchers either have designed an adequate process for obtaining informed consent to participate in the study or that the study satisfies specific regulatory conditions for waiving the usual informed consent requirements. In studies for which informed consent is required, the regulations allow for consent to be provided by either the participant or the participant’s “legally au­thor­ ized representative,” defined as “an individual or judicial or other body authorized under applicable law to consent on behalf of a prospective subject to the subject’s participation in the procedure(s) involved in the research.”12 In most cases, the “applicable law” refers to state laws governing guardianships and surrogate decision-­making for individuals who lack decision-­making capacity. Subject to limited exceptions, the process of informed consent must be documented in writing. Since 2019, the Common Rule has required informed consent forms to begin with “a concise and focused presentation of the key information that is most likely to assist a prospective subject or legally authorized representative in understanding the reasons why one might or might not want to participate in research.”13 The purpose of this provision is to give prospective participants something more useful than lengthy lists of every conceivable risk that a study might involve. The revisions that took effect in 2019 also included a requirement to post consent forms for federally funded studies on a publicly available website.14 In most situations, federal regulations specify nine basic elements of informed consent, including the following: • information about the nature and purposes of the research; • the reasonably foreseeable risks or discomforts; • any benefits that may reasonably be expected from the research; • appropriate alternative procedures or courses of treatment, if any; • a statement describing the extent, if any, to which confidentiality of identifying records will be maintained; • for studies involving more than minimal risk, an explanation as to whether compensation and/or medical treatments will be made available in cases of injury; • an explanation of whom to contact with questions or in case of injury; 12  45 C.F.R. § 102(i).

13  45 C.F.R. § 116(a)(5)(i).

14  45 C.F.R. § 116(h).

640   Carl H. Coleman • a statement that participation is voluntary and may be discontinued at any time and that refusal to participate will involve no penalty or loss of benefits to which the participant is otherwise entitled; and, • in studies involving the collection of identifiable private information or identifiable biospecimens, either a statement that the information or biospecimens might be anonymized and used in research without additional informed consent or a statement that the information and biospecimens will not be used in future research even if identifiers are removed.15 The Common Rule also identifies several additional elements of informed consent that must be provided whenever applicable. These additional elements include information about • potentially unforeseeable risks; • circumstances under which participation in the study may be terminated by the investigator; • any additional costs that might be incurred as a result of participating in the study; • the consequences of deciding to withdraw from the research; • a statement that participants will be informed of any significant new findings developed during the study that might influence their willingness to continue participation; • the approximate number of study participants; • a statement that participants’ biospecimens may be used for commercial profit, and whether or not participants will share in this profit; • a statement regarding whether, and if so when, clinically relevant research results will be disclosed to participants; and • for research involving biospecimens, whether the research will or might include whole-­genome sequencing.16 Since 2019, the Common Rule has provided an alternative approach to informed consent, known as “broad consent,” which can be used in studies involving the storage, maintenance, and secondary research use of identifiable information or identifiable ­biospecimens.17 This option enables participants or their representatives to consent to broad categories of future research with their identifiable information or biospecimens before the specific details of those future studies are known. The advantage of broad consent is that it enables researchers to conduct large-­scale studies involving identifiable data or biospecimens obtained in previous studies without having to go back to each of the participants to obtain their informed consent again. However, keeping track of which data or biospecimens were provided with broad consent and which were not is administratively complex and probably not possible without sophisticated and interoperable medical records systems. In addition, asking participants to provide broad 15  45 C.F.R. § 116(b).

16  45 C.F.R. § 116(c).

17  45 C.F.R. § 116(d).

Research with Human Participants in the United States   641 consent to future research comes with a risk to the researcher: if an individual is asked to provide broad consent but refuses to do so, the researcher no longer has the option of seeking a waiver of informed consent in future studies involving that person’s identifiable data or biospecimens. In light of these limitations, it is unclear whether the option of broad consent will be frequently used. The Common Rule allows IRBs to waive or alter the requirement to obtain informed consent in two circumstances. First, informed consent may be waived or altered in studies related to public benefit or service programs if the studies are conducted by or subject to the approval of state or local governments.18 This type of waiver might be used in a study designed to compare different procedures for obtaining benefits or services under a particular program. Second, informed consent may be waived or altered in research involving “minimal risk” if (1) the research could not practicably be carried out without the requested waiver or alteration; (2) for studies using private information or identifiable biospecimens, the research could not practicably be carried out without using such information or biospecimens in an identifiable format; (3) the waiver or alteration will not adversely affect the rights and welfare of the participants; and (4) whenever appropriate, the participants or their legally authorized representatives will be provided with additional pertinent information after participation.19 This type of waiver is often used in epidemiological studies involving the review of large quantities of medical records or other identifiable information, where it would be difficult, if not impossible, to track down and obtain the permission of everyone whose records are being reviewed. It is also used in so-­called deception research, in which providing false information to participants about the nature or purpose of the study is an inherent part of the research design (e.g., telling participants that they will be working cooperatively with other participants, when in fact they will be interacting with and evaluated by study personnel). The Common Rule’s waiver provisions do not currently appear in the FDA human subject protection regulations, but, in 2019, the FDA issued a proposed rule to adopt the waiver applicable to certain minimal risk studies.20 In addition to the Common Rule waivers, the FDA regulations allow for waivers of informed consent in some types of emergency research. This waiver is available if the IRB determines that participating in the research “holds out the prospect of direct benefit to subjects” because (1) they are facing a life-­threatening situation that necessitates intervention, (2) available treatments are “unproven or unsatisfactory,” (3) preclinical evidence shows that the experimental intervention is promising, and (4) the risks associated with the research “are reasonable in relation to what is known about the medical condition of the potential class of subjects, the risks and benefits of standard therapy, if any, and what is known about the risks and benefits of the proposed intervention or activity.” Waivers of informed consent in emergency research also require public 18  45 C.F.R. § 116(e). 19  45 C.F.R. § 116(f). 20  Food and Drug Administration, “Institutional Review Board Waiver or Alteration of Informed Consent for Minimal Risk Clinical Investigations,” November 15, 2018. https://www.federalregister.gov/ documents/2018/11/15/2018-­24822/institutional-­review-­board-­waiver-­or-­alteration-­of-­informed-­ consent-­for-­minimal-­risk-­clinical.

642   Carl H. Coleman disclosure and consultation with the community in which the study will be conducted and from which participants will be drawn.21 Although the emergency research waivers are not part of the Common Rule, DHHS has adopted identical waiver provisions pursuant to its authority to waive particular Common Rule provisions.22 A 2018 review of research conducted in the 20 years since the emergency waiver provisions were adopted identified 28 studies in which these waivers were used; the most common areas studied were cardiac arrest, hemorrhagic shock, and traumatic brain injury.23

4.3  Recruiting and Paying Participants Neither the Common Rule nor the FDA regulations explicitly addresses the process of recruiting research participants. However, FDA guidance provides that direct advertising for participants constitutes the start of the informed consent process and is therefore subject to IRB review.24 In addition, oversight of the recruitment process can be seen as part of IRBs’ responsibility to ensure the equitable selection of research participants.25 Federal research regulations are silent on the permissibility of paying research participants. In a 2018 guidance document, the FDA stated that payment is generally an “acceptable practice” but that IRBs should consider whether payments that exceed reimbursement for expenses “could present an undue influence, thus interfering with the potential subjects’ ability to give voluntary informed consent.”26 However, it is questionable whether the concern about undue inducement is relevant in most types of research. As Ezekiel Emanuel has argued, “ethical concerns about undue inducement arise only when the imprudence might eventuate in substantial risk of serious discomfort or harms,”27 and studies with such a high level of risk are unlikely to pass IRB review. In some cases, the use of financial incentives to encourage the recruitment of research participants can raise issues under federal or state anti-­fraud laws. For example, a physician who refers study participants to research conducted at a hospital with which the

21  21 C.F.R. § 50.24. 22  Ellis, Gary B, and Melody H Lin, “Informed Consent Requirements in Emergency Research (OPRR Letter, 1996).: https://www.hhs.gov/ohrp/regulations-­and-­policy/guidance/emergency-­ research-­informed-­consent-­requirements/index.html#:~:text=The%20waiver%20authorization%20 applies%20to,authorized%20person%20to%20represent%20them. 23  Klein, L, J Moore, and M Biros, “A 20-­year Review: The Use of Exception From Informed Consent and Waiver of Informed Consent in Emergency Research” (2018) Acad Emerg Med 25, 1169–1177. 24  Food and Drug Administration, “Recruiting Study Subjects,” US Food and Drug Administration, April 18, 2019. https://www.fda.gov/regulatory-­information/search-­fda-­guidance-­documents/ recruiting-­study-­subjects. 25  45 C.F.R. § 46.111(a)(3). 26  Food and Drug Administration, “Payment and Reimbursement to Research Subjects,” US Food and Drug Administration, April 18, 2019. https://www.fda.gov/regulatory-­information/search-­fda-­ guidance-­documents/payment-­and-­reimbursement-­research-­subjects. 27  Emanual, Ezekiel J, “Ending Concerns About Undue Inducement Independent Article” (2004) J L Med Ethics 32(1), 100–105.

Research with Human Participants in the United States   643 physician has a financial relationship may violate the federal Stark Law.28 In addition, payments made to investigators based on the number of participants recruited may violate the federal Anti-­Kickback Statute if any part of the study will be paid for through the Medicare or Medicaid programs.29 Even payments to research participants (including waivers of otherwise applicable copayment obligations) could constitute Anti-­Kickback violations if “one purpose” of those payments is to encourage enrollment in study for which Medicare or Medicare will cover some of the costs.30

4.4  Confidentiality and Data Protection Federal research regulations protect confidentiality in two ways. First, they require IRBs to review proposed studies to ensure that “there are adequate provisions in place to protect the privacy of subjects and to maintain the confidentiality of data.”31 These provisions might include measures such as requirements to store identifiable data in locked cabinets or password-­protected files or to destroy research data after a specified period of time. Second, the Common Rule (but not the FDA regulations) exempts from coverage many types of studies in which identifying information will not be recorded. For example, research that is limited to educational tests, survey or interview procedures, or observation of public behavior is exempt if “the information obtained is recorded by the investigator in such a manner that the identity of the human subjects cannot readily be ascertained, directly or through identifiers linked to the subjects.”32 The availability of this exemption creates an incentive for researchers to design studies so that identifying information is recorded only when absolutely necessary. Researchers can provide additional confidentiality protections to research participants by obtaining a certificate of confidentiality from DHHS.33 A certificate of confidentiality prevents the researcher from being compelled to disclose identifying information about participants in any federal, state, or local civil, criminal, administrative, legislative, or other proceeding. Certificates are available for any study, regardless of funding source, in which the disclosure of identifying information could have adverse consequences for participants or damage their financial standing, employability, insurability, or reputation.

28  Rusczek, Julie M, and Andrew P Rusczek, “Fraud and Abuse in Clinical Research: Three Case Studies” (2010) ABA Health eSource 6(10), 1–5. 29  McAllister, Debbie K, “Gimme Shelter: Anti-­Kickback Safe Harbors and Clinical Trial Agreements” (2008) J Clin Res Best Pract 4(5), 1–11. 30  Persad, Govind, “Paying Patients: Legal and Ethical Dimensions” (2018) Yale J L Techn 20, 177–233. 31  45 C.F.R. § 46.111(a)(7). 32  45 C.F.R. § 104(d)(2)(i). 33  Office for Human Research Protections, “Certificates of Confidentiality - Privacy Protection for Research Subjects: OHRP Guidance (2003).” February 25, 2016. https://www.hhs.gov/ohrp/regulations-­ and-­policy/guidance/certificates-­of-­confidentiality/index.html.

644   Carl H. Coleman Confidentiality and data protection in research are also governed by the Privacy Rule promulgated under the Health Insurance Portability and Accountability Act (HIPAA), which applies to, among other entities, any healthcare provider that transmits information in electronic form.34 Under the Privacy Rule, most uses of protected health information (PHI) require the individual’s prior, written authorization. The definition of PHI is sufficiently broad to encompass virtually all identifiable medical records likely to be of interest to researchers. In most situations, researchers subject to the HIPAA Privacy Rule will need to obtain the participant’s written authorization for the use or disclosure of their PHI. In some cases, however, obtaining a HIPAA authorization from every participant may not be a realistic option, as in studies involving retrospective review of large sets of medical records. In these situations, it is possible to obtain a waiver or alteration of the authorization requirement, either from the IRB or the institution’s privacy board. Alterations or waivers are available if the use or disclosure of PHI involves no more than minimal risk to the subjects, the research could not practicably be conducted without the waiver or alteration, and the research could not practicably be conducted without access to and use of the PHI.35 The HIPAA Privacy Rule contains provisions allowing researchers to review PHI for  purposes “preparatory to research,” such as preparing a research protocol. The researcher must file a request for access to the information with an office designated by the institution, representing that the request is sought solely to review PHI preparatory to research; that no PHI will be removed from the institution in the course of the review; and that the PHI for which use or access is sought is necessary for the research purposes.36 Researchers also have the option under HIPAA of creating a “limited data set,” which consists of data that have had most, but not all, of their identifying information removed. Unlike fully de-­identified information, a limited data set may include dates related to the individual (such as date of birth or dates of hospital admission and discharge), geographic information other than street address (such as zip code), and a unique identifying code. The regulations permit researchers to use or disclose limited data sets without complying with the usual requirements for the use or disclosure of PHI. However, the option is available only if the researcher has signed a data use agreement with the institution, which places limits on who may receive the information and how it may be used.37 The potential consequences of HIPAA violations include criminal penalties and civil fines. The criminal penalties, which include imprisonment, apply to anyone who knowingly obtains PHI in violation of the law, even for those who were not aware that their 34  Office for Civil Rights, Department of Health & Human Services, “Privacy Rule General Overview.” November 5, 2015. https://www.hhs.gov/hipaa/for-­professionals/privacy/guidance/ general-­overview/index.html. 35  Office for Civil Rights, Department of Health & Human Services, “Research.” May 7, 2008. https:// www.hhs.gov/hipaa/for-­professionals/special-­topics/research/index.html. 36 Id. 37 Id.

Research with Human Participants in the United States   645 actions were unlawful.38 HIPAA does not authorize individuals to sue for damages based on a violation of the regulatory requirements. However, some courts have allowed plaintiffs to rely on HIPAA to establish a minimum standard of care in cases alleging breach of confidentiality under state-­law tort theories.39

5  Additional Protections for Special Populations In addition to the Common Rule, separate subparts of the DHHS’s regulations on research with human participants provide additional protections for studies involving pregnant women and fetuses (Subpart B), prisoners (Subpart C), and children (Subpart D).40 The FDA, as well as the Department of Education, have adopted Subpart D for research that they regulate involving children as participants. However, because these additional subparts are not part of the Common Rule, most other federal agencies have not adopted them. A common feature of the additional subparts is that, unlike the Common Rule, they impose upper limits on the levels of risk to which research participants can be exposed, which vary depending on the nature of the research. For example, under Subpart B, research that does not hold out a prospect of direct benefit to either the pregnant woman or the fetus may not be approved if it poses more than a minimal risk to the fetus.41 Under Subpart C, prisoners may participate only in specified categories of research, including research “on the possible causes, effects, and processes of incarceration, and of criminal behavior,” research on “prisons as institutional structures or of prisoners as incarcerated persons,” research on “conditions particularly affecting prisoners as a class,” and research on practices that “have the intent and reasonable probability of improving the health or well-­being of the subject.” The first two of these categories of permitted research may not involve more than minimal risk or inconvenience.42 Finally, with limited exceptions, Subpart D allows children to be exposed to research involving more than minimal risks only if the study offers the prospect of direct benefit to the child or if the study involves only a “minor increase over minimal risk” and is “likely to yield generalizable knowledge about the subject’s disorder or condition.”43 Some of the additional subparts also include special requirements for obtaining informed consent to research. For example, Subpart D replaces the usual informed consent rules with a requirement to obtain permission from one (or in some cases, both) 38  United States v. Zhou, 678 F.3d 1110 (9th Cir. 2012). 39  E.g., Byrne v. Avery Center for Obstetrics & Gynecology, P.C., 102 A.3d 32 (Conn. 2014). 40  Office for Human Research Protections, “Federal Policy for the Protection of Human Subjects (‘Common Rule’).” June 23, 2009. https://www.hhs.gov/ohrp/regulations-­and-­policy/regulations/ common-­rule/index.html. 41  45 C.F.R. § 46.204(b). 42  45 C.F.R. § 46.307(a). 43  45 C.F.R. § 46.405–46.406.

646   Carl H. Coleman of the child’s parents or guardians.44 In addition, with limited exceptions, assent must be solicited from the children if they are capable of providing it.45 Under Subpart B, research with pregnant women that “holds out the prospect of direct benefit solely to the fetus” is permitted only if informed consent is obtained from both the pregnant woman and the father of the fetus, unless he is unavailable or unable to provide consent or the pregnancy resulted from rape or incest.46 This requirement has been subject to extensive criticism, but efforts to eliminate it toward the end of the Clinton administration were abandoned after the Bush administration took office.47 Finally, the additional subparts provide a variety of heightened procedural protections. For example, under Subpart B, IRBs reviewing research with prisoners must contain a prisoner or prisoner representative as a member, and certain studies may be approved only after consultation with appropriate experts and publication of notice in the Federal Register.48 Similarly, under Subpart D, research with children involving more than a “minor increase over minimal risk” and no prospect of direct benefit to the children may be approved only after consultation with a panel of experts and an opportunity for public review and comment.49 Some state laws provide additional limitations on research with some of these populations. For example, although Subpart B does not prohibit research involving fetal material after delivery, some states have laws that prohibit research with fetal tissue under any circumstances.50 Court decisions in some states may impose further limitations. For example, in the case of Grimes v. Kennedy Krieger Institute, the Maryland Supreme Court held that researchers may not expose children to more than minimal risks in “nontherapeutic” studies.51 As a result, even though Subpart D allows a “minor increase over minimal risk” in studies “likely to yield generalizable knowledge about the subject’s disorder or condition,” researchers in Maryland are required to follow the stricter state law standard. While the additional subparts provide important protections for specific areas of research, their effectiveness is limited by the fact that they apply only to research conducted or funded by a few federal agencies. In addition, the subparts do not address other population groups that could also benefit from special protections in research. One notable gap is the absence of specific protections for research involving adults who lack decision-­making capacity. As noted earlier, the federal regulations defer to state law in this area, but only a few states have enacted laws that specifically address research involving decisionally incapacitated adults.

44  45 C.F.R. § 46.408(b).

45  45 C.F.R. § 46.408(a).

46  45 C.F.R. § 46.204(e).

47  “Bush Replaces Clinton Rule, Requires Paternal Nod for Some Studies on Fetuses” (November 14, 2001) Health Care Daily Rep (BNA). 48  45 C.F.R. § 46.304(b), 46.306(iii)–(iv). 49  45 C.F.R. § 46.406. 50  National Conference of State Legislatures, “Embryonic and Fetal Research Laws.” Accessed June 23, 2020. https://www.ncsl.org/research/health/embryonic-­and-­fetal-­research-­laws.aspx. 51  782 A.2d 807 (Md. 2001).

Research with Human Participants in the United States   647

6  Inclusion of Women and Racial Minorities in Research Policies concerning the participation of women and members of racial minority groups in research have evolved considerably over the past several decades. In the 1970s, FDA policy explicitly prohibited the inclusion of women of childbearing potential in most Phase 1 and early Phase 2 clinical trials, allowing exceptions only in the case of drugs intended for life-­saving or life-­prolonging treatment. The term “women of childbearing potential” was defined broadly to include all premenopausal women physiologically capable of becoming pregnant, regardless of whether they used contraception or were even sexually active with male partners. The policy allowed women of childbearing potential to be included in later phase studies only if adequate information on the effectiveness of the drug being studied existed and animal teratogenicity and the female part of animal fertility studies had already been completed.52 The FDA’s policy was designed to guard against the danger that experimental drugs might have harmful consequences for the offspring of women who became pregnant. However, this policy was counterproductive because manufacturers could still get approval for drugs that doctors would be free to prescribe to women who might become pregnant while taking them. The exclusion of women from clinical trials simply meant that the consequences of those drugs for women and their children would never be studied. In addition, the FDA’s policy also denied women the right to make their own assessments of the risks of research participation as compared to the potential benefits. In 1993, the FDA withdrew its restrictions on women’s participation in clinical trials. It noted its expectation that “in accordance with good medical practice, appropriate precautions against becoming pregnant and exposing a fetus to a potentially dangerous agent during the course of study will be taken by women.” It also emphasized that efforts should be made to ensure that women participating in clinical trials are not pregnant at the time the trial begins.53 The FDA issued additional guidelines in 1998 that require sponsors to document the safety and effectiveness of new drugs for demographic subgroups, including women of childbearing potential. Moreover, it stated that it might withhold final approval of a new drug if the approval application did not analyze the clinical trial data according to gender.54 The revised FDA policies do not, however, expressly require the inclusion of women in particular studies.

52  Food and Drug Administration, “Gender Studies in Product Development: Historical Overview.” February 28, 2020. https://www.fda.gov/science-­research/womens-­health-­research/ gender-­studies-­product-­development-­historical-­overview. 53  Food and Drug Administration, “Guideline for the Study and Evaluation of Gender Differences in the Clinical Evaluation of Drugs; Notice” (July 1993) Fed Reg 58(139), 39406–39416. 54  Food and Drug Administration, “Investigational New Drug Applications and New Drug Applications: Final Rule” (1998) Fed Reg 63, 6854–6862.

648   Carl H. Coleman The 1998 FDA guidelines also require all new drug applications to include safety and efficacy data broken down by racial and ethnic subgroups. In 2005, the FDA recommended that researchers use the demographic categories created by the Office of Management and Budget to collect and report race and ethnicity information. These categories were most recently updated in 1997 and classify five races—White, Black or African American, Asian, American Indian and Alaska Native, Native Hawaiian or other Pacific Islander—and two ethnicities: Hispanic or Latino and not Hispanic or Latino.55 In contrast to the FDA guidelines, the NIH requires clinical trials that it funds to include women and members of racial minority groups in some situations. Under NIH guidelines, if there is reason to believe that men and women will respond differently to an intervention, Phase 3 studies of that intervention must be designed to assess its safety and efficacy for men and women separately, with adequate sample sizes for each gender. If the data from prior studies neither strongly support nor strongly negate the existence of significant gender-­based differences, the Phase 3 trials must still be analyzed according to gender-­based subgroups, but the trials do not need to provide high statistical power for each subgroup. NIH-­funded clinical trials may proceed without gender-­based subgroups only if data from prior studies “strongly support no significant differences of clinical or public health importance in intervention effect between subgroups.” Even in these situations, however, the guidelines “strongly encourage” the collection of gender-­based data.56 Similarly, under the NIH guidelines, if data from prior studies strongly indicate the existence of significant differences of clinical or public health importance in intervention effect among racial or ethnic subgroups, studies must be specifically designed to examine those differences. The NIH also encourages researchers to “develop appropriate and culturally sensitive outreach programs and activities” in order to “actively recruit and retain the most diverse study population consistent with the purpose of the research project.” The NIH recognizes that it may not always be possible to obtain appropriate representation of minority groups in research, particularly in certain parts of the country. However, if the representation of minority participants in a study is expected to be limited, the NIH requires investigators to provide a justification for the disparity, which will be considered as part of the NIH’s funding evaluation.57 Despite the FDA and NIH policies, substantial gender- and race-­based disparities in research continue to exist. Some researchers and advocates are working to address this problem through, for example, efforts to develop inclusive and culturally competent recruitment practices. The FDA could significantly enhance these efforts by requiring

55  Rugnetta, Michael J, and Desai, Khusboo, “Addressing Race and Genetics: Health Disparities in the Age of Personalized Medicine” (June 2011) Sci Prog, https://cdn.americanprogress.org/wp-­content/ uploads/issues/2011/06/pdf/race_genetics.pdf. 56  National Institutes of Health, “NIH Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research” 2001. https://grants.nih.gov/policy/inclusion/women-­and-­ minorities/guidelines.htm. 57 Id.

Research with Human Participants in the United States   649 the inclusion of diverse participant populations in clinical trials as a condition of product approval.

7  Compensation for Research Injuries Although international ethical guidelines provide that participants who are harmed as a result of participating in a study should receive compensation, no right to compensation for research injuries exists as a matter of US law. Instead, the Common Rule and FDA regulations require only that researchers inform participants whether any compensation and/or medical treatments will be available if an injury occurs and, if so, what they consist of and where further information may be obtained. Some research institutions in the United States have voluntarily adopted no-­fault compensation policies for research participants, but the majority have not. In the absence of a comprehensive compensation system for research-­related injuries, participants who suffer injury must either bear the costs themselves, rely on their own insurance, or seek damages through tort litigation. Both the Common Rule and the FDA human subject protection regulations prohibit researchers from asking participants to waive their right to sue.58 Potential tort claims include medical malpractice, ordinary negligence, failure to obtain informed consent, and products liability, among others. Establishing liability can be difficult, however, because of the need to establish both that the researcher deviated from the applicable standard of care and that this deviation was the cause of injuries the participant otherwise would not have experienced. The absence of a no-­fault compensation system for individuals who experience ­injuries while participating in research is a significant defect of the US regulatory s­ ystem. Ideally, Congress would pass legislation requiring all research sponsors to provide ­medical care and compensation to any participants who are injured as a result of participating in a study. Until that happens, IRBs should consider requiring provisions for compensation as a condition of approving any study involving more than minimal risk.

8  Clinical Trial Registration and Results Reporting Clinical trial registration began in the United States in 2000, when the NIH created the website clinicaltrials.gov, pursuant to a 1997 law that required it to establish a registry for effectiveness studies of investigational drugs for serious or life-­threatening conditions.59 58  45 C.F.R. § 46.116(a)(6),21 C.F.R. § 50.20(a). 59  National Institutes of Health, “ClinicalTrials.gov Background - ClinicalTrials.gov.” Accessed June 23, 2020. https://clinicaltrials.gov/ct2/about-­site/background.

650   Carl H. Coleman The primary rationale for the law was to provide information about ongoing trials for the benefit of persons who might be interested in becoming participants. While this rationale remains important, it is now recognized that clinical trial registration also serves another purpose: it makes it more difficult for researchers and sponsors to hide negative study results. In 2007, Congress enacted the FDA Amendments Act (FDAAA), which expanded the requirements for registering studies on clinicaltrials.gov. Whereas the 1997 legislation required registration only of effectiveness studies of drugs designed to treat serious or life-­threatening conditions, FDAAA expanded the requirement to include controlled clinical investigations of any drugs and biologics subject to FDA regulation, other than Phase 1 investigations, as well as most controlled trials of medical devices subject to FDA regulation. Even if the trial is conducted entirely outside of the United States, registration in clinical trials.gov is required if the drug, device, or biologic involved is manufactured in the United States, or if the study is being conducted pursuant to an investigational new drug application or an investigational device exemption.60 Although clinicaltrials.gov is maintained by the NIH, all studies that meet the FDAAA criteria are subject to registration, even studies conducted and funded entirely by private entities. However, since 2017, the FDAAA criteria no longer apply to NIH-­ funded research; instead, all studies funded in whole or in part by NIH now must comply with registration requirements, regardless of study phase or type of intervention.61 In addition to expanding the type of clinical trials that had to be registered, FDAAA also required the reporting of clinical trial results. The results reporting requirement was adopted to address the fact that a large percentage of clinical trial results are never published in the scientific literature. Under FDAAA, sponsors must report information about the progress of participants through each stage of the study, baseline characteristics of all participants, outcome measures and statistical analysis, a summary of all anticipated and unanticipated serious adverse events, and a summary of other adverse events exceeding a specific frequency threshold. For trials completed before January 18, 2017, results must be reported if the study involved a drug, biological, or device that was approved, licensed, or cleared by the FDA. For trials completed on or after January 18, 2017, results must be reported whether or not the investigational product was approved, licensed, or cleared.62 The failure to comply with registration or results reporting requirements can lead to civil monetary penalties. In addition, as of January 2015, the Centers for Medicare and Medicaid Services (CMS) has required a clinical trial identifier number to be reported 60  National Institutes of Health, “FDAAA 801 and the Final Rule - ClinicalTrials.gov.” Accessed June 23, 2020. https://clinicaltrials.gov/ct2/manage-­recs/fdaaa. 61  National Institutes of Health, “NOT-­OD-­16-­149: NIH Policy on the Dissemination of NIH-­ Funded Clinical Trial Information.” Accessed June 23, 2020. https://grants.nih.gov/grants/guide/ notice-­files/NOT-­OD-­16-­149.html. 62  Clinical Trials Registration and Results Information Submission, 81 FR 64981 (2016). https://www. federalregister.gov/documents/2016/09/21/2016-­22129/clinical-­trials-­registration-­and-­resultsinformation-­submission.

Research with Human Participants in the United States   651 on all billing claims for items or services related to qualifying clinical trials. This means that Medicare will not pay for the routine costs of any clinical trials that have not been properly registered.63 Failure to register a trial upon inception also may make it impossible to publish the trial’s results in a peer-­reviewed medical journal.64 However, despite the potential penalties, compliance with registration requirements—and, even more so, with the requirement to report the results of completed trials—remains poor. One reason is that regulators have taken no action to enforce the requirements; despite widespread violations, no fines have yet been imposed. Unless the registration and results reporting requirements are actively enforced, transparency in clinical research will be difficult to achieve.

9  Disclosing and Managing Conflicts of Interest Financial conflicts of interest in research have grown in the past several decades as a result of changes in the way that research is financed.65 Before the 1980s, the majority of clinical research in the United States was funded by the federal government, particularly the NIH. Today, the federal government funds only about one-­third of all biomedical research; most of the remaining funding is provided by the pharmaceutical and medical device industries. These commercial sponsors have a financial incentive to ensure that investigational products are brought to market as quickly as possible, which requires the prompt completion of clinical trials demonstrating the new product’s safety and efficacy. To achieve this goal, some sponsors may provide incentives to researchers to speed up trial enrollment, such as finder’s fees for identifying or referring potential participants, payments to researchers based on the number of subjects enrolled, or bonuses for meeting certain benchmarks in recruiting or retaining subjects. Commercial sponsors also may give researchers or research institutions a financial stake in the outcome of a study—for example, by providing compensation in the form of equity interests in the company. In addition to payments directly related to research, companies also provide financial benefits to researchers in the form of consulting arrangements or participation in paid speakers’ bureaus or advisory boards. These financial incentives create conflicts of interest because they have the potential to lead to decisions that are not necessarily

63  Centers for Medicare and Medicaid Services, “Mandatory-­Clinical-­Trial-­Identifier-­Number-­ QsAs.pdf,” 2014. https://www.cms.gov/Medicare/Coverage/Coverage-­with-­Evidence-­Development/ Downloads/Mandatory-­Clinical-­Trial-­Identifier-­Number-­QsAs.pdf. 64  International Committee of Medical Journal Editors, “ICMJE Recommendations: Clinical Trials.” Accessed June 23, 2020. http://www.icmje.org/recommendations/browse/publishing-­and-­editorial-­ issues/clinical-­trial-­registration.html. 65  For a more general discussion of conflicts of interest, see the chapters by Saver and Roksandić in this volume.

652   Carl H. Coleman consistent with the primary goals of promoting the scientific integrity of the research and protecting participants’ well-­being. Three main sets of regulations govern conflicts of interest in clinical research: FDA rules governing disclosure and management of investigators’ financial relationships66; DHHS regulations designed to promote objectivity in research funded by the Public Health Service (PHS), which includes the NIH as well as other entities such as the Centers for Disease Control and Prevention67; and disclosure requirements for pharmaceutical and medical device companies under the Physician Payment Sunshine provisions of the Affordable Care Act.68 The FDA conflict of interest requirements apply to studies that are submitted in support of marketing applications for approval of FDA-­regulated products. They require sponsors to include in their applications a list of investigators who worked on any covered clinical studies, as well as information about (1) financial arrangements where the value of the compensation could be influenced by the outcome of the research; (2) significant payments of other sorts, such as consulting or speaking fees, in excess of $25,000 from the sponsor to the investigator; (3) any proprietary interest, such as a patent or trademark, in the tested product; and (4) any significant equity interest in the sponsor, defined to include equity in excess of $50,000 in a publicly traded corporation or any equity in an entity that is not readily valued by reference to public prices. Notably, the regulations exclude from the disclosure requirements any payments to cover “the costs of conducting the clinical study or other clinical trials.” In addition to these disclosure requirements, the FDA regulations also require sponsors to report steps taken to minimize the potential for bias in research. In a 2013 guidance document, the FDA emphasized that sponsors can guard against bias by ensuring the proper design of clinical trials. For example, it noted that “having someone with no financial interests or arrangements evaluate study endpoints, especially in an unblinded study, can help minimize potential bias in assessing therapy outcomes.”69 If an investigator’s disclosure to the FDA raises data integrity questions, the FDA may decide to audit the data, require further analyses or studies, or decline to credit the study’s results. The DHHS conflict of interest regulations require institutions that receive PHS funding to maintain and enforce written policies that reduce or eliminate financial conflicts of interest, including provisions for monitoring and reporting investigators’ significant financial interests. The term “significant financial interest” is defined to include salaries, royalties, or other payments in excess of $5,000; equity interests in excess of $5,000 or 66  Food and Drug Administration, “Guidance for Clinical Investigators, Industry, and FDA Staff: Financial Disclosure by Clinical Investigators,” 2013. https://www.fda.gov/media/85293/download. 67  Responsibility of Applicants for Promoting Objectivity in Research for Which Public Health Service Funding Is Sought and Responsible Prospective Contractors: Final Rule, Federal Register 76 (2011): 53256–53293. 68  42 C.F.R. Part 403. 69  Food and Drug Administration, “Guidance for Clinical Investigators, Industry, and FDA Staff: Financial Disclosure by Clinical Investigators,” 2013. https://www.fda.gov/media/85293/download.

Research with Human Participants in the United States   653 that represent more than a 5% ownership interest in a single entity; and intellectual property rights, such as patents and copyrights. The regulations provide examples of conditions or restrictions that might be imposed to manage financial conflicts of interest, including disclosures to the public or research participants, appointment of an in­de­ pend­ent monitor for the study to protect against bias, modification of the research plan, changes in personnel, disqualification of personnel from the research or a portion of it, and/or reduction or elimination of the financial interest. In addition, institutions must make researchers’ financial conflicts of interest available to the public on request. Finally, under the Physician Payments Sunshine provisions of the Affordable Care Act, applicable manufacturers and group purchasing organizations are required to report to CMS any transfers of value to physicians and teaching hospitals, as well as any ownership or investment interests held by physicians or their immediate family members. CMS then posts this information on a searchable website as part of what it has named the Open Payments Program. The regulations require reporting of any transfers of value in excess of $10 per instance or $100 per year (adjusted annually for inflation), except for a few limited categories such as the provision of free product samples or gifts of small incidental items provided at large conferences. Each transfer must be designated as falling within one of several reportable categories. One of these categories is “research,” which includes payments that are subject to a written research agreement or protocol. Unlike other transfers of value, which are published on an annual basis, research payments will not be published until the manufacturer obtains FDA approval for the product or 4 years following the date of the payment.70 The delay in publication is designed to address manufacturers’ concern that the immediate disclosure of research payments would alert competitors to confidential business information. The delayed publication does not apply to payments related to studies of off-­label uses of products that are already on the market. Whether conflict of interest disclosure requirements have been effective is difficult to determine. On one hand, researchers who know that their financial relationships will be subject to public disclosure may be more reluctant to enter into relationships that could compromise their scientific objectivity. On the other hand, some researchers might see disclosure as a way to “sanitize” potentially questionable relationships, which could have the opposite effect. Moreover, it is doubtful that most research participants routinely check the Open Payments Database, and, even if they do, it is not clear how they are supposed to use the information to make decisions about study participation. Overall, while disclosure requirements can increase transparency about conflicts of interest in clinical research, it seems unlikely that disclosure alone will eliminate these conflicts’ pernicious effects.

70  42 C.F.R. 403.910.

654   Carl H. Coleman

10  Conclusion As the foregoing discussion shows, most types of research involving human participants in the United States is subject to significant ethical oversight. However, the oversight system has major gaps, particularly for privately funded studies not involving FDA-­regulated products. In addition to ethics oversight, the United States has also taken important steps toward addressing other research-­related issues, but many of these efforts have been insufficient, either because of inadequate laws or lack of enforcement. Priorities for reform include stronger policies regarding the inclusion of women and racial minorities in research, a national requirement for no-­fault compensation for research-­ related injuries, enforcement of laws requiring clinical trial registration and prompt reporting of results, and efforts to address conflicts of interest in research that go beyond disclosure requirements. Greater attention to these issues is essential to ensuring that research with human participants in the United States adheres to international ethical norms.

Bibliography Overview of the Regulatory Framework Coleman, Carl H, ‘Rethinking the Regulatory Triggers for Prospective Ethics Review” (June 1, 2019) J L Med Ethics 47(2) 247–253. Loe, Jonathan D, D Alex Winkelman, and Christopher T Robertson, “An Assessment of the Human Subjects Protection Review Process for Exempt Research” (September 2016) J L Med Ethics 44(3) 481–491. Meyer, Michelle N, Mark A Rothstein, and John T Wilbanks, “There Oughta Be a Law: When Does(n’t) the US Common Rule Apply?” (March 2, 2020) J L Med Ethics 48, 60–73. Tovino, Stacey A, Mark A Rothstein, and John T Wilbanks, “Mobile Research Applications and State Research Laws” (March 2, 2020) J L Med Ethics 48, 82–86.

The IRB’s Place in the Research Oversight System Emanuel, Ezekiel J, Trudo Lemmens, and Carl Elliot, “Should Society Allow Research Ethics Boards to Be Run As For-Profit Enterprises?” (July 2006) PLoS Medicine 3(7). Klitzman, Robert, Ekaterina Pivovarova, Alexandra Murray, Paul S Appelbaum, Deborah F Stiles, and Charles W Lidz, “Local Knowledge and Single IRBs for Multisite Studies: Challenges and Solutions” (2019) Ethics Hum Res 41(1), 22–31. Menikoff, Jerry, “The Paradoxical Problem with Multiple-IRB Review” (2010) N Engl J Med 363(17), 1591–1593. Peckman, Steven, “Local Institutional Review Boards,” in Ethical and Policy Issues in Research Involving Human Participants (Bethesda, MD: National Bioethics Advisory Commission 2001), K1–K35.

Research with Human Participants in the United States   655 Wolinetz, Carrie D, and Francis S Collins, “Single-Minded Research Review: The Common Rule and Single IRB Policy” (July 2017) Am J Bioethics 17(7), 34–36.

Key Ethical Considerations for IRB Review Risk-­Benefit Assessment Caulfield, Timothy, Nola Ries, and Graham Barr, “Variation in Ethics Review of MultiSite Research Initiatives Health Care, Bioethics and the Law: Opinion Article” (2011) Amsterdam L Forum 3(1), 85–100. Eyal, Nir, Marc Lipsitch, Till Bärnighausen, and Dan Wikler, “Opinion: Risk to Study Nonparticipants: A Procedural Approach” (August 7, 2018) Proc Natl Acad Sci 115(32), 8051–8053. Mello, Michelle M, and Leslie E Wolf, “The Havasupai Indian Tribe Case — Lessons for Research Involving Stored Biologic Samples” (2010) N Engl J Med 363, 204–207.

Informed Consent Lynch, Holly Fernandez, Leslie E Wolf, and Mark Barnes, “Implementing Regulatory Broad Consent Under the Revised Common Rule: Clarifying Key Points and the Need for Evidence” (June 1, 2019) J L Med Ethics 47(2), 213–231.

Additional Protections for Special Populations Louie, Mikaela LJ, “Respecting the Right to Research: Proxy Consent and Subject Assent in Alzheimer’s Disease Clinical Trials” (2019) Washington L Rev 94, 887–926.

Inclusion of Women and Racial Minorities in Research Borno, Hala T, et al, “A Step Towards Equitable Clinical Trial Recruitment: A Protocol for the Development and Preliminary Testing of an Online Prostate Cancer Health Information and Clinical Trial Matching Tool” (November 7, 2019) Pilot Feasibility Stud 5(1), 123. https:// doi.org/10.1186/s40814-019-0516-4. Dresser, Rebecca, “Wanted: Single, White Male for Medical Research” (January 2, 1992) Hastings Cent Rep 22(1), 24–29. Loree, Jonathan M, et al, “Disparity of Race Reporting and Representation in Clinical Trials Leading to Cancer Drug Approvals from 2008 to 2018” (October 1, 2019) JAMA Oncol 5(10), e191870–e191870. Rothenberg, Karen, “Gender Matters: Implications for Clinical Research and Women’s Health Care” (1996) Houston L Rev 32, 1201–1272.

Compensation for Research Injuries Henry, Leslie Meltzer, Megan E Larkin, and Elizabeth R Pike, “Just Compensation: A No-Fault Proposal for Research-Related Injuries” (August 12, 2015) J L Biosci 2(3), 645–668.

656   Carl H. Coleman Clinical Trial Registration and Results Reporting DeVito, Nicholas J, Seb Bacon, and Ben Goldacre, “Compliance with Legal Requirement to Report Clinical Trial Results on ClinicalTrials.gov: A Cohort Study” (February 1, 2020) Lancet 395(10221), 361–369. Lemmens, Trudo, and Candice Telfer, “Access to Information and the Right to Health: The Human Rights Case for Clinical Trials Transparency” (2012) Am J L Med 38 (1), 63–112.

Disclosing and Managing Conflicts of Interest Dorsey, E Ray, Jason de Roulet, Joel P Thompson, Jason I Reminick, Ashley Thai, Zachary White-Stellato, Christopher A Beck, Benjamin P George, and Hamilton Moses, “Funding of US Biomedical Research, 2003–2008” (January 13, 2010) JAMA 303 (2), 137–143. Institute of Medicine Committee on Conflict of Interest in Medical Research, Education, and Practice; Lo B, Field MJ, editors. Conflict of Interest in Medical Research, Education, and Practice. (Washington DC: National Academies Press 2009).

chapter 32

Bioba n ks Graeme T. Laurie and Carl H. Coleman

1  The Phenomenon of Biobanking The advent of large-­scale biobanks—collections of biomedical samples linked to ­personal data and other resources such as health records—has come to represent a watershed in health research regulation in many countries. Notable examples include UK Biobank, which holds the data and biological samples of more than 500,000 UK citizens; the public biobank established by the Estonian Human Genes Research Act of 2000, which holds data and tissue from more than 50,000 Estonians while offering personal access and genetic testing; and the Mayo Clinic Biobank in the United States, which consists of blood, blood derivatives, and health information donated by Mayo Clinic patients. Biobanks come in many shapes and sizes, from the population-­based initiatives just mentioned down to collections of samples in a research lab fridge. While biobanks are not novel per se, the ability to build and sustain such resources on a large scale, for use in prospective and as-­yet undetermined research objectives, has grown substantially in recent years, raising a range of challenges for regulation and governance. As this chapter demonstrates, the search for effective biobank governance has often involved a “reflexive turn” in oversight mechanisms: that is, there is less emphasis on biobank-­specific hard laws and more attention to soft law models, such as ethics and governance frameworks and/or stakeholder engagement strategies that require ongoing reflection and deliberation on the ethical and social implications of these projects. The European Commission’s 2012 study into biobanks reminds us of the diversity of collections that can be grouped under the nomenclature of “biobank.”1 These include (i) population-­based biobanks, (ii) disease-­oriented biobanks, (iii) case-­control biobanks, (iv) tissue banks, (v) biobanking as part of clinical trials, and (vi) diverse other collections, 1  European Commission, “Biobanks for Europe: A Challenge for Governance” (Directorate-­General for Research and Innovation, EUR 25302 EN, 2012) https://www.coe.int/t/dg3/healthbioethic/ Activities/10_Biobanks/biobanks_for_Europe.pdf; accessed 8 October 2018.

658   Graeme T. Laurie and Carl H. Coleman including cord blood banks, stem cell banks, and newborn genetic screening collections. As a result, there is no universally agree definition of biobanks, and the Council of Europe’s (CoE) 2006 and revised 2016 Recommendation talk instead of “biological materials of human origin.” Regardless of terminology, it is the prospect of opening collections to a potentially limitless category of future research that focuses our attention here. The viability and sustainability of biobanks is dependent on three crucial technical considerations: (i) the quality of the banked materials, which include biomedical specimens, personal data about the source of the specimens, and metadata to make sense of the significant volumes of data being held in newly generated databases (or linkages from and to biobanks across existing databases of health-­related data); (ii) traceability of data and materials when sharing and using biobank resources; and (iii) interoperability of standards, protocols, and—arguably—governance mechanisms to ensure that biobanks and associated datasets can interact to promote the utmost value from the collections. The biobanking community is typified by a series of well-­funded, high-­profile initiatives that represent multidisciplinary efforts to meet these challenges. In Europe, for example, the BBMRI-­ERIC consortium involves biobanks across 20 countries, with infrastructure funding from the European Union and with a claim to be the world’s largest biobank directory.2 In the United States, the National Institutes of Health’s “All of Us” initiative is seeking to collect 35 specimens from at least 1 million individuals, with a particular emphasis on groups that have traditionally been underrepresented in medical research.3 Self-­evidently, all biobanks must comply with local and international legal regimes governing core elements of the resource. These include privacy and security laws, data protection requirements, laws governing human organs and tissues, and laws relating to research with human participants. These aspects of regulation are examined elsewhere in this volume and will not be repeated here. Instead, this chapter examines biobanks as a case study of multiple legal and governance challenges that cut across conventional jurisdictional lines. In particular, we stress two features of biobanks that raise questions about the role of law and regulation: first, we examine the extent to which existing legal frameworks must be adapted to accommodate particular features of biobanking (and how far this is successful), and second, we highlight examples of ways in which some biobanks have developed their own governance mechanisms—in addition to legal com-

2  This is the Biobanking and BioMolecular Resources Research Infrastructure-­European Research Infrastructure Consortium’s (BBMRI-­ERIC) mission statement: BBMRI-­ERIC is a European research infrastructure for biobanking. We bring together all the main players from the biobanking field—researchers, biobankers, industry, and patients—to boost biomedical research. To that end, we offer quality management services, support with ethical, legal and societal issues, and a number of online tools and software solutions. Ultimately, our goal is to make new treatments possible. BBMRI-­ERIC Homepage: http://www.bbmri-­eric.eu/. 3  National Institutes of Health, All of Us Research Program Backgrounder, https://allofus.nih.gov/ news-­events-­media/media-­toolkit/all-­us-­research-­program-­backgrounder.

Biobanks   659 pliance in the host country—that are aimed at promoting engaged and ethically robust biobanking practices.

2  Overview of Legal and Ethical Responses There have been a wide variety of ethico-­legal responses to biobanks globally, and only a few of these have taken the form of bespoke law (i.e., specific biobank legislation). Examples of countries that have taken this route include Estonia4 and Iceland,5 which were among the first to pass legislation arising from concerns about local population-­ based collections. Other jurisdictions have treated aspects of biobanking as matters of bioethics and public health law (France6), data protection and human tissue regulation (United Kingdom7), and/or as examples of general biomedical research regulation (Spain8) or genetic and health information regulation (Portugal9). Most recently, Singapore has become the first country in the world to require consent for secondary research on anonymized tissue under its regulatory regime that came into full force in 2019.10 “Secondary research” relates to research that is unconnected with the primary purposes for which tissue are provided (e.g., for diagnostic purposes). A common reaction has been for stakeholders to look beyond the realm of law to offer better support for the biobanking community and to protect citizens whose material and data become part of such collections. For example, the European Commission’s 2012 report called for “embeddedness of new governance bodies, and biobanks themselves, into research and healthcare delivery systems.” Likewise, the CoE has issued a generic framework document on how to build an optimal biobank collection, conforming not only to the European Convention on Human Rights, but also to the ever-­ changing technological needs of the biobanking community. In 2009, the Organisation for Economic Co-­operation and Development (OECD) adopted its Guidelines on Human Biobanks and Genetic Research Databases, which was designed primarily to

4  Estonian Human Genes Research Act 2000. 5  Icelandic Act on Biobanks no 110/2000, 13 May 2000. 6  Law no 2011–814, 7 July 2011 on Bioethics; JORF no 0157 of 8 July 2011, 11826. 7  Data Protection Act 1998, as amended 2018, and the Human Tissue Act 2004; for current Health Research Authority approaches in the UK, https://www.hra.nhs.uk/about-­us/news-­updates/biobanks-­ and-­how-­were-­helping-­uk-­become-­world-­leader-­life-­sciences/; accessed 8 October 2018. 8  Biomedical Research Act 2007. 9  Law no 12/2005, 26 January (Act on personal genetic information and information regarding health). 10  Human Biomedical Research Act (Act no 29 of 2015) and related legislation, such as the Human Biomedical Research Regulations 2017 and the Human Biomedical Research (Restricted Research) Regulations 2017.

660   Graeme T. Laurie and Carl H. Coleman assist biobanks and researchers in setting up new biobanks for research purposes.11 Its “Principles + Best Practices” approach has the benefit of avoiding overly prescriptive, highly specific, and rule-­like diktats while encouraging due consideration of the core values underpinning good governance.12 The reasons this kind of approach has emerged (i.e., one that seems to go above and beyond what is required by or can be proscribed by law) are manifold. They include the fact that few jurisdictions were willing or able to pass biobank-­specific laws despite the fact that the international bioethics community was highlighting numerous ways in which biobanking could not necessarily be accommodated within existing norms, such as informed consent (see further discussion later in this chapter). Moreover, the longitudinal nature of biobanks (i.e., that they are set up specifically to attract future, as-­yet unknown research projects) not only caused problems in informing participants about what precisely might be done with their data and samples, but it also meant that laws were often considered to be too blunt a tool with which to anticipate future uses or likely future harms. Many international and national agencies therefore sought to lay out broad principles-­based considerations for novel biobanking initiatives that supported this open-­endedness of scientific purpose while also ensuring that oversight mechanisms had due regard to participants’ rights and interests. It should be self-­ evident that any biobank must, as a minimum, conform with existing local laws, such as those related to the storage, management, and use of personal data and human tissue. This flexible approach is also well suited to the inherently cross-­jurisdictional nature of biobanking, where the cumulative effect of multiple legal regimes can be significantly disruptive and unduly burdensome. A common feature to emerge in many biobank initiatives has been the notion of interoperable governance, whereby translatable policies are developed that can be adapted to specific contexts, and/or meta-­institutions, such as meta-­data access committees, which seek to ensure the consistent application of standards in different environments.

3 Consent One of the core challenges arising from contemporary biobanking is the role of consent. This is so for multiple reasons. First, historical biomedical collections might have been established without consent (or at least not with consent that would meet modern ethical and legal requirements). Blood samples collected as part of newborn screening programs, many of which exist in many countries and date back to the 1960s, are an 11  Organisation for Economic Co-­operation and Development, “Guidelines on Human Biobanks and Genetic Research Databases” http://www.oecd.org/health/biotech/44054609.pdf; accessed 8 October 2018. 12  Graeme Laurie and Nayha Sethi, “Towards Principles-­Based Approaches to Governance of Health-­Related Research Using Personal Data” (2013) 4(1) EJRR 43, doi: 10.1017/S1867299X00002786; accessed 8 October 2018.

Biobanks   661 example of this.13 It is commonly the case that such historical collections are not based on consent from the parents simply because the ethical ethos at the time they were established did not require informed consent. Almost all such collections now actively seek consent to retain blood samples, but there are millions of samples held around the world without such consent. This issue has sometimes resulted in destruction of such collections.14 Second, consent might have been given for one purpose (e.g., medical care) and this might not extend to a new purpose, such as medical research. Third, the ethical and legal requirement that consent be informed is particularly challenging with biobanks given that the reason they exist is to promote prospective, as-­yet unspecified research. How, then, can biobankers adequately inform participants of what will be done with their samples and data?

3.1  United States 3.1.1  The Common Rule The primary source of US law concerning consent to the use of stored biospecimens is the Federal Policy for the Protection of Human Subjects, known as the “Common Rule.”15 The Common Rule applies to any human subject research conducted or funded by any of the 20 federal agencies that adhere to the Rule. Even biobanks that do not themselves receive federal research funding need to ensure that their policies and practices conform to the Common Rule if they want to be able to provide their samples to researchers in federally funded studies. 3.1.1.1  The Scope of the Common Rule’s Consent Requirement The Common Rule’s applicability to biospecimen research turns on the regulatory ­definition of “human subject.” Under the Rule, a “human subject” is defined as “a living individual about whom an investigator . . . obtains (1) data through intervention or interaction with the individual; or (2) identifiable private information.”16 Thus, individuals whose biospecimens are used in research are considered human subjects if the researcher will have access to identifiable private information about them, regardless of whether they interact with the researcher directly. 13  Graeme Laurie, Kathryn Hunter with Sarah Cunningham-­Burley, Guthrie Cards in Scotland: Ethical, Legal and Social Issues (Scottish Government, 2013) https://www.gov.scot/ Resource/0044/00441799.pdf; accessed 8 October 2018; US examples involving destruction: Mary Carmichael, “Newborn Screening: A Spot of Trouble” (2011) 475 Nature 156, doi: 10.1038/475156a; and Michelle H Lewis et al, “Blood Samples: State Laws Regarding the Retention and Use of Residual Newborn Screening Blood Samples” (2011) 127 Paediatrics 703. 14  Also, Katherine Drabiak-­Syed, “Legal Regulation of Banking Newborn Blood Spots for Research: How Bearder and Beleno Resolved the Question of Consent” (2011) 11 Hous J Health L & Pol’y 1; Michelle H Lewis, “Lessons from the Residual Newborn Screening Dried Blood Sample Litigation” (2015) Spring J Law, Med., Ethics 32. 15  45 C.F.R. Part 46. 16  45 C.F.R. § 46.102(e).

662   Graeme T. Laurie and Carl H. Coleman Human subject research covered by the Common Rule normally requires the subject’s informed consent unless the institutional review board (IRB) reviewing the study has waived the consent requirement. This means that researchers seeking to use identifiable biospecimens from a biobank in federally funded research must be able to provide documentation that either (1) the source of the biospecimens provided informed consent to the study, or (2) the IRB has waived the informed consent requirement. Conversely, if the research does not have access to identifying information, the Common Rule is inapplicable—meaning that no documentation of consent or a waiver is required at all (absent some other legal basis requiring consent, such as state law). In 2004, the Office for Human Research Protections (OHRP), the federal agency in charge of Common Rule enforcement, issued a guidance document elaborating on the meaning of “identifiable” in biospecimen research.17 The guidance explained that “private information or specimens” are not identifiable if “they cannot be linked to specific individuals by the investigator(s) either directly or indirectly through coding systems.” Under this interpretation, the focus is not on whether it is possible for someone to connect the specimens to identifying information, but whether the identifying information will be available to the investigators. Thus, even when specimens are stored in a biobank with links to identifying information, they will be considered nonidentifiable in the context of a particular study as long as the researcher has no ability to access the links. In 2015, federal regulators published a notice of proposed rulemaking outlining potential revisions to the Common Rule.18 Perhaps the most dramatic of the suggested changes was a proposal to treat all research with biospecimens as “human subject” research, regardless of whether the biospecimens are linked to identifiable information. Had this change been adopted, all biospecimen research would have presumptively required the consent of the individuals from whom the biospecimens were obtained— even research with samples for which all identifying information had been permanently removed. In the preamble to the proposal, the agencies gave two rationales for the change.19 First, they suggested that technological changes made it questionable whether a biospecimen could ever be permanently de-­identified. Second, they suggested that “people want to be asked for their permission,” pointing to “[a] growing body of survey data show that many prospective participants want to be asked for their consent before their biospecimens are used in research.”20 However, when the final revisions to the Common Rule were released in 2017, the proposal to treat all research with biospecimens as human subject research was not included among the changes. The preamble to the Final Rule explained that the premise 17  Office for Human Research Protections, Coded Private Information or Specimens Use in Research, Guidance (2008), available at https://www.hhs.gov/ohrp/regulations-­and-­policy/guidance/research-­ involving-­coded-­private-­information/index.html. (This 2008 document updates and replaces the original 2004 guidance.) 18  Notice of Proposed Rulemaking, Federal Policy for the Protection of Human Subjects, 80 Fed Reg 173, 8 September 2015, at 53933. 19  Id. at 53943. 20  Id. at 53938.

Biobanks   663 of the proposed change—“that continuing to allow secondary research with biospecimens collected without consent for research places the publicly funded research enterprise in an increasingly untenable position because it is not consistent with the majority of the public’s wishes”—was belied by the public comments to the proposal, most of which opposed expanding the Rule to cover research with de-­identified biospecimens.21 At the same time, recognizing that technological developments might someday make it possible to re-­identify information previously considered to have been permanently de-­identified, the Final Rule directs agencies to regularly review the definition of “identifiable private information” and “identifiable biospecimen” in consultation with relevant experts and, if appropriate, to “alter the interpretation of these terms, including through the use of guidance.”22 3.1.1.2  Options for Obtaining or Waiving Consent Prior to the 2017 amendments to the Common Rule, researchers seeking to use ­identifiable biospecimens were limited to two options: (1) show that the sources of the biospecimens provided informed consent to the specific study in question (following disclosure of the full details of the study, such as how long it would last and the procedures that would be used) or (2) seek a waiver of the consent requirement from the IRB. Thus, unless the researcher was prepared to go back to the sources of all the samples to obtain study-­specific informed consent, the only way to use the samples would be to convince the IRB to grant a waiver of consent. As discussed further below, the 2017 amendments now adds the additional option of relying on “broad consent” as a basis for using the samples, but only for specimens collected in 2019 or later. The Common Rule authorizes IRBs to waive the informed consent requirement if they determine that the study involves no more than minimal risk, that it could not “practicably be carried out” without the waiver, and that the waiver “will not adversely affect the rights and welfare of the subjects.”23 The term “minimal risk” is defined in the regulations to mean that “the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.”24 IRBs vary in their willingness to grant waivers of informed consent for research with identifiable biospecimens. A 2015 study found that “[n]early 37 percent of IRBs prohibit or discourage a request for a waiver of informed consent whereas almost 24% encourage or require this option when faced with a request for a new use outside the scope of the original consent used for specimen collection.”25 Disagreements may involve whether

21  Final Rule, Federal Policy for the Protection of Human Subjects, 82 Fed Reg 12, 19 January 2017, at 7149, 7168. 22  45 C.F.R. 46.102(e)(7). 23  45 C.F.R. 46.116(f)(3). 24  45 C.F.R. 46.102(j). 25  Goldenberg AJ et al, “IRB Practices and Policies Regarding the Secondary Research Use of Biospecimens” (2015, 8 May) BMC Med Ethics 16.

664   Graeme T. Laurie and Carl H. Coleman the study could not practicably be carried out without obtaining individual consent, or whether the particular study involves more than minimal risk. The 2017 revisions to the Common Rule (effective in 2019) reduced the significance of waiver determinations by authorizing individuals to provide “broad consent” to the future use of their identifiable biospecimens in research. Before providing broad consent, the subject must be given a general description of the types of research that may be conducted with the biospecimens, which must include “sufficient information such that a reasonable person would expect that the broad consent would permit the types of research conducted.” Because the revisions are still new, it is too early to tell how IRBs will interpret the “sufficient information” requirement. Before the regulations were adopted, some commentators suggested that individuals providing broad consent “should be told that any research [is] possible unless specifically limited in the consent form or overruled by the oversight body.” Others called for providing “a broad but non-­ exhaustive description of possible research topics, including the possibility of genetic analyses and keeping cells for indefinite periods, as well as other techniques to be developed.”26 As part of the process of obtaining broad consent, individuals must be told that they should not expect to receive the results of this research, even if it yields clinically relevant research results. In addition, they must be told that the biospecimens may be used for commercial profit, whether they will share in this profit, and whether the research will or might include whole-­genome sequencing.27 The revised Rule contains two additional provisions designed to reduce the reliance on consent waivers. First, it provides that waivers of informed consent should not be granted in research with identifiable biospecimens unless the IRB determines that “the research could not practicably be carried out without using such . . . biospecimens in an identifiable format.”28 Second, the Rule provides that “[i]f an individual was asked to provide broad consent for the storage, maintenance, and secondary research use of identifiable private information or identifiable biospecimens in accordance with the requirements . . . of this section, and refused to consent, an IRB cannot waive consent for the storage, maintenance, or secondary research use of the identifiable private information or identifiable biospecimens.”29

3.1.2  Other Sources of Consent Obligations in US Law Even if a study is not subject to the Common Rule, researchers may be subject to state law obligations to ensure that informed consent to the use of biospecimens has been obtained. Relevant state laws include medical records statutes, health privacy laws, genetic testing laws, and laws on human subject protection.30 These statutes may expand 26  Grady C et al., “Broad Consent for Research with Biological Samples: Workshop Conclusions” (2015, 2 September) 15(9) Am J Bioethics 38. 27  45 C.F.R. 46.116(d). 28  45 C.F.R. 46.116(f)(3)(iii). 29  45 C.F.R. 46.116(f)(1). 30  Hakimian R et al, “50-­State Survey of Laws Regulating the Collection, Storage, and Use of Human Tissue Specimens and Associated Data for Research” (2004) National Cancer Institute Cancer Diagnosis Program.

Biobanks   665 the consent requirement to research not supported by federal funding, as well as to research involving coded samples, which would not be considered “human subject research” under OHRP’s interpretation of the Common Rule. Medical records statutes in all states generally require informed consent to the release of medical information, including biospecimens and their associated data. Several state statutes impose specific consent requirements for the disclosure of genetic i­ nformation.31 Most of these statutes have exceptions to the consent requirement for disclosures to ­scientific researchers, provided the information is not identifiable.32 In addition to medical records and privacy laws, some states have human subject protection laws that apply to all human subject research performed in the state, regardless of funding source.33 These laws typically work by extending the Common Rule to privately funded studies that would be covered if they were supported by federal funds. Such laws are particularly important for biobanks not affiliated with the kind of entities likely to receive funding from Common Rule agencies, such as hospitals and medical centers.

3.2  The European Union and Its Member States As will be clear from the preceding discussion, there is no single, uniform model of regulation or governance that applies to biobanks throughout the European Union, and this is equally true of approaches to consent. While the 1995 Data Protection Directive34 did provide that consent is a core and legitimate legal basis for data processing, its unifying effects were diluted by the fact that member states were at liberty to implement the Directive into national laws as they saw fit within their margin of appreciation; moreover, any given state could impose higher standards if it chose to do. The General Data Protection Regulation (GDPR),35 which came into force in May 2018, is partly designed to address variation in standards across Europe, but as shown in the later discussion, its provisions on consent are challenging to biobanking initiatives, and another legal basis might be preferable.

3.2.1  National Legislation and Practices A range of legal measures are deployed by countries in tackling the role of consent to biobanking. Existing constitutional (Germany) or Civil Code (Netherlands) requirements influence the tenor of the specific rules: for example, the German constitutional 31  Electronic Privacy Information Center, State Genetic Privacy Policy, available at https://www.epic. org/state-­policy/genetic-­privacy/. 32  Hakimian et al, “50-­State Survey.” 33 Id. 34  Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data [1995] L281/31. 35  Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data [2016] L119/1.

666   Graeme T. Laurie and Carl H. Coleman position is that specific informed consent is required for the taking of samples for research—requiring information on the “purpose, significance, and implications of the encroachment”—but once tissue has been removed from the body it is treated as property, and downstream uses of it are subject to contractual arrangements. Only Finland and Norway have passed specific legislation dealing with consent requirements for both data and tissue together. In all cases, exceptions to reliance on consent are possible (subject to the German constitutional requirements for consent to obtain material for research). Indeed, countries seem to base their models on some version of the tripartite dynamic of “consent-­ anonymize-­authorize.” For example, in the United Kingdom, consent for tissue use is not required so long as the tissue is appropriately anonymized and suitable research ethics committee (REC) approval is provided. Here, “anonymized” means “that it is not reasonably possible to trace the person from whose body the material has come” (Human Tissue Act 2004, s.7(1)(b)). All countries, except Germany, allow the transfer of clinical samples to research biobanks without consent, using variations of the tripartite model. It is considered good practice to require REC approval for research using biobanks, but the prospect of variation in REC processes and decisions is well-­recognized.36 This applies also when consent is the legitimating factor and RECs must determine what amounts to valid consent. There is, therefore, an important distinction between the text of law and the context in which it is interpreted and applied. Models for mutual recognition of ethics review are explored fully in the literature.37 The BioSHaRE-­EU project, a consortium to facilitate biobank collaboration across Europe, undertook a study of approaches to consent in Finland, France, Germany, the Netherlands, Norway, and the United Kingdom.38 The conclusion was that, despite extensive variation in the legal means by which the consent requirement was met, there was “remarkable uniformity” in the consent processes themselves.

3.2.2  Conformity with GDPR The GDPR is designed to promote uniformity of standards in personal data processing in Europe and internationally.39 It makes no specific mention of biobanks, and indeed its application to (international) human health research is a matter of interpretation and application of the general provisions of the Regulation to specific research initiatives.40 36  David Townend et al, “Streamlining Ethical Review of Data Intensive Research” (2016) 354 BMJ doi: 10.1136/bmj.i4181; accessed 8 October 2018. 37  Edward S Dove et al, “Ethics Review for International Data-­intensive Research” (2016) 351(6280) Science 1399, doi: 10.1126/science.aad5269; accessed 8 October 2018. 38  Jane Kaye et al, “Consent for Biobanking: The Legal Frameworks of Countries in the BioSHaRE-­EU Project” (2016) 14(3) Biopreservation and Biobanking. ahead of print doi: 10.1089/ bio.2015.0123; accessed 8 October 2018. 39  Michael Morrison et al, “The European General Data Protection Regulation: Challenges and Considerations for iPSC Researchers and Biobanks” (2017) 12(6) Regen Med 693, doi: 10.2217/ rme-­2017-­0068; accessed 8 October 2018. 40  For discussion, see Edward S Dove, “The EU General Data Protection Regulation: Implications for International Scientific Research in the Digital Era” (2018) 46(4) J Law Med Ethics 1013.

Biobanks   667 That is, the GDPR relates to the regulation of the processing of personal data across all domains, scientific, commercial, and otherwise. The GDPR defines consent as “any freely given, specific, informed and unambiguous indication of the data subject's wishes by which he or she, by a statement or by a clear affirmative action, signifies agreement to the processing of personal data relating to him or her.”41 Manifestly, this definition raises challenges for any biobanking initiative seeking to rely on “broad consent” to future, as-­ yet-­unspecified research projects because doubts are raised both about whether this is adequately informed and sufficiently specific. To an extent, this is recognized by Recital 33 of the GDPR, which states that It is often not possible to fully identify the purpose of personal data processing for scientific research purposes at the time of data collection. Therefore, data subjects should be allowed to give their consent to certain areas of scientific research when in keeping with recognised ethical standards for scientific research. Data subjects should have the opportunity to give their consent only to certain areas of research or parts of research projects to the extent allowed by the intended purpose.

This provision, however, must be read together with the Article 29 Working Party guidance on consent, which explains that Recital 33 does “not disapply the obligations with regard to the requirement of specific consent.” The Working Party guidance provides a helpful checklist of considerations for parties seeking to rely on consent as the lawful basis for conducting research involving personal data. It emphasizes the following points: • Research projects must have a well-­described purpose. If the precise purpose cannot be specified at the outset, the data controller “must seek other ways to ensure the essence of the consent requirements are served best, for example, to allow data subjects to consent for a research purpose in more general terms and for specific stages of a research project that are already known to take place at the outset. As the research advances, consent for subsequent steps in the project can be obtained before that next stage begins.” • Transparency is an important safeguard when research does not initially allow for a specific consent (e.g., having a comprehensive research plan that specifies research questions or working methods and to which participants have ready access). It is also good practice to appoint a designated contact person for data subjects to approach with questions.42 • As research progresses, data controllers should provide more information so that— over time—consent becomes more specific.

41  Article 4(11), GDPR. 42  See Article 14(1) of the Personal Data Act of Finland (Henkilötietolaki, 523/1999).

668   Graeme T. Laurie and Carl H. Coleman • Anonymization is the preferred solution as soon as the purpose of the research can be achieved without the processing of personal data. • The right to withdraw must always remain. Notwithstanding, concerns remain in some quarters that (i) “rolling” consent and repeat contacts with participants might be necessary under the GDPR,43 and (ii) this is at odds with the US revised Common Rule, which allows a one-­time authorization of open-­ended categories of research without any requirement for “specificity” in the European sense. All of this having been said, it must also be remembered that consent is only one of a range of lawful bases for processing personal data in Europe. Other bases for processing personal data are considered under the next section.

4  Privacy Protections All US and European jurisdictions recognize the central importance of privacy protection when set against the promise of public benefits from biobanking. The specific approaches are detailed here.

4.1  United States 4.1.1  The Common Rule As discussed earlier in connection with consent, federally funded research with identifiable biospecimens is subject to the federal human subject protection regulations, known as the Common Rule. Under the Common Rule, most research must be reviewed by an IRB, which must determine, among other things, that “there are adequate provisions to protect the privacy of subjects and to maintain the confidentiality of data.”44 Before the Common Rule was revised in 2017, research with identifiable biospecimens was exempt from IRB review if the biospecimens were publicly available or the investigator recorded only non-­identifying information about the biospecimens. Under the 2017 changes (effective in 2019), a new exemption is available for studies involving identifiable biospecimens for which broad consent was provided. However, IRBs must still conduct a “limited review” of such studies to ensure that adequate privacy and confidentiality protections are in place.45

43  Nicholas Wallace, David Peloquin, Leslie Thornton and Mark Barnes, “Final Guidelines on Consent Requirements Under the EU General Data Protection Regulation: Implications for Scientific Research” (2018) 17 Med Res Law & Pol’y Rep 192. 44  45 C.F.R. 46.111(a)(7). 45  45 C.F.R. 46.104(d)(8).

Biobanks   669

4.1.2  The HIPAA Privacy Rule The Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule applies to biobanks that are part of “covered entities,” a term that includes most healthcare providers.46 Thus, a biobank that is affiliated with a hospital or medical center would be covered by HIPAA; an unaffiliated biobank established by a private foundation would not be. However, even biobanks that are not directly covered by HIPAA need to be concerned about complying with the Privacy Rule if they ever want to share samples with HIPAA-­covered researchers. HIPAA generally requires an individual’s written authorization for the use of individually identifiable health information, which would include biospecimens associated with identifying information. Until 2013, HIPAA required authorizations for research to be specific to the particular study, a position that precluded individuals from providing broad authorizations that would cover future research uses. However, under changes made in 2013, individuals may now authorize the use of their identifiable biospecimens in future research as long as the authorization describes the future research “such that it would be reasonable for the individual to expect that his or her protected health information could be used or disclosed for such future research.”47 If no authorization has been obtained, biospecimens may be used or disclosed by HIPAA-­covered entities only if (1) an IRB or privacy board waives the authorization requirement (based on standards similar to those governing IRBs’ waiver of the informed consent requirement under the Common Rule), (2) the specimens are released in a de-­identified format, or (3) only a “limited data set” is released (a limited data set consists of general information such as dates of admission and discharge but excludes “facial” identifiers such as name and address).48 The concept of de-­identified information is defined more strictly under HIPAA than it is in the Common Rule. As discussed earlier, OHRP has determined that biospecimens are not identifiable if the identifying information is protected by a code to which the researcher will never be given access. However, the existence of such a code would not be sufficient to render the biospecimens de-­identified for purposes of HIPAA.49 Thus, even though a study using coded biospecimens would not require informed consent pursuant to the Common Rule—because no “human subjects” would be involved— it would require a HIPAA authorization (assuming HIPAA applies to the study).

46  Centers for Medicare and Medicaid Services, Are You a Covered Entity? available at https://www. cms.gov/Regulations-­and-­Guidance/Administrative-­Simplification/HIPAA-­ACA/ AreYouaCoveredEntity.html. 47  Grady C et al, “Broad Consent,” at 34–42. 48  Department of Health & Human Services, Health Information Privacy: Research, available at https://www.hhs.gov/hipaa/for-­professionals/privacy/guidance/research/index.html. 49  Harrell HL, Rothstein MA, “Biobanking Research and Privacy Laws in the United States” (2016) J Law, Med. Ethics 116.

670   Graeme T. Laurie and Carl H. Coleman HIPAA grants individuals a right to access individually identifiable health information about them held by a covered entity.50 Thus, biobanks that are subject to HIPAA must have policies ensuring that individuals who provide specimens have access to relevant information. This includes access to results of testing the specimen “when the laboratory can authenticate that the test result pertains to the patient.”51

4.1.3  State Laws As noted earlier, many states have enacted genetic privacy laws, which typically impose informed consent requirements for disclosure of genetic information. These laws may grant additional privacy protections to individuals whose identifiable biospecimens are stored in a biobank. Some state laws may grant further protections, such as a right to be notified of all uses of a biospecimen or rights to access information held by the biobank.52 State law rights of access to biobank records are particularly important for individuals whose specimens are held by biobanks that are not subject to HIPAA.

4.2  European Union As indicated earlier, the GDPR is now the unifying legal instrument with respect to data protection of personal data in Europe and beyond. It is sometimes forgotten, however, that this operates under the twin principles of both protection and promotion, the latter being concerned with supporting data flows that conform to the appropriate standards. Thus, while there must be a lawful purpose for data processing, there are also exemptions that apply to certain provisions of the GDPR and data subjects’ rights for matters such as historical, statistical, and scientific research. The lawful basis of consent was covered earlier. Other bases include the public interest (for public authorities)53 and the legitimate interests of non-­public authorities.54 In all cases, when special categories of data, such as health data, are held and used for research purposes, it must be shown that this is necessary for those purposes and that the processing is subject to appropriate safeguards.55 This last criterion is not specifically defined in the GDPR, so it falls to member states to lay out what should be covered. Compliance with professional and common law standards of confidentiality are an important element here. Other considerations include whether and when anonymized or pseudonymized data could be used for health-­related research. If this is not possible, then technical and organizational measures should apply to ensure respect of fundamental rights and the data protection principles. In designing and assessing safeguards, a holistic approach should be adopted. All the relevant factors, purposes, and conditions 50  Department of Health & Human Services, Individuals’ Right under HIPAA to Access Their Health Information, available at https://www.hhs.gov/hipaa/for-­professionals/privacy/guidance/access/ index.html 51  Harrell, Rothstein, “Biobanking Research and Privacy Laws,” 114. 52  Hakimian et al, “50 State Survey.” 53  Article 6(1)(e) GDPR. 54  Article 6(1)(f) GDPR. 55  Article 89(1) GDPR.

Biobanks   671 regarding data processing should be taken into account. The effectiveness of these safeguards should be assessed and kept under review. In the research context, Article 9(2)(j) provides that56 [P]rocessing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject.

In such circumstances, key features of the exemption include • Further processing of data for research beyond the original purposes for which they were obtained is possible; • Consent to research-­related processing is not required; • Data subjects need not be given a right of access to their data, and • Data subjects do not have a right to object when the processing is necessary for the performance of a task carried out for reasons of public interest. Numerous European biobanking initiatives have attempted to be transparent about the ways in which they have addressed the new requirements of the GDPR, especially as these relate to privacy protection. Thus, both the BBMRI-­ERIC initiative mentioned earlier and the UK Biobank set up webpages on their home sites to explain their approaches. Consider, for example, UK Biobank,57 which offers the following explanation of the relevant terms of the GDPR and how the biobank has sought to comply: A legal requirement of the GDPR is that we tell you about the legal basis on which UK Biobank will process your personal data. This includes information about you (like your name and address), the information that you told us about (such as answers to questions about your lifestyle), and information that we have received from others (such as your health records). Because of the health-­related nature of the personal information that you have provided to us, there are two lawful bases for our processing of it, which are referred to in the legislation as “legitimate interests” and “consent.” UK Biobank believes that both of these are valid reasons for processing your data and, as is required by the law, we have set our reasoning below.

The reader is invited to visit the site to examine the full reasoning of UK Biobank. As a further indication of transparency, UK Biobank also includes an update on its 56  For a critical appraisal, see Kärt Pormeister, “Genetic Data and the Research Exemption: Is the GDPR Going Too Far?” (2017) 7(2) Int’l Data Priv Law 137. 57  See the UK Biobank Information Notice, available at https://www.ukbiobank.ac.uk/gdpr/.

672   Graeme T. Laurie and Carl H. Coleman i­ nterpretation of the new rights enjoyed by data subjects under the GDPR. These relate specifically to the right to access to one’s personal data, the “right to be forgotten,” and the right to withdraw or request erasure of data. In both respects, UK Biobank invokes the exemptions of the GDPR that relate to scientific research. Thus, the stated position is that participants are not entitled to feedback about their health from analysis of data and samples (this is in conformity with the long-­stated “no feedback” policy of UK Biobank); moreover, given that participants can withdraw from UK Biobank at any time and for any reason, the view is taken that UK Biobank is fully in conformity with the rights relating to erasure and being forgotten. Only a court challenge would provide further clarification on this particular interpretation of the GDPR. It is, however, likely to be typical for all European biobanks; to take a contrary view would make the biobanking endeavor impossible. As a further rejoinder, it is crucial to remember that data protection is only one of a family of legal considerations in Europe that have privacy protection as their object: human rights and national laws on privacy and confidentiality are also in play. Thus, the right to respect for family and private life under Article 8(1) of the European Convention of Human Rights must be considered with the provisions of Article 8(2) that allow legitimate interference by a public authority with private life on a range of public interest grounds so long as any given initiative is necessary and proportionate. Some European states are also signatories to the Council of Europe Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine (Oviedo Convention),58 together with its associated Protocols (notably on Biomedical Research59) and Recommendations.60 National laws will also provide for the protection of confidentiality, and sometimes these go beyond data protection requirements. For example, in England and Wales, uses of confidential information require authorization in the public interest by the Confidentiality Advisory Group (CAG) operating under the National Health Research Authority.61 A register of publicly approved projects is available that demonstrates the kinds of initiatives that have been deemed to be in the public interest. CAG also operates a Precedent Set Review Pathway, whereby similar applications can be streamlined and reviewed in a timely manner if they raise similar issues to those previously addressed by the Group. This is an example of research governance proportionality in action.

58  Council of Europe, Oviedo Convention (1997) ETS no 164. 59  Council of Europe, Additional Protocol to the Convention on Human Rights and Biomedicine, concerning Biomedical Research (2005) ETS no 195. 60  Council of Europe, Committee on Bioethics: Treaties and Recommendations https://www.coe.int/ en/web/bioethics/treaties-­recommendations; accessed 8 October 2018. 61  See s251 of the Health and Social Care Act 2006, as amended, and the Health Service (Control of Patient Information) Regulations 2002.

Biobanks   673

5 Governance As stated earlier, a key feature of the rise of large-­scale prospective research biobanks has been the reflexive turn that has come to typify many governance arrangements. This refers to the act of self-­reflection about one’s conduct, intentions, and actions, and, in governance terms, it is often seen in oversight mechanisms that provide biobank managers with means to engage on and work through novel issues that arise as the biobanking enterprise progresses over time. In this section, we elaborate on key governance issues and offer some examples that illustrate the nature and potential added value of different arrangements.

5.1  OECD Guidelines (2009) Key principles laid out in the OECD Guidelines include transparency and accountability of governance arrangements and management structures. Oversight mechanisms should be designed to deliver on all aspects of the biobanking enterprise, including management and operations and access and use procedures, and should extend to the discontinuation of the biobank itself. Demonstrable provision must be made to comply with legal requirements and ethical principles. Best practices around governance include (i) robust review processes to respond to situations where data and materials are used in ways not anticipated in the original consent; (ii) mechanisms to reflect on the need for, and manner of, any re-­consent; (iii) diverse personnel responsible for governance oversight, selected from a wide range of expertise reflecting the purposes of the collection; (iv) publicly available policies and procedures; and (v) ongoing review and revision of all governance arrangements. Note how regular review of policies and practices is a central feature here. This captures the reflexive turn mentioned earlier; it also demonstrates how governance can be contrasted with law in this context. While laws require compliance and remain relatively static over time (absent formal reform through laborious processes), governance arrangements can be flexible so that they can respond to unanticipated eventualities that do not necessarily implicate a legal response. For example, given that most biobank projects operate on a broad notion of consent, the decision about whether, when, and how to inform participants of new uses of a biobank and perhaps to seek re-­consent will benefit from ethically sound reflection among well-­informed stakeholders and biobank managers.

5.2  United States Biobanks in the United States are characterized by a wide diversity of organizational characteristics. Some exist as part of the federal government, while others are affiliated

674   Graeme T. Laurie and Carl H. Coleman with academic medical centers or private research institutions. A majority have affiliations with one or more other organizations, such as an academic medical center and a hospital. In a 2012 survey of 456 US biobanks, only 5% of responding biobanks were for-­ profit organizations. Regardless of organizational form, 80% of responding biobanks have “internal oversight boards of some kind.”62 Many biobanks in the US rely on a governance structure that includes a combination of IRBs, biospecimen access committees, and community advisory boards. IRBs are responsible for reviewing the risks and benefits of research involving biospecimens and ensuring that researchers adhere to basic ethical principles, such as informed consent and confidentiality. Biospecimen access committees review and approve individual requests for access to specimens and data and “may serve to review the science within the protocols as well as providing stewardship of finite biological samples.”63 Community advisory boards provide ongoing guidance to biobank management. In addition, community boards “can help translate basic discovery into clinical medicine by outlining the needs of both the individual and the community.”64 The Mayo Clinic Biobank, which includes biospecimens donated by Mayo Clinic patients and other volunteers, illustrates this division of responsibilities.65 Its gov­ern­ ance structure was established following a 4-­day deliberative community engagement process, during which 20 citizens, chosen to represent the local population, participated in educational activities and discussions about biobanking and genetics research. The engagement process resulted in a number of specific recommendations about how the biobank should be governed, including the importance of maintaining an active community advisory board.66

5.3 Europe 5.3.1  CoE Recommendation 2016 The 2016 CoE Recommendation largely reflects the OECD Guidelines. It does, however, go further in some particular respects. For example, Article 16 expands on the importance of transparency, extending it to all aspects of governance, quality assurance, and data/ tissue transfer and traceability arrangements, and calling on biobanks to issue public reports on access granted to the resource, including a summary of findings from completed research projects. Article 20 emphasizes the need for independent oversight of any 62  Henderson GE et al, “Characterizing Biobank Organizations in the US: Results from a National Survey” (2013) 5(1) Genome Med 3. 63  Olson JE, Bielinski SJ, Ryu E, Winkler EM, Takahashi PY, Pathak J, et al, “Biobanks and Personalized Medicine” (2014, July) 86(1) Clin Genet 50–55. 64 Id. 65  Mayo Clinic, Mayo Clinic Biobank, available at http://www.mayo.edu/research/centers-­programs/ mayo-­clinic-­biobank/overview. 66  Olson JE et al, “The Mayo Clinic Biobank: A Building Block for Individualized Medicine” (2013, September) 88(9) Mayo Clinic Proc 952–962.

Biobanks   675 proposal to establish a collection. After establishment, each collection should have independent proportionate oversight of its operations, aimed in particular at safeguarding the rights and interests of participants. As a particular example of reflexivity, the ­recommendation provides that “[o]versight mechanisms should be able to adapt to possible evolutions of the collection and of its management.”

5.3.2  UK Biobank Ethics and Governance Framework and Council UK Biobank began as the Population Biomedical Collection toward the end of the 20th century. It was a pioneer in addressing the ethical, legal, and governance challenges of large-­scale collections. Early in its work, an Interim Advisory Group recommended the creation of an Ethics and Governance Framework (EGF) to outline the biobank’s responsibilities to participants and society and for the establishment of an independent Ethics and Governance Council (EGC) to act as a “critical friend” to UK Biobank through its recruitment stage (2007–2010) and beyond into the access phase. UK Biobank now has more than 500,000 participants.67 The EGF adds value beyond mere legal compliance with the myriad regulations that exist in the United Kingdom. As a living instrument, the EGF is designed to evolve over time and with experience. The EGC operated from 2004 until 2018 on an open dialogical basis with UK Biobank to advise on developments and revisions of the biobank and the EGF through the life cycle of the enterprise. The model was one of “reflexive gov­ern­ ance”: “[it] is both about partnership in governance in the face of future uncertainty and the facilitation of mutual learning for experience over time.”68 After years of arm’s-­ length governance, it was mutually agreed between the EGC, UK Biobank, and the funders that oversight could be continued through an advisory body to the UK Biobank Board. This role is now discharged by the UK Biobank Ethics Advisory Committee.69

6 Engagement 6.1  Ethical and Social Imperative The importance of public, community, and stakeholder engagement in biobanks now seems self-­evident, but this is a relatively recent phenomenon. Once again, the example of UK Biobank serves well. Although the funders also sought early public consultation on the idea of establishing such a large-­scale biobank, critics argued that outreach efforts were poorly designed and that stakeholders were not given sufficient opportunity to 67  UK Biobank http://www.ukbiobank.ac.uk/; accessed 8 October 2018. 68  Graeme Laurie, “Reflexive Governance in Biobanking: On the Value of Policy Led Approaches and the Need to Recognise the Limits of Law” (2011) 130 Hum Genet 347, 356, doi: 10.1007/ s00439-­011-­1066-­x; accessed 8 October 2018. 69  https://www.ukbiobank.ac.uk/ethics/; accessed 5 June 2019.

676   Graeme T. Laurie and Carl H. Coleman influence the proposal more directly.70 Since then, extensive engagement exercises have been conducted with a range of biobanks. An important driver behind many engagement initiatives, often reflected in their own findings, is the importance of trust in biobanking itself. Eschewing the hackneyed and inaccurate metaphor of the need to “build trust,” commentators stress the importance of demonstrating trustworthiness.71 Robust community and stakeholder engagement is one important means to do so.

6.2  United States A 2014 study reported the results of interviews about community engagement with 24 people from a diverse group of six US biobanks. While all of the biobanks undertook community engagement efforts, there was considerable variation in the specific practices they used as well as their rationales for using them. “Despite recommendations from federal agencies to conduct community engagement,” the authors concluded, “its meaning is widely interpreted among biobank employees, thus affecting how its goals are perceived and ultimately how it is practiced.”72 Types of community engagement reported in the interviews included establishing community advisory boards, sending out newsletters or publishing articles in local papers, setting up focus groups, and talking with people at community venues, such as churches and schools. An important goal of these activities was to establish relationships in the community and maintain community trust. Other goals included providing education and “understanding public perceptions of the biobank via direct gauge of its ‘temperature.’” One interviewee suggested that biobanks see community engagement processes as a kind of stealth public relations mechanism, enabling the biobank to “monitor the community to ensure that their (currently positive) view of the biobank is not changing.” A key challenge for all biobanks is ensuring that the individuals with whom they engage—whether as members of a community advisory board or at community forums and other events—are genuinely representative of the larger community. In some cases, participants in community engagement activities may have idiosyncratic views that do not always reflect prevailing community attitudes. In addition, “[w]hen bank participants are drawn from very large and diverse populations within a hospital system or geographic area,” they note, “it may be exceedingly difficult to adequately represent the 70  For an account of the processes, see https://egcukbiobank.org.uk/sites/default/files/Public%20 attitudes%20to%20biobank%20and%20related%20ethics%20and%20governance%20issues_0.pdf; accessed 8 October 2018; for commentary, Helen Busby, “Biobanks, Bioethics and Concepts of Donated Blood in the UK” (2006) 28(6) Soc Health & Ill 850, and Oonagh Corrigan O and Richard Tutton, “Biobanks and the Challenges of Governance, Legitimacy and Benefit” in Paul Anthony Atkinson, Peter Egon Glasner and Margaret Lock (eds) Handbook of Genetics & Society: Mapping the New Genomic Era (Routledge 2009). 71  Onora O’Neill, Autonomy and Trust in Bioethics (Cambridge University Press 2002). 72  Haldeman KM et al, “Community Engagement in US Biobanking: Multiplicity of Meaning and Method” (2014) 17(2) Public Health Genomics 84–94.

Biobanks   677 various religious, cultural, and ethnic groups that make up the communities from which donors are recruited.”73 The University of California’s Deliberative Community Engagement (DCE) project, completed in 2013, illustrates one approach to community engagement by a major US biobank. The goal of the project was “to engage input from ordinary Californians about how to govern biobanking at UC in order to assure ongoing public trust and engagement in UC translational research.” Following the principles of the DCE method, project organizers recruited a sample of relevant stakeholders to serve as participants, engaged in educational activities regarding both technical and social considerations, facilitated discussions in which participants were encouraged to clarify their own values and to understand others’, and developed specific recommendations related to issues important to the group. The project yielded specific recommendations on a variety of issues related to biobank governance, including policies on sample and data sharing, informed consent, and return of research results. In addition, a majority of participants supported the ongoing involvement of lay community representatives in biobank governance.74

6.3  European Union The 2010 Eurobarometer on biotechnology showed that two-­thirds of those interviewed had not heard of biobanks, and only 17% could be said to have engaged actively on the subject, for example, by seeking information or via discussions.75 However, once asked, almost one in two Europeans said they would participate in a biobank project. A report from the Working Group on Ethical, Legal and Social Issues (ELSI) relating to BBMRI had one clear over-­arching message: “[i]n order to be successful in the long term, biobanks must engage with the public.”76 Three examples raise valuable points for the future success of biobanks. • Engagement with publics and participants involved in the Generation Scotland project—an initiative similar to UK Biobank but concentrating on families— showed that concern about a profit dimension in biobanking is not necessarily a barrier to involvement or trust. Rather, it is the prospect of excess profit that causes most concern. Moreover, policies that involve benefit sharing to the wider community might go far in alleviating any such concerns.77 73 Id. 74  Dry SM et al, “Community Recommendations on Biobank Governance: Results from a Deliberative Community Engagement in California” (2017, February 24) 12(2) PLoS One e0172582. 75  European Commission, Europeans and Biotechnology in 2010—Winds of Change? (European Commission Directorate-­General for Research 2010) 61. 76  Available at http://www.bbmri-­eric.eu/wp-­content/uploads/BBMRI-­Biobanks-­and-­the-­Public. pdf, 5; accessed 8 October 2018. 77  Gill Haddow et al, “Tackling Community Concerns About Commercialisation and Genetic Research: A Modest Interdisciplinary Proposal” (2007) 64 Soc Sci Med 272.

678   Graeme T. Laurie and Carl H. Coleman • A German postal survey on public attitudes toward biobanks and participation showed that levels of awareness remain low, harboring serious challenges both for future engagement exercises and biobanks themselves.78 • An EU Horizon 2020 Project surveyed operators of European biobanks to understand perceptions of the legal and regulatory requirements for return of research results to participants. Respondents, who were drawn from 13 countries participating in BBMRI-­ERIC,79 showed varied understandings of the relevant laws and regulations, as well as confusion about how to interpret them.

7 Access It is been nearly two decades since the earliest large-­scale biobanks were established, which means that many biobanks are well into their access phase. As we move further from set-­up and considerations of legal compliance, the need for robust and transparent access policies becomes all the more important to biobanks’ long-­term success. If trust is undermined, participants might eventually decide to withdraw their specimens, representing the end to many years of investment and hard work, not to mention a serious betrayal of participants’ contributions. Equally, the pressures on biobank operators and researchers seeking access are considerable. “Proportionality” is an important watchword in this context. Unduly burdensome and bureaucratic access processes will seriously damage research and undermine public benefit. Seeking a balance of interests is crucial, but so, too, might be some degree of reciprocity, either in the form of benefit sharing or in the further opening up and sharing of resources. One review of US-­based biobanks identified three types of access models: open access (unrestricted access by anyone through mechanisms such as a public website), controlled access (access only to certain approved researchers and protocols), and tiered access (different degrees of access based on factors such as “donor consent, the nature and content of the data or specimens, or the type of research use of the specimens and data”). Almost half of biobanks reviewed required IRB approval of research projects, and many had panels or committees to review specimen requests. More than half of the biobanks rejected at least some requests for specimens, “most commonly because of a lack of scientific merit, unavailability of specimens, or donor limitations.”80 78  Sabine Bossert, Hannes Kahrass and Daniel Strech, “The Public’s Awareness of and Attitude Toward Research Biobanks—A Regional German Survey” (2018) 9 Front Genet 190, doi: 10.3389/ fgene.2018.00190; accessed 8 October 2018. 79  Minna Brunfeldt et al, “Perceptions of Legislation Relating to the Sharing of Genomic Biobank Results with Donors: A Survey of BBMRI-­ERIC Biobanks” (2018) 26(3) Eur J Hum Genet 324, doi: 10.1038/s41431-­017-­0049-­3; accessed 8 October 2018. 80  HL Harrell and MA Rothstein, “Biobanking Research and Privacy Laws in the United States” (2016) 44(1) J Law, Med. Ethics 106–127, doi:10.1177/1073110516644203.

Biobanks   679 The 2018 Revision to the International Society for Biological and Environmental Repositories (ISBER) Best Practices for Repositories calls on repositories to • Develop data sharing policies describing what type of specimen (or sample) associated data can be accessed and provided. • Establish policies and procedures for acknowledgment in publications and for reporting requirements about the use of the specimens and the data provided by the repository; these requirements should be specified in Material Transfer Agreements. • Ensure that, when specimens and data are imported from abroad, fair and equitable benefits are made available to the providing country. With respect to data, the establishment of METADAC in the United Kingdom is a good example of economies of scale and efficiency in access decision-­making. The METADAC infrastructure—for Managing Ethico-­social, Technical and Administrative issues in Data ACcess—is the United Kingdom’s only independent data and sample access platform. An ethnographic study involving key actors within the infrastructure identified three structural features as the foundation for responsible data sharing: independence and transparency, interdisciplinarity of members contributing from a range of perspectives, and participant-­centric decision-­making.81 A recent European analysis found some differences in approach among biobanks’ access policies with respect to issues such as application processes, eligibility criteria, the conditions for REC approval, and the terms of agreements on sharing materials and data.82 High-­level principles are, however, the same: access for research in the public interest, trustworthy applicants, scientific merit, and conformity with the biobanks’ operating objectives. As such, the next step is to consider closer approximation of arrangements, such as a single centralized application process or a model of mutual recognition of approval by one access body in the consortium. This having been said, some empirical evidence suggests that both local and global considerations around access must be carefully balanced in a context-­specific way. For example, the Generation Scotland biobank was established at much the same time as UK Biobank and with a similar objective: to create a resource for health-­related research. However, Generation Scotland differs in two important respects: first, it recruited genetically related families rather that a cross-­section of the population, and, second, its mission statement has always been determinedly about improving the health of Scotland. This has created challenges around access, both in managing expectations of family members that benefits might accrue to them directly, and around the national message about how the resource will be managed—an early policy statement said that samples 81  M Murtagh et al, “Better Governance, Better Access: Practising Responsible Data Sharing in the METADAC Governance Infrastructure” (2018) 12 Hum Genomics 24, doi: 10.1186/s40246-­018-­0154-­6. 82  Jane Kaye et al, “Access Governance for Biobanks: The Case of the BioSHaRE-­EU Cohorts” (2016) 14(3). Biopreserv Biobank 201, doi: 10.1089/bio.2015.0124; accessed 8 October 2018.

680   Graeme T. Laurie and Carl H. Coleman would not leave domestic shores. This has had a direct impact on access policies and practices, as well as (re)consent processes to delicately balance local participants’ protection with wider, global aspirations to maximize the value of the resource.83 Access management strategies to emerge have included (re)consent of around half the participant cohort to allow samples to leave the United Kingdom, requiring additional research ethics committee approval and delays in access as a result. More collaboratively, the Generation Scotland Access Committee and related management bodies now require there to be some form of collaboration agreement between Generation Scotland and researchers seeking access, both to manage the resource and to realize benefit in keeping with the original terms of reference.

8 Conclusion Biobanks represent particularly acute challenges to health research regulation. However, while there is a diversity of approaches across jurisdictions—and particularly between the United States and Europe—there is also much scope for common lessons to be learned about what counts as good governance. As this chapter has demonstrated, no biobank is above the law, and, despite quite different legal regimes governing consent and privacy protection as between the US and Europe, the initial challenges thought to be created for law by the advent of biobanking have been accommodated. Few jurisdictions have felt the need to legislate specifically for biobanks, and there is no evidence of which we are aware that suggests there is any particular advantage or disadvantage of doing so. Rather, reflecting the solidaristic ethic that is found in many European health- and public health-­related initiatives, there are multiple examples that promote biobanking as a collective enterprise, such as the CoE Recommendation and the European Commission funding of BBMRI-­ERIC. The collective spirit is also reflected in US initiatives like the All of Us project, in which a significant amount of public resources have been invested to promote access to a research cohort of at least 1 million Americans, with a particular emphasis on individuals from historically underrepresented groups. A key feature of much biobanking practice that has emerged, however, is the distinction between law and governance. As demonstrated earlier, there are multiple international examples of agencies seeking to lay down generic principles for good biobanking practice that are designed to sit on top of—and in addition to—local legal requirements. Importantly, these do not merely relate to aspirational ethical standards, but also are often designed with attention to technical interoperable standards that will facilitate the international sharing of data and samples. Examples include the International Society for Biological and Environmental Repositories (ISBER) Best Practices for Repositories and the infrastructure funding for BBMRI-­ERIC, which is based on international 83  C Heeney and SM Kerr, “Balancing the Local and the Universal in Maintaining Ethical Access to a Genomics Biobank” (2017) 18(1) BMC Med Ethics 80, doi: 10.1186/s12910-­017-­0240-­7.

Biobanks   681 collaboration. More widely still, initiatives such as the Global Alliance for Genomic Health (GA4GH)84 has adopted a multiprong governance and data sharing framework—based around international human rights and the right to participate in science—that arguably acts as the best possible incentive to biobankers to share their resources. GA4GH does not mandate a compliance culture, but rather it provides technical, ethical, regulatory, and governance tools to make cooperation easier by providing a lingua franca of standards for international biobanking. Its success is measured in large part by its membership of more than 200 organizations. Another vital benefit from such initiatives is that they help to promote and check the quality of data and samples derived from biobanks, of which there is considerable variability. At the more local level—where regional and national laws begin to bite—more research is needed on whether and how far particular approaches to consent, privacy, or access act as incentives or disincentives to biobankers seeking to set up or collaborate with colleagues in a particular country. Despite the global aspirations held by many for open sharing in biobanking, what evidence does exist suggests that access policies cannot be blind to local considerations. Once again, these transcend the purely legal to include the expectations of participants themselves. Indeed, this last point raises an issue that is of universal concern to all who are involved in the biobank enterprise: the crucial importance of trustworthiness. Whatever any given law in a particular country might permit, the long-­term and unceasingly uncertain enterprise of biobanking relies entirely on the ability of the biobankers to demonstrate their trustworthiness to their participants and society more generally. This chapter has sought to explore some of the key elements that are common in addressing that challenge, including who is a stakeholder in a biobanking initiative and how diverse interests can be acknowledged and accommodated.

84  See https://www.ga4gh.org/.

PH A R M AC EU T IC A L AND MEDICAL D E V I C E S L AW

chapter 33

I n troduction to M edica l Products L aw Erika Lietzan, Aurélie Mahalatchimy, and Patricia J. Zettler

Policymakers around the world have long recognized the need to regulate medical products, meaning interventions intended to prevent, diagnose, treat, or cure diseases and other health-related conditions. Generally speaking, these interventions can be subdivided into medicinal products (sometimes known as drugs or medicines, and hereafter “medicines”) and medical devices, although the distinctions between these categories can be blurred and some commercialized products comprise both. In most of the world, the rules for these two broad categories differ somewhat, although many of the basic principles are the same. Furthermore, these basic principles inform medical products law throughout the developed and developing world. At the same time, the regulation of medical products can present a variety of ethical, legal, and social issues that countries will choose to resolve differently. As a result, even in the developed world, regulation of medical products can differ in important ways from one country to the next. The four chapters that follow explore medical products regulation in the United States and the European Union. We consider several broad themes, including the challenges of approving products with inherent risks, some of which are unknown and possibly unknowable; the challenges of ensuring that the regulation of medical products encourages the discovery and development of treatments that address pressing public health needs; the need to find an appropriate level of regulation that does not impose more burdens or barriers on innovation or access to the market than necessary; and the desire to ensure robust competition on the basis of price as well as (where applicable) accurate information about differing product features. With respect to each jurisdiction, we consider recent controversies and ongoing debates as well as product categories that have presented special difficulties for policymakers. These include human cell- and tissue-based products, gene therapy products, genetic testing, and 3D printing of medical devices.

686   Erika Lietzan, Aurélie Mahalatchimy, and Patricia J. Zettler Over the past century, regulators in the United States and in the European Union have developed complex frameworks for regulation of medicines and medical devices marketed within their borders. In both jurisdictions, these frameworks have evolved in response to public health crises, such as discoveries that marketed products were not nearly as safe and effective as previously assumed, as well as genuine scandals—that is, serious situations in which a product’s manufacturer bears fault. To give a well-known example, the primary US law governing medical products dates to 1938—one year after an antibiotic mixed with a sweet but toxic solvent (without any malicious intent) killed more than 100 people in the United States. Major medicines legislation in the United States in 1962 and in the European Union in 1965 followed on the heels of the thalidomide tragedy—after widespread use of this treatment for morning sickness led to severe birth defects in roughly 10,000 babies, mostly in the European Union, Australia, and Japan. And the discovery of a French company illegally manufacturing and selling breast implants made of industrial rather than medical-grade silicone, and that were associated with increased risks of rupture, prompted changes in European rules governing medical devices. Other scandals, which have similarly informed the development of medical products law in both jurisdictions, have involved the claims made by companies about their products. Some have involved fraudulent or otherwise unsupported claims of cure—such as drugs marketed in the United States in the early twentieth century as effective in curing cancer, even though the companies knew the claims were false. In light of this history, medical products regulation in both jurisdictions rightly focuses on protection of the public health. Generally speaking, both jurisdictions seek assurance that medicines—drugs and biologics in the United States, medicinal products in the European Union—are safe and effective for their intended purposes. Both require medical devices to be safe, and the United States also requires an assurance of medical device effectiveness. Moreover, in both jurisdictions, the requirement to establish safety (and, where applicable, effectiveness) applies before product launch. Thus many medical products must be the subjects of extensive applications presenting the results of studies performed and asking permission to launch from regulatory agencies (mainly the European Medicines Agency [EMA] and national medicines agencies in the European Union and the US Food and Drug Administration [FDA] in the United States). For other medical products (medical devices in the European Union), the conformity to essential legal requirements has to be established by the manufacturers, sometimes via the competent authorities’ authorizations. To ensure product safety, both jurisdictions also impose “quality” requirements, a prerequisite which generally means that the manufacturer must tightly control the manufacturing process in order to prevent errors and ensure consistency in product characteristics: ensuring product quality means, in part, ensuring its conformity to the claimed formula or design. In addition to protecting the public health, EU law also aims to create a single European market for both medicines and medical devices through generally prohibiting certain barriers to trade and ­promoting the free movement of goods, capital, services, and persons across Member States’ boundaries.

Introduction to Medical Products Law   687 Where both safety and effectiveness must be assessed prior to marketing, a medical product may be launched only if its benefits outweigh its risks, based on the information available at the time. This broad principle reflects two facts: first, that no medical product is perfectly safe or perfectly effective, and, second, that a product’s safety and effectiveness cannot be fully understood on the basis of any reasonably sized premarket research program. Risks may be inherent in a product, or they could result from external factors such as misuse (or use in a way that was not studied or intended). The risks acceptable in a particular medical product—and the level of uncertainty about risk that is acceptable—will often depend on the expected benefits of the product. It may also differ over time or, indeed, depend on the jurisdiction. And while it might be possible to generate additional information about a medical product’s safety (and, where relevant, effectiveness) prior to launch, requiring additional study imposes costs of its own in the form of delay in treatment and expenses which must be borne by the payer in the end. In both jurisdictions the medical products regulatory framework at any given time reflects the then-acceptable level of uncertainty about the safety and effectiveness of particular interventions given the cost and delay that would be involved in generating additional evidence. Further reflecting the challenge of striking the right balance between ensuring further study, on the one hand, and providing patients with access to even with incomplete information, on the other, each jurisdiction has adopted programs to expedite the availability of medical products for patients with serious or terminal illnesses who lack satisfactory treatment options. Nor do the risks end with product approval: a medical product may trigger adverse events (anticipated or even unanticipated) even when used properly, just as it might trigger adverse events when used incorrectly. The laws of both jurisdictions also seek to manage the risks of medical products after approval through regulatory requirements and (where applicable) liability rules. The necessary premarket research and development required in both jurisdictions not only delays patient access to new medical treatments, but it also imposes costs and risk on innovators. These costs must be recovered during the marketing of the resulting product. High cost and risk can also discourage innovation, particularly if there are questions about the recovery of research and development costs. In some cases, cost and risk might steer researchers toward less risky projects or projects that are more likely to produce commercially successful products. Policymakers in both jurisdictions therefore seek to ensure that regulatory requirements applicable to medical products do not impose undue barriers that might operate as an unnecessary disincentive to innovate. This requires striking a balance between ensuring quality and robust premarket testing, on the one hand, and ensuring manageable costs and time investment, on the other hand. Similar considerations apply to post-market regulation, which can discourage post-market innovation (such as the development of newer manufacturing techniques) if unduly burdensome. The laws of both jurisdictions also provide incentives for innovation—patents for new and useful inventions, temporary protection of testing data, and special regulatory incentives intended to encourage specific types of research (such as exclusivity for development of treatments for rare diseases)—although the data protection and regulatory exclusivity rules differ. At a high level, the principles of patent law

688   Erika Lietzan, Aurélie Mahalatchimy, and Patricia J. Zettler are harmonized between the two jurisdictions. Although the chapters that follow do not discuss patent law in any detail, we briefly touch on the special relationship between patents and US and European drug regulatory requirements. Policymakers in both jurisdictions, as elsewhere in the world, must also strike a balance between encouraging innovation and ensuring a steady supply of newer medical treatments, on the one hand, and ensuring that available medical treatments are in fact affordable for payers and patients, on the other hand. With respect to medicinal products, regulators in the United States and Europe facilitate the affordability of medical treatments by approving copies (generic drugs and biosimilar biologics, known as “similar biological medicinal products” in Europe) on the basis of smaller applications that are less expensive and time-consuming to prepare. Patents, data protection, and targeted incentives encourage innovation by providing a period of exclusivity in the ­market, before these copies can be approved. For instance, a patent protecting the active ingredient of a new medicinal product and protection of the testing data that supported approval of the medicinal product can work together to preclude copies of the product until both the patents and data protection expire. As a result, although robust incentives ensure a steady stream of new medical products, they inherently delay the introduction of lower cost alternatives. Of course, pricing and reimbursement rules also play a role in patients’ access to medical treatments and may address affordability concerns, including during the exclusivity period for medicines. Patients’ access to a medicine or device is a function both of whether the relevant requirements for regulatory authorization have been met and whether patients can afford the treatment. But pricing and reimbursement rules are complex and differ substantially between the United States and Europe, as well as between the Member States of the European Union (which implement their own heterogeneous healthcare systems), and these are outside the scope of these chapters. Although innovators may recover their research and development costs during the exclusivity term through higher prices than are available after generics and biosimilars reach the market, medical products may experience some competition from other innovative medical products during exclusivity periods. The medical products laws of both jurisdictions address the functioning of this competitive marketplace, focusing on protection of consumers from economic harm and protection of the marketplace from unfair competition, although other regulators may also play a role. We will focus mainly on US law, which requires, for instance, that manufacturers provide truthful and non-misleading information about their medical products, whether marketing to healthcare providers or directly to patients. European law prohibits direct-to-consumer advertising of prescription drugs, to protect the public health. Competition (antitrust) authorities also play a role in both jurisdictions, and those laws are beyond the scope of these chapters. Continuing innovation in medical products—particularly significant innovations that fundamentally change the mode of intervention—can challenge existing regulatory programs. Over the past century, the medical products laws of both jurisdictions have evolved and adapted to new medical knowledge and new scientific capabilities. For instance, after the biotechnology revolution of the 1970s and 1980s, regulators in the European Union and the United States adapted their approval processes to the special

Introduction to Medical Products Law   689 issues presented by recombinant protein products. The sequencing of the human genome in 2001 has made it possible for researchers to develop pharmacogenetic medicines and, recently, gene therapies. Introduction of the personal computer in the 1980s began an electronic and engineering technology revolution that has given rise to mobile health “apps” (“e-health”) as well as 3D printing, among other things. Advances in microbiology have led to cellular-based therapies and may also lead to microbiomebased therapies. These newer treatment modalities—which could also be called “medical technologies”—may not fit neatly within existing medical products law. Policymakers and regulators must determine whether existing laws and programs can be adapted to oversee novel technologies or whether new laws and regulatory approaches are needed. In the four chapters that follow, we explore the principles that guide medical products law in the European Union and the United States as well as the approaches taken by regulators in the European Union and the United States to reconcile these differing goals of medical products oversight. We devote two chapters to each jurisdiction: one for medicines and one for medical devices. Although many of the overarching principles for regulation are the same, there are significant differences between the medical device laws of the two jurisdictions, and each jurisdiction treats medicines and medical devices somewhat differently. Within each chapter, we explain the threshold concepts and principles governing regulation, the legal rules governing access to the market, the rules relating to innovation and competition, and the regulatory approach to risk management. As a general rule, we focus on the work of the European Medicines Agency (EMA) for medicines and the US Food and Drug Administration (FDA) for both medicines and medical devices. Because, unlike the FDA, the EMA has no role in the regulation of medical devices, we focus on the European Union’s reliance on balancing the roles of private and public actors in regulated medical devices. Additionally, further differences between the European and US chapters result from the diverging political structures of the jurisdictions. Where relevant, therefore, we note the role of state law in the United States or the law of individual Member States in the European Union. Each chapter concludes with a discussion of innovative medical treatments—such as 3D printing, cell and tissue therapies (known with gene therapies as advanced therapy medicinal products in the European Union)—that challenge the current paradigm. We use these examples to explore what the regulatory approach to recent and emerging breakthrough innovations suggests about the future of medical products law.

Chapter 34

R egu l ati ng M edici n es i n the U n ited States Erika Lietzan and Patricia J. Zettler

This chapter and the chapter on “Regulating Medical Devices in the United States” (Zettler and Lietzan, this volume) illustrate that although the details of the US schemes for regulating medicines and medical devices differ, the core goals of medicine and device regulation are the same. US law aims to ensure that medical products—whether medicines or medical devices—are safe and effective for their intended uses, that they become available for patients promptly, and that companies manufacturing and marketing these products (whom we call “manufacturers” in these chapters) provide truthful, non-misleading, and complete information about the products.1 European medicines law has similar objectives although the phrasing is slightly different.2 Individual states play a role in regulating medical products, most notably through establishing rules for product liability.3 But the US scheme for achieving the primary goals of medical product regulation is largely federal—governed by the Federal Food, 1  See, e.g., 21 USC. § 393. 2  For instance, the overall goal of the European medicines framework is usually said to be ensuring the safety, efficacy, and quality of medicines. Quality is a broad concept that refers generally to the identity, strength, purity, and similar characteristics designed to ensure that a medicine reaches the required levels of safety and effectiveness. Generally, in the United States, policymakers simply refer to ensuring the safety and effectiveness of medicines. In fact, though, US medicines law places the same high emphasis on product quality as European medicines law. 3  See, e.g., Patricia J. Zettler, Pharmaceutical Federalism, 92 Ind. L.J. 845 (2017). In some situations, however, state law may not impose liability on medicine manufacturers. The US Constitution’s “supremacy clause” makes clear that when state and federal law conflict, federal law preempts state law—that is, the state law has no effect. US Const. Art. VI, cl. 2. The US Supreme Court has held that the FDCA impliedly preempts many claims against generic medicine manufacturers, and, at the time of writing, a new case regarding the preemption of claims against brand-name medicine manufacturers is pending before the Court. Mut. Pharm. Co. v. Bartlett, 133 S. Ct. 2466 (2013); PLIVA, Inc. v. Mensing, 564 US 604 (2011); Merck Sharp & Dohme Corp v. Albrecht, 138 S. Ct. 2705 (2018). Preemption of claims against medical device manufacturers are discussed briefly in the chapter by Zettler and Lietzan, “Regulating Medical Devices in the United States,” in this volume.

692   erika lietzan and patricia j. zettler Drug, and Cosmetic Act (FDCA) and two provisions of the Public Health Service Act (PHSA), administered by the US Food and Drug Administration (FDA), a federal executive branch agency within the US Department of Health and Human Services. The FDA accomplishes its mission by regulating medical products throughout their lifecycle, from the early stages of research and development through their use in m ­ edical care. Thus federal law generally requires research before a medical product enters the market, FDA authorization before market launch affirming that there is sufficient evidence of the product’s safety and effectiveness, and continued monitoring and regulation after launch.4 These requirements and how they are implemented, however, reflect values and goals that, in some circumstances, can be in tension. This chapter examines how the regulatory scheme for medicines seeks to reconcile these sometimes competing values and goals. It starts by describing what constitutes a medicine subject to FDA regulation and then explains how medicines reach the market, how the FDA and manufacturers assess and manage risks, and the incentives for innovation and rules for competition. The chapter concludes by focusing on innovative medical technologies that challenge the traditional US regulatory scheme, using those examples to explore the future of medicines regulation.

1  Defining Medicines The FDA’s medicines regulatory scheme applies to items that travel (or have a component that travels) in interstate commerce and that meet the definition of a “drug” or “biological product” in US federal law. A “drug” is an “article” “intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease” or “intended to affect the structure or any function of the body.”5 Thus, the “intended” use of an item—most often demonstrated by a manufacturer’s statements about the item—is generally the key to determining whether it constitutes a medicine.6 This means, for example, that when a 4  The relationship between the states and the federal government in the United States with respect to medicines law is fundamentally different from the relationship between the Member States of the European Union, on the one hand, and European Union institutions, on the other hand. For instance, in Europe, products liability is primarily governed by a European Directive. Also, there are numerous pathways to market in Europe, only one of which involves review by the European Medicines Agency, the equivalent of the US FDA. See Mahalatchimy, “Regulating Medicines in the European Union,” in this volume. 5  21 USC. § 321(g). Food, which of course is intended to affect the structure of function of the body, is expressly excluded from the definition of a drug. Id. 6  21 C.F.R. § 201.128. In rare instances, the FDA has concluded that an item is intended for use as a drug based not on the manufacturer’s statements but instead on the circumstances surrounding the item’s distribution, which can include the manufacturer’s knowledge of how physicians, patients, or consumers actually use the product. The agency’s authority to rely on certain evidence other than a manufacturer’s statements is, at the time of writing, controversial. Clarification of When Products Made or Derived From Tobacco Are Regulated as Drugs, Devices, or Combination Products; Amendments to Regulations Regarding “Intended Uses”; Partial Delay of Effective Date, 83 Fed. Reg. 11639 (Mar. 16, 2018).

Regulating Medicines in the United States   693 charcoal manufacturer markets its product as a fuel, the charcoal is not a medicine and is outside FDA jurisdiction; but when a manufacturer markets charcoal as an emergency treatment for poisoning, it is an FDA-regulated medicine.7 Also, because “drugs” can be items intended to affect the structure or function of the body, drugs can include both items commonly understood to be medicines—such as antibiotics—as well as items that may not have therapeutic purposes and may not be commonly thought of as medicines—such as nonprescription products that men use to regrow lost hair. Within this broad drugs category, certain types of articles—including vaccines, viruses, proteins, therapeutic serums, and analogous products—also meet the definition of “biological product” in federal law.8 Modern biotechnology-derived therapeutic protein products, such as monoclonal antibodies administered to treat cancer, are biological products as well as drugs under US law. Gene therapies and certain products composed of human cells and tissue, likewise, generally are both biological products and drugs.9 A biological product that is also a drug is regulated as a drug and thus (for example) must be shown safe and effective before market launch. The requirements for drugs and biological drug products are largely the same, and, in this chapter, we use the term “medicines” to include both.

2  Bringing a Medicine to Market US law charges the FDA with two goals related to bringing a new medicine to market: ensuring that marketed medicines are safe and effective and helping to make medicines available promptly.10 These goals are sometimes in tension. We require FDA permission to market medical products on the theory that this will protect patients from unsafe and ineffective products, as well as on the theory that it will ensure the production of information about the products that healthcare professionals and patients need.11 But developing clinically meaningful information about a medical product’s benefits and risks can take a long time. Premarket testing requirements inherently delay patient access to safe and effective medicines. Indeed, patients who face serious or terminal conditions are sometimes willing to forgo the greater certainty that comes with extensive testing and may push for earlier access to specific medicines. Stakeholders and scholars have long debated whether the FDA sets the bar for evidence of safety and effectiveness too 7 FDA Memorandum: Public Health Interests and First Amendment Considerations Related to Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products 4–5 (Jan. 2017), https://www.regulations.gov/document?D=FDA-201 6-N-1149-0040 [https://perma.cc/FTD8-LRES]. 8  42 USC. § 262(i). 9  Some human cell- and tissue-based products are not regulated as drugs. We discuss this further in section 5. 10  21 USC. § 393. 11  See, e.g., Rebecca S. Eisenberg, The Role of the FDA in Innovation Policy, 13 Mich. Telecomm. Tech. L. Rev. 345 (2007).

694   erika lietzan and patricia j. zettler high, thus unduly delaying patient access, or, instead, too low, thus permitting ineffective or unsafe medicines on the market. The inherent tradeoff between greater certainty and earlier access has prompted policymakers in the United States to make many changes to US law over the years. Today the FDA’s laws and regulations provide several pathways and incentives to bring new medicines (or medicine-like products) to the market, all of which seek to find the right balance between robust evidence of safety and effectiveness, on the one hand, and timely patient or consumer access, on the other hand.

2.1  Standard Approval Process for New Medicines Before new medicines—medicines that are not “generally recognized as safe and effective” for the uses recommended in their labeling—may enter the market, the FDA must approve them.12 A new drug may not be marketed until the FDA approves a new drug application (NDA) demonstrating that the medicine is safe and effective for its intended use.13 A biological drug product may not be marketed until the FDA approves a biologics license application (BLA) showing that the medicine is safe, pure, and potent—a standard generally considered to mean safe and effective for its intended use.14 Medicine manufacturers generate the safety and effectiveness data needed for an application to the FDA in stages, typically beginning with laboratory and animal (“preclinical”) testing. Preclinical testing gathers basic information about a medicine to determine whether research with human subjects may begin. The FDA regulates preclinical testing through its “good laboratory practice” regulations primarily to ensure the scientific validity of data.15 Other agencies have primary responsibility for regulating the risks to animals and laboratory workers.16 If the results of preclinical testing suggest conducting clinical trials in humans would be scientifically worthwhile and ethically permissible, the manufacturer submits an investigation new drug (IND) application to the FDA.17 The agency reviews the IND to ensure that the research includes appropriate protections for subjects and to help ensure that study designs are adequate to permit evaluation of the medicine’s effects.18 FDA ­regulations also impose many obligations on the company responsible for a clinical trial (called “the sponsor”) and on a trial’s investigators, including the obligation to obtain informed consent from research subjects and permission from an institutional review 12  21 USC. §§ 321(p), 331(d), 355(a). To avoid categorization as a new medicine requiring approval, the medicine also must have been marketed to a material extent and for a material time for its intended use. 21 USC. § 321(p). To satisfy that standard, the FDA generally requires that the medicine have been legally marketed in sufficient quantities for at least 5 continuous years. FDA, Guidance for Industry: Time and Extent Applications for Nonprescription Drug Products (Sept. 2011). 13  21 USC. § 355(d). 14  42 USC. § 262(a). 15  21 C.F.R. pt. 58. 16  See, e.g., 7 USC. §§ 2131-2159; 42 USC. § 351A. 17  21 USC. § 355(i); 21 C.F.R. §§ 312.1, 312.20. European medicines law similarly requires a clinical trial authorization application. See Mahalatchimy, “Regulating Medicines in the European Union,” in this volume. 18  21 C.F.R. § 312.22.

Regulating Medicines in the United States   695 board (IRB) to conduct the research.19 Once the IND is authorized—30 days after the FDA receives it, unless the agency tells a sponsor otherwise—and research starts, the agency may stop the clinical trials if certain concerns arise, including that subjects are exposed to unreasonable or significant risks.20 FDA regulations describe three phases of premarket clinical trials, with research generally starting in small groups of healthy subjects and moving to large controlled trials in subjects with the relevant disease as more is learned about the medicine’s effects.21 The traditional approach is to first introduce a medicine into humans in small phase 1 studies of approximately 20–80 healthy subjects, to gather basic information on safety and how humans process the medicine. Phase 2 studies involve larger groups of subjects who often have the disease being studied, and these are designed to produce preliminary evidence of effectiveness as well as additional safety information. Phase 3 studies are large, controlled trials with hundreds or thousands of subjects that are designed to produce substantial evidence of the medicine’s effect on a clinical outcome that is sufficient to support approval. In practice, however, in light of the expense and time involved with conducting this research and the pressure to speed development of important medicines, modern clinical trials are more varied. Clinical trials may be designed to combine aspects of different phases, to proceed seamlessly from one phase to the next, or, in some circumstances, including for certain truly novel biological medicines, to start with a “phase 0” trial of a very small dose in just a few subjects. Regardless of the precise design of the research program, at the end of this process the manufacturer submits to the FDA a marketing application describing the medicine, its manufacturing process, and the data establishing that it is safe and effective.22 Like European law, however, US law recognizes that no medicine can be perfectly safe and effective. All medicines are associated with risks, and some risks probably cannot be uncovered during a reasonable premarket research program. Moreover, no medicine is equally effective in all patients. FDA approval of a new medicine therefore reflects the agency’s judgment call that the currently available evidence indicates that the medicine’s benefits outweigh its risks. The FDA also does not weigh the medicine’s benefits and risks in some general sense. Instead, it assesses the medicine’s safety and effectiveness for the use described in its labeling—to treat a particular disease or condition, in a specific patient population, in a specified dose and dosage form. Thus when the FDA approves a new medicine, it approves that medicine for the specific use described in the labeling based on the positive benefit-risk balance for only that use.23 And in some cases the use for which there is a positive balance might be quite narrow—only patients of a particular 19  21 C.F.R. pts. 50, 56, 312. European medicines law similarly requires review by an ethics committee. See Mahalatchimy, “Regulating Medicines in the European Union,” in this volume. 20  21 C.F.R. § 312.42. 21  21 C.F.R. § 312.21. As in Europe, clinical trials performed after marketing approval are considered “phase 4” trials. 22  21 C.F.R. § 505(d). 23  The FDA does not have statutory authority to consider the cost-effectiveness of a treatment when making this decision.

696   erika lietzan and patricia j. zettler age who have already failed three other specified treatment regimens, for instance, or only patients with a particular condition and genotype. Public health stakeholders, and the FDA in particular, like its counterparts in Europe, are typically willing to accept greater risks with greater benefits. For example, a medicine that effectively cures 5% of patients with metastatic pancreatic cancer but that is also associated with a significant risk of death might still have a favorable benefit-risk ratio because survival rates for pancreatic cancer are so low.24 If the same risk were associated with a different medicine intended to treat minor aches and pains, however, the benefit-risk ratio would not be favorable. And, if the same medicine also treated minor aches and pains, the benefit-risk ratio for that second use would not be favorable. Once the FDA approves a medicine as safe and effective for a specific use, however, healthcare professionals generally may prescribe the medicine for any use, including unapproved uses known as “off-label uses.”

2.2  Expediting Access The standard approval process, while providing the benefit of producing robust information about a new medicine, takes significant resources and time. One study estimated that, for new chemical entities, the process takes an average of 8 years from the start of clinical trials to FDA approval.25 The lengthiness of the process poses particular challenges for patients with serious or terminal illnesses who do not have adequate treatment options among marketed therapies and who may want to try medicines with uncertain risks and benefits. Over the past several decades, the FDA and other policymakers have developed numerous programs to expedite access to important new medicines still under investigation, balancing the need for evidence of safety and effectiveness, on the one hand, with the needs of desperately ill patients and a strong US culture of respecting individual autonomy to assume risk, on the other hand.26 They generally fall into two categories: programs to shorten the research or approval process and programs to provide patients with unapproved medicines. The chief program for shortening the research and development process—accelerated approval—changes the data requirements for approving medicines intended to treat a serious or life-threatening condition.27 Under this program, a manufacturer may demonstrate that its medicine has an effect on a surrogate or intermediate clinical endpoint that is reasonably likely to predict clinical benefit rather than providing direct evidence of a clinical benefit. For instance, the FDA often permits manufacturers studying 24  Henry T. Greely & Judy Illes, Neuroscience-Based Lie Detection: The Urgent Need for Regulation, 33 Am. J.L. & Med. 377, 416 (2007). 25  Joseph A. DiMasi et al., Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs, 47 J. Health Econ. 20 (2016). 26  Janet Woodcock, Expediting Drug Development for Serious Illness: Trade-Offs Between Patient Access and Certainty, 15 Clinical Trials 230 (2018). 27  21 USC. § 356; 21 C.F.R. part 314 subpart H; 21 C.F.R. part 601 subpart E.

Regulating Medicines in the United States   697 drugs for solid tumor cancers to demonstrate either a durable objective overall response rate or progression-free survival rather than demonstrating an increase in overall survival. Using a surrogate or interim endpoint can shorten the duration of the pivotal trials for approval, and medicines approved through the accelerated process generally reach the market more quickly than those approved through the standard process.28 The manufacturer, however, must conduct research after approval to show that the medicine does provide a clinical benefit as predicted. For example, the FDA recently approved nivolumab (Opdivo) for the treatment of unresectable or metastatic melanoma based on tumor response rate and durability of response. The manufacturer must still submit the results of a multicenter randomized trial or trials establishing the superiority of Opdivo over standard therapy, and it will use an overall survival endpoint. Most of the medicines that reach the market through the accelerated approval program treat cancer. One of the most significant, for instance, was the 2001 approval of imatinib mesylate (Gleevec) for the treatment of patients with chronic myeloid leukemia based on overall hematologic and cytogenetic response rates; it ultimately was shown to dramatically improve the odds for patients. This medicine also benefitted from another program at the agency to speed access to medicines: “priority review” of the submitted application. Priority review is available for a medicine that treats a serious condition and offers a significant improvement over existing therapies.29 In addition to accelerated approval and priority review, various other FDA programs are designed to expedite research and approval. These include programs that encourage flexibility in research design and that commit the FDA to providing additional guidance to a manufacturer during research and development.30 Moreover, recent changes to federal law have encouraged the FDA to consider “real-world evidence”—generated from sources other than clinical trials, including electronic health records, claims and billing data, and product and disease registries—when evaluating new uses of already approved medicines.31 The hope is that real-world evidence can both save time and resources and be more generalizable than the evidence developed from tightly controlled clinical trials. But these recent changes did not change the standard for approval: federal law ­continues to require adequate and well-controlled investigations demonstrating effectiveness for new uses. And senior scientists at the agency remain cautious about the reliability and accuracy of real-world data.32 For now, real-world data are most likely to play an indirect role in supporting regulatory decisions, for instance, by informing patient 28  Erika Lietzan, The Drug Innovation Paradox, 83 Mo. L. Rev. 39, 66, 98 (2018). 29 FDA, Guidance for Industry: Expedited Programs for Serious Conditions—Drugs and Biologics (May 2014). 30 Id. 31  21 USC. § 355g; Jonathan P. Jarow et al., Multidimensional Evidence Generation and FDA Regulatory Decision Making, 318 JAMA 703 (2017). 32  E.g., Rachel E. Sherman et al., Real-World Evidence: What Is It and What Can It Tell Us?, 375 nejm 2293, 2295 (2016) (“Although access to real-world data adds important dimensions to the assessment of therapies and important progress is being made in the methodologic arena, these factors do not yet suffice to fully overcome the fundamental issues of confounding, data quality, and bias, unless other, specific countervailing features of the evaluation are relevant.”)

698   erika lietzan and patricia j. zettler selection for clinical trials and perhaps by supporting approval of new uses that have also been studied in clinical trials. They might also be used to support the expansion of existing indications to new patient groups, particularly for targeted therapies.33 Although most new medicines now benefit from at least one expedited development or review program, and although the FDA appears to be reviewing marketing applications somewhat faster than the European Medicines Agency (EMA) on average,34 some patients need or want access to medicines more quickly than even these programs can enable. In recognition of this reality, the FDA has long allowed patients with serious or terminal illnesses and without adequate treatment options to receive investigational medicines through its expanded access program.35 Expanded access, also known as compassionate use and preapproval access, refers to the use of an investigational medicine outside of clinical trials for treatment rather than research purposes.36 There are now two avenues for expanded access in the United States. First, access may be available under the FDA’s expanded access regulations.37 Generally, the patient, healthcare professional, and manufacturer work together to create an investigational protocol that covers the use in question. The FDA asks for information about the indication (proposed treatment use), for instance, as well as a brief clinical history of the patient and description of the treatment plan (dose, route and schedule of administration, duration of treatment, monitoring procedures, and planned modifications in the event of toxicity). For the overwhelming majority of expanded access programs, the manufacturer may charge the patient only the direct costs of manufacturing, shipping, and storing the drug, and this calculation must be audited by an independent certified public accountant and submitted to the FDA.38 And an IRB generally must review the informed consent documents and ensure that the rights and welfare of the patient are adequately protected. Although the manufacturer cannot provide access without FDA permission, the agency currently authorizes about 99% of requests, typically within hours or days and often with meaningful suggestions for improving the access pro33  For example, in 2017, the FDA extended the indication of ivacaftor (Kalydeco) from 10 cystic fibrosis mutations to 33 based on registry data in combination with laboratory studies and evidence from earlier clinical trials. 34  Advancing Health 6/16/2020Innovation: 2017 New Drug Therapy Approvals, available at https:// www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDER/ ReportsBudgets/UCM591976.pdf; Nicholas S. Downing et al, Regulatory Review of New Therapeutic Agents—FDA versus EMA, 2011–2015, 376 NEJM 1386 (2017). 35  The FDA has other policies that allow certain patients to access unapproved medicines, including its policies permitting pharmacists to compound medicines in certain circumstances (discussed in the next subsection) and permitting patients to import medicines approved in other countries for their personal use. 36  Laura L. Kimberly, et al., Pre-Approval Access Terminology: A Cause for Confusion and a Danger to Patients, 51 Therapeutic Innovation & Regulatory Science (2017). 37  21 C.F.R. §§ 312.300–312.30; FDA, Expanded Access to Investigational Drugs for Treatment Use—Questions and Answers (Oct. 2017). 38  21 C.F.R. § 312.8. For the small number of expanded access programs that involve large groups of patients, the manufacturer also may recover certain administrative costs, such as costs associated with monitoring the expanded access program. Id.

Regulating Medicines in the United States   699 gram.39 Generally, the agency prefers the drug to have completed phase 2 trials, but it can—and does—permit access earlier in some cases. Patients, their caregivers, and advocacy groups have been frustrated by their lack of earlier access to investigational medicines since the 1970s, however, leading to reform efforts and (largely unsuccessful) lawsuits against manufacturers and the agency.40 The second access pathway is a result of the recent “right to try” movement, which culminated in enactment of a new federal law in 2018 permitting access to investigational medicines that have completed phase 1, without FDA authorization in certain circumstances.41 An eligible patient must have a life-threatening disease or condition, must have exhausted approved treatment options, and must be unable to participate in a clinical trial involving the drug. In truth, this legislation may do more harm than good. To begin with, it may not actually provide patients with access to more drugs than FDA’s existing practices already provided; although it explicitly applies to drugs after completion of phase 1 testing, in practice, the agency already permits access to investigational drugs at any stage of development. In addition, the FDA’s requirements are usually not the impediment to patient access. Instead, manufacturers choose not to provide access. And although the new law attempts to address some manufacturer concerns, it does not address all of the concerns. For example, the 2018 law provides manufacturers and prescribers with protection from liability arising out of use of medicine administered under a right-to-try program.42 But manufacturers have also cited concerns that adverse events in expanded access programs will delay approval. The 2018 law describes the circumstances in which the FDA may use clinical outcomes in its review of investigational drugs, but it states—as it should—that the agency may use these data if they are critical to determining the safety of the drug. In other cases, a manufacturer might lack an adequate supply of a medicine to respond to all requests and complete its clinical trials on schedule, or it may be concerned that an expanded access program will make it difficult to enroll patients in its clinical trials. Thus it is not clear that the new law will actually provide patients with access to any drugs that they could not have received under the FDA’s expanded access program. At the same time, it deprives patients of the benefits of agency oversight.43 In the classic expanded access scenario, a treating physician works with the manufacturer, a local IRB, and the agency to ensure that the patient has no other treatment options, that the patient has provided informed consent, that the patient is charged only permissible costs, and that the proposed treatment plan has a positive benefit-risk profile for the patient. Under 39  Jonathan P. Jarow, et al., Overview of FDA’s Expanded Access Program for Investigational Drugs, 51 Therapeutic Innovation & Regulatory Science 177 (2017). 40  See, e.g., Patricia J. Zettler & Henry T. Greely, The Strange Allure of State Right-to-Try Laws, 174 JAMA Internal Med. 1885 (2014). 41  21 USC. § 360bbb-0a. 42  Pub. L. No. 115–176 § 2(b), 132 Stat. 1374 (protection from liability unless the relevant conduct constitutes reckless or willful misconduct, gross negligence, or an intentional tort). 43  E.g., Alison Bateman-House, “Right To Try” Is Law, Now What?: Part 1, Health Affairs Blog (Oct. 25, 2018).

700   erika lietzan and patricia j. zettler the 2018 law, however, only a manufacturer and physician are involved, which raises the risk that desperate patients will fall prey to the unscrupulous. It is not yet clear whether or how this law will change the landscape of pre-approval access, but anecdotal evidence suggests that many manufacturers prefer the formal FDA expanded access program and handle incoming requests through that program, regardless of how the requests are formulated.

2.3  Other Routes to Market There also are ways to market medicines without FDA approval of full applications containing the results of premarket safety and effectiveness testing. One such way is an abbreviated application, which is also an option in the European Union. Instead of containing data directly demonstrating safety and effectiveness, abbreviated applications make a comparative showing sufficient to establish a bridge to research submitted by another company (sometimes called the “innovator”) that submitted a full application.44 To approve an abbreviated application, the FDA relies on the comparative showing, the fact that the agency approved the reference product, and (indirectly) the research performed by the innovator. Abbreviated applications are less expensive to prepare than full applications, which generally means the resulting medicines are less expensive for payers and patients. The abbreviated pathways thus accomplish a key goal for US health policy: cost-savings for the healthcare finance system. There are three types of abbreviated application for medicine in the United States. First, the FDA can approve an abbreviated new drug application (ANDA) for a generic version of a “drug” that is not a biological product.45 The generic drug approval requirements in the United States are generally the same as those in Europe. A generic drug generally has the same active ingredient, route of administration, dosage form, and strength as the innovator drug it relies on, and it must be bioequivalent to that drug.46 With a few narrow exceptions, a generic drug bears the same labeling for prescribers as its reference listed drug.47 Second, a manufacturer that wants to market a drug that differs from the previously approved innovator drug may submit a 505(b)(2) application, named after the provision of the FDCA in which it is described.48 For instance, the manufacturer might propose a new use, a new dosage form or formulation, or even a change to the active ingredient.49 FDA approved levocetirizine dihydrochloride (Xyzal) for seasonal allergic rhinitis, to give an example, based on a 505(b)(2) application that cited cetirizine hydrochloride (Zyrtec). Paclitaxel protein-bound particles for injectable suspension (Abraxane) was approved for the treatment of breast cancer on the basis of a 505(b)(2) application that cited paclitaxel (Taxol). The manufacturer’s goal had been to 44  Erika Lietzan, The Uncharted Waters of Competition and Innovation in Biological Medicines, 44 Fla. St. L. Rev. 883, 888 (2017). 45  21 USC. § 355(j). 46  21 USC. § 355(j)(2)(A). 47  21 USC. § 355(j)(2)(A)(v). 48  21 USC. § 355(b)(2). 49 FDA, Guidance for Industry, Applications Covered by Section 505(b)(2) (draft) (Oct. 1999).

Regulating Medicines in the United States   701 develop a formulation that omitted cremophore, an inactive ingredient associated with some toxicity. Like a hybrid application in the European Union, the 505(b)(2) application relies partly on the results from testing the referenced drug and partly on new clinical data. Third, “biosimilar” biological products are approved on the basis of abbreviated applications filed under the Public Health Service Act.50 Like a generic drug, a biosimilar biologic is meant to be a replica of an innovator product. For example, adalimumab-atto (Amjevita) is a biosimilar version of adalimumab (Humira). The biosimilar application differs from an ANDA, however, in part because biologics differ meaningfully from most chemically synthesized drugs. They are larger, more complex, and often incompletely characterized or not well understood.51 They also have the potential to trigger undesirable (and potentially life-threatening) immunogenic reactions, which can be impossible to predict without human trials.52 The biosimilar licensure requirements in the United States are similar to those in Europe, and, in fact, the United States developed its pathway after Europe did.53 US law presumes that a biosimilar application will contain comparative analytical studies, animal studies, and clinical studies, and that the clinical studies generally will include clinical immunogenicity studies, pharmacokinetic or pharmacodynamic studies, and comparative clinical outcomes data relevant to at least one indication. Biosimilar biologics can be expensive and risky to develop, but—like generic drugs— they have the potential to meaningfully lower healthcare costs. Very few biosimilar biologics have been approved in the United States to date, leading some to criticize the FDA for imposing unnecessarily cautious testing requirements.54 Others, however, believe that even more caution may be warranted.55 Approving biosimilars has required the FDA to walk a difficult line in its mission between ensuring safety and effectiveness, on the one hand, and ensuring prompt access to medicines, on the other hand. The agency tries to reduce the burden on biosimilar applicants as much as possible, consistent with protecting the public health. To date, it has mostly focused on fostering a collaborative process with companies that tailors data requirements as a company proceeds.56 50  42 USC. 262(k). 51  Janet Woodcock et al., The FDA’s Assessment of Follow-On Protein Products: A Historical Perspective, 6 Nat. Revs. Drug Discovery 437, 438 (June 2007). 52  See generally FDA, Guidance for Industry: Immunogenicity Assessment for Therapeutic Protein Products (Aug. 2014). 53  Krista Carver, Jeffrey Elikan, & Erika Lietzan, An Unofficial Legislative History of the Biologics Price Competition and Innovation Act of 2009, 65 Food & Drug L. J. 671, 691–95 (2010). 54  See, e.g., Yaniv Heled, Follow-On Biologics Are Set Up to Fail, 2018 U. Ill. L. Rev. Online 113 (2018). 55  Other regulators have sometimes been more conservative, as when Health Canada approved biosimilar infliximab for rheumatic and related uses but declined to approve four gastrointestinal indications without additional data. See Lietzan, Uncharted Waters, supra note 44, at 900 n.71. 56 FDA, Guidance for Industry: Scientific Considerations in Demonstrating Biosimilarity to a Reference Product (Apr. 2015); FDA, Guidance for Industry: Formal Meetings Between the FDA and Biosimilar Biological Product Sponsors or Applicants (Nov. 2015).

702   erika lietzan and patricia j. zettler Innovative medicines, generic drugs, and biosimilars are all considered “new” medicines in the United States, and they require approved marketing applications. Other medicines, however, may reach the market without FDA approval if they are “generally recognized” by qualified experts as “safe and effective” for their labeled uses and have been used for a material time and to a material extent for those uses. This standard is met only when the data supporting safety and effectiveness are at least as scientifically rigorous as the data that would support premarket approval.57 That said, the FDA has concluded that several hundred active ingredients are “generally recognized as safe and effective” for over-the-counter (OTC) use when they are marketed as described in detailed agency regulations known as “OTC monographs.”58 As a result, most nonprescription medicines on the market in the United States are not the subject of premarket approval. Within just the category comprising cough and cold medicines, for instance, this includes common antihistamines (such as brompheniramine maleate and diphenhydramine citrate), expectorants (such as guaifenesin), oral decongestants (such as pseudoephedrine), topical decongestants (such as oxymetazoline hydrochloride), and antitussives (such as codeine and dextromethorphan). For many of these drugs, the manufacturer simply complied with an OTC monograph. Pharmacies also sometimes prepare customized medicines for patients without obtaining FDA approval. Typically this involves combining, mixing, or changing the ingredients of a medicine to meet the specific needs of a patient, for instance because the patient is allergic to an inactive ingredient used by the manufacturer or because the patient cannot swallow the capsules marketed by the manufacturer. Compounding practices require policymakers to balance competing policy goals—ensuring patients have access to medicines they can use, on the one hand, while ensuring they receive safe and effective treatments, on the other hand. FDA regulation of compounding, in turn, raises federalism concerns because compounding is viewed as a traditional pharmacy practice, and individual states, rather than the federal government, generally regulate pharmacies. If a state-licensed pharmacy compounds a medicine in response to a valid patient-specific prescription, the FDA does not approve the medicine (and thus does not verify its safety and effectiveness), nor does it require the pharmacy to use current good manufacturing practices.59 Some companies also compound medicines in bulk, for shipment to individual pharmacies and for healthcare providers (such as hospitals) to keep on hand, but historically these companies operated without federal oversight. After contaminated steroid injections compounded and shipped by the New England Compounding Center in 2012 led to more than 800 cases of fungal meningitis, including more than 70 deaths, Congress amended the law. The FDCA now permits companies to register as “outsourcing facilities” to compound medicines in larger supply for healthcare providers (such as hospitals) to keep on hand.60 These facilities are subject to current good manufacturing practices and regular FDA inspection, but the medicines 57  Weinberger v. Hynson, Wescott & Dunning, Inc., 412 US 609 (1973). 58  See generally 21 C.F.R. part 330. 59  21 USC. § 353a. 60  21 USC. § 353b.

Regulating Medicines in the United States   703 are not subject to FDA approval. More significantly, registering is voluntary, which means that Congress has not yet fully addressed the risk presented by companies that compound medicines in bulk without any federal regulatory oversight. Patients in the United States also purchase unapproved homeopathic medicines, which tend to be available without a prescription. Homeopathic medicine is based on two principles: that a substance that causes symptoms in a healthy person can be diluted to treat the person, and that a substance is more potent if it is more diluted.61 For instance, caffeine might be diluted to treat sleeplessness, or cantharsis (a caustic chemical extracted from a poisonous beetle) might be diluted to treat blisters. Many homeopathic medicines are, when properly prepared, so diluted that no detectable molecules of the original substance remain. Because they are labeled for treatment of medical conditions, however, homeopathic medicines are considered “drugs” under US law and technically require FDA approval. Yet none is the subject of an approved application, and it is unlikely that any could satisfy the standard for approval. The FDA historically exercised its discretion not to enforce approval requirements, perhaps out of deference to a strong culture of autonomy in communities that rely heavily on homeopathic medicine. This is concerning from a health policy perspective because it means consumers continue to spend money on treatments that have no pharmacological action and cannot provide safe and effective relief for their conditions. The FDA has historically taken action, however, when safety issues emerged—as might be the case if a homeopathic product actually contained detectable amounts of the symptom-causing molecule or if it contained other harmful ingredients or was contaminated during manufacturing. But, even still, the agency sometimes only issued warnings. For example, in late 2016, it simply warned against the use of homeopathic teething tablets and gels containing belladonna, which can trigger unpredictable reactions in children under 2 years of age. These products were associated with serious adverse events, including seizures and death. In late 2017, concerned about the increasing use of homeopathic drugs, the FDA finally revised its policy for homeopathic medicine, making it clear that it will not only focus oversight on products with reported safety concerns but also on products containing high-risk ingredients, products for vulnerable populations, and products intended for serious or life-threatening conditions. It lacks the resources and may lack the political cover to forcefully remove homeopathic products from the market altogether, however. Finally, consumers may purchase “dietary supplements,” which are not subject to ­premarket approval by FDA. Dietary supplements generally contain vitamins, minerals, or herbs, and they are often presented in traditional medicine forms, such as capsules or tablets.62 A manufacturer marketing a supplement with a well-established ingredient may have no premarket interaction with the agency. A manufacturer that wishes to market a new ingredient must notify the agency before launch, providing evidence that the

61 FDA, Guidance: Drug Products Labeled as Homeopathic (draft) (Dec. 2017). 62  21 USC. § 321(ff).

704   erika lietzan and patricia j. zettler ingredient is reasonably expected to be safe.63 A manufacturer selling a dietary supplement may claim that the product affects the structure or function of the human body—for instance, that it helps to maintain healthy heart function (rather than that it treats coronary heart disease).64 US law also permits a dietary supplement manufacturer to make a “health claim,” which links the supplement to a specific diseases—such as a claim linking calcium and osteoporosis.65 This claim would turn most other products into a medicine requiring premarket approval, but US law carves out dietary supplements, permitting such claims based on lower levels of evidence. There is a question whether US dietary supplement regulation adequately serves the public health, in part because it is unclear whether US consumers understand the different levels of scientific support for the claims made about medicines and about dietary supplements. At the same time, the US culture of personal autonomy with respect to preventive care and self-treatment for minor conditions (and resistance to regulating products perceived to be “natural”) makes more aggressive regulation of dietary supplements a difficult choice for policymakers.

3  Risk Management As noted, no medicine is effective (or effective the same way) in every patient, and no medicine is perfectly safe. Premarket safety and effectiveness testing generates a substantial body of data for the FDA’s first, and most important, risk management decision: whether to approve the new medicine in the first instance. But it is impossible to know everything about a medicine’s clinical profile based on a reasonable premarket program. Premarket clinical trials may not have uncovered all risks, particularly those that are rare or have long latency periods. In addition, they are highly controlled and tend to involve homogenous populations. Clinical use after approval—in a larger and heterogeneous population of real-world patients—may uncover new risks (and new benefits). Tools for risk management, which includes monitoring and assessing risk, are therefore implemented at the time of approval and are the dominant feature of postmarket ­regulation of new medicines.

3.1  Managing Risk at the Time of Approval The FDA makes several risk management decisions when it approves a medicine. First, it decides whether to limit the medicine to prescription status.66 FDA does so if the 63  21 USC. § 350b(a); 21 C.F.R. § 190.6; FDA, Guidance: Dietary Supplements: New Dietary Ingredient Notifications and Related Issues (draft) (2016). 64  21 C.F.R. 101.93. 65  21 USC. § 343(r)(3); 21 C.F.R. § 101.14(a)(1). 66  Separate from the FDA’s authority, the federal Controlled Substances Act, administered by the Drug Enforcement Administration, limits the refilling of prescriptions for certain controlled substances based on their potential for abuse. 21 USC. § 829.

Regulating Medicines in the United States   705 medicine is not safe for use without the supervision of a healthcare professional licensed (under state law) to administer the medicine—for instance, because of its toxicity or potential for other harmful effects.67 As a practical matter, the FDA limits almost every new medicine in the United States to prescription status. Unlike Europe, the United States currently has no official “behind the counter” or “pharmacy only” class for medicines, meaning no class of medicines that can be purchased without a prescription but only after consultation with a pharmacist. A medicine generally is either nonprescription (in which case it may be sold in any store, including a supermarket or gas station) or prescription (in which case it must be dispensed—and thus sold—by a pharmacist licensed under state law). Second, the FDA manages risk when it approves labeling for prescribers—also known as the package insert or professional labeling. The labeling distills the safety and effectiveness information in the marketing application into detailed, but accessible instructions and guidance, sufficient to allow the prescriber to make an informed decision about whether, and how, to use the medicine for an individual patient.68 FDA review of a marketing application includes close scrutiny and often heavy revision of the applicant’s proposed professional labeling, and, when the agency approves a medicine, it approves every word of that labeling. As a result, the professional labeling is the FDA’s primary risk management tool for a new medicine. Third, in some cases the FDA will require a formal risk management program— known as a risk evaluation and mitigation strategy (REMS)—when it approves a ­medicine. By law the agency must require a REMS if one is necessary to ensure that a medicine’s benefits outweigh its risks.69 European law, by way of contrast, requires a Risk Management Plan for every approved medicine. US law describes two categories of the risk management measures that the FDA can impose. One category is enhanced communications with patients and healthcare professionals. For instance, the agency might require special labeling for patients that describes a particular risk, or it may require that information be sent to healthcare providers. The REMS for 1% testosterone (Androgel), for example, requires that a Medication Guide be dispensed with the drug, warning patients about the risk to children and women who are accidentally exposed to the gel through skin-to-skin contact. The REMS for naltrexone (Vivitrol) informs patients and healthcare providers about severe injection site reactions associated with use of the drug. The second category goes further. If a medicine is associated with a serious adverse experience, the FDA can limit access to the medicine. For instance, the agency can require that healthcare professionals have particular training before prescribing the medicine, it can require that dispensing pharmacies have special certifications, and it can require that the medicine be dispensed only in some settings or only to patients who satisfy specific conditions (such as laboratory test results).70 The agency can also require that patients be monitored or entered into a registry.71 For example, FDA requires a REMS for natalizumab (Tysabri), which is approved for treatment of multiple sclerosis 67  21 USC. § 353(b)(1). 68  71 Fed. Reg. 3922, 3922 (Jan. 24, 2006). 70  21 USC. § 355-1(f)(3)(A)-(C). 71  21 USC. § 355-1(f)(3)(F).

69  21 USC. § 355–1.

706   erika lietzan and patricia j. zettler and Crohn’s disease due to the risk of progressive multifocal leukoencephalopathy (PML), an opportunistic viral infection of the brain that usually leads to death or severe disability. Healthcare professionals who prescribe Tysabri must become certified to prescribe the drug, counsel each patient about the risk before initiating treatment, enroll that patient in the REMS Program, and assess the patient regularly for signs and symptoms of PML. Pharmacies that dispense Tysabri must also receive training from the manufacturer and enroll in the REMS Program. The FDA has imposed restrictions on the use and distribution of roughly four dozen medicines. These restrictions can be controversial because they may reduce access to the medicine and can limit a healthcare professional’s ability to prescribe an approved medicine according to his best judgment about the needs of his patient. Regulation of the practice of medicine usually falls to state law.72 Finally, when there are known risks or signals of risks, the FDA may require a company to conduct additional safety-related research—“postmarketing” studies and clinical trials—after approval.73 Postmarketing studies and trials include prospective clinical trials in humans as well as the full range of other studies a company might perform, including observational epidemiological studies in humans, animal studies, and laboratory experiments.74 But the statute effectively requires the agency to first use the least burdensome approach to studying the risk in question—adverse event reporting in the first instance, and studies other than clinical trials next.75 After reviewing the results of this research, the agency might require additional risk information in the labeling of the medicine or a change in the medicine’s approved conditions of use. Risk management decisions at the time of a medicine’s approval—beginning with the approval decision itself—are meant to ensure that a medicine’s benefits outweigh its risks. Making these decisions entails difficult judgment calls that are complicated by at least three factors. First, the decisions generally must be made without complete information about the medicine’s benefits and risks. The agency must also decide what level of uncertainty is acceptable. Second, views about the acceptable level of risk—within the agency but also among patients, the broader public health community, and policymakers—vary considerably. They may vary over time and by stakeholder, by product type, and in accordance with factors that the FDA cannot predict. Third, each risk management measure itself imposes costs on the healthcare system. This can occur in obvious ways—such as when use and distribution restrictions in REMS limit access—and in less obvious ways, such as if numerous warnings confuse healthcare providers and obscure the primary risks. Reaching an acceptable risk management decision about a new medi-

72  Patricia J. Zettler, Toward Coherent Federal Oversight of Medicine, 52 San Diego L. Rev. 427, 430–31 (2015). 73  21 USC. § 355(o)(3). 74 FDA, Guidance for Industry: Postmarketing Studies and Clinical Trials— Implementation of Section 505(o)(3) of the Federal Food, Drug, and Cosmetic Act (Apr. 2011), at 4. 75  21 USC. § 355(o)(3)(D).

Regulating Medicines in the United States   707 cine, given these considerations, is a difficult task for the agency that can lead to criticism from the public or other government entities no matter how the FDA rules.

3.2  Monitoring and Managing Risk After Approval Regardless of the other risk management measures the FDA might have imposed, any manufacturer that markets a medicine in the United States has an ongoing responsibility to report and respond to safety information that it learns about while the product is on  the market. The FDA expects each company to design and maintain an effective “pharmacovigilance” system, which will allow investigation and reporting of any adverse event associated with use of the medicine, no matter how the company learned the information and no matter how the adverse event occurred.76 Serious unexpected “adverse drug experiences”—adverse events not included in the approved labeling that lead to death or hospitalization, for instance—must be reported to the FDA within 15 days.77 All other adverse events must also be reported, albeit less frequently, through quarterly or annual reports analyzing risk information. This passive approach to monitoring for new risk information has inherent limitations. Many adverse events are not reported to healthcare professionals or manufacturers, and even when they are reported there may be inadequate information to evaluate the risk. The FDA has therefore also launched the “Sentinel Initiative,” which allows it to more proactively monitor the safety of medicines after they have reached the market.78 This system gives the agency access to electronic healthcare data from a variety of data partners (including medical schools and insurers). The goal of pharmacovigilance is to ensure that, as more is known about a medicine’s clinical profile, the benefits of the medicine continue to outweigh its risks. Ultimately, therefore, additional safety information must be analyzed and placed within the broader context of what is known already about the medicine. If a safety signal is detected, therefore, it may be appropriate for the company to conduct follow-up research, and in some cases the FDA can require additional research.79 Additional safety information also may warrant a change to the professional labeling. When the FDA becomes of aware of new safety information, it can require changes to the labeling.80 For example, in August 2018, the agency notified the manufacturer of ertugliflozin and metformin hydrochloride (Segluromet), a recently approved diabetes treatment, that it was required to add information about the risk of necrotizing fasciitis of the perineum, also known as Fournier’s gangrene.81 A manufacturer independently has an obligation to revise its labeling to include a warning about a clinically significant hazard as soon as there is “reasonable evidence” of a 76  21 C.F.R. § 314.80. 77  21 C.F.R. § 314.80(c)(1). 78  21 USC. § 355(k)(B)(i)–(ii). 79  21 USC. § 355(o)(3). 80  21 USC. § 355(o)(4). 81  See Letter from William Chong, CDER, to Merck Sharp & Dohme Corp., NDA 209806/S-001 (Oct. 26, 2018) (approving the supplemental NDA submitted by Merck to include the requested labeling change).

708   erika lietzan and patricia j. zettler causal association between the medicine and the hazard, and it can do so without seeking prior approval from the FDA.82 The company takes some risk in this case that  the agency will disapprove the change after the fact, forcing the company to stop distribution with the new labeling.83 But, as a practical matter, companies must make labeling changes immediately to reduce their exposure under state products liability law.84 Currently, however, this dilemma only applies to innovators. Generic drugs must have the same labeling as the corresponding brand products, and the FDA takes the position that generic companies may not make unilateral safety-related labeling changes.85 As a result, generic companies do not face exposure under products liability law for risks known to them but not yet included in the labeling. The FDA has considered making changes to these rules that presumably would result in generic manufacturers being exposed to product liability, but stakeholders have strongly differing views about the right policy outcome. In some situations, new safety information can lead to more significant changes— a new or modified risk management program, for instance, or even withdrawal from the market. Typically product withdrawal is voluntary; although the FDA has the power to withdraw approval of marketing applications,86 companies generally agree with the agency (or even reach the decision on their own initiative) that a product’s benefits no longer outweigh its risks. Market withdrawals, whether voluntary or effectively required by the FDA, can be controversial because they end patient access to a previously available medicine. Particularly if a medicine satisfies an otherwise unmet medical need relating to a serious condition, patients may still be willing to accept the risk and may organize to persuade the FDA and other policymakers that  the medicine should remain available. According to senior leadership at the FDA, for example, patient advocacy was key to the relaunch of alosetron (Lotronex) in 2002, after this medicine for irritable bowel syndrome was withdrawn from the market.87 The outcomes from ischemic colitis (a known side effect) had been more serious than expected, and several patients had experienced serious complications from constipation. Deaths had been reported. The  manufacturer did not push for reintroduction but eventually agreed to work with the FDA to design a risk management program that could permit relaunch. Biogen withdrew natalizumab (Tysabri) from the market when patients in a clinical trial developed PML, a fatal brain ­infection, but patients pressed the company for access. After the company worked with the FDA to design a risk management program that would detect infections

82  21 C.F.R. §§ 201.57(c)(6), 314.70(c)(6)(iii). 83  21 C.F.R. § 314.70(c)(7). 84  Wyeth v. Levine, 555 US 555 (2009) (holding that state law failure-to-warn product liability claim was not preempted by federal law because NDA holder could have unilaterally amended its labeling to include the warning in question). 85  21 USC. § 355(j)(2)(A)(v); Pliva v. Mensing, 564 US 604, 614 (2011). 86  21 USC. § 355(e). 87  Janet Woodcock, Commentary: The Reintroduction of Lotronex for Diarrhea-Predominant Irritable Bowel Syndrome, 2 bmj usa 637 (2002).

Regulating Medicines in the United States   709 early, patients spoke forcefully in support of the product, and the agency approved reintroduction.88

4  Innovation and Competition Premarket testing of new medicines imposes costs on manufacturers that must be recouped in the marketplace. Various aspects of US law make this possible by providing the manufacturer with a period for exclusive sales in the market. Such exclusivity, however, can lead to higher prices. Higher pricing for medical products raises public health and social policy concerns: FDA authorization is not sufficient for patients to have access to the benefits that a medical product offers; the product must also be affordable for patients and payers. At least some exclusivity is needed, however, so that manufacturers can justify investing in some types of innovative products, given the length of time required for premarket testing.89 The US regulatory scheme for medicines, like the European regulatory scheme, must balance providing and protecting incentives for manufacturers to spend their resources developing important new medical products, on the one hand, with facilitating rapid market entry of lower cost versions of medical products, on the other hand.

4.1  Incentives for Innovation The primary incentive to discover and develop a new medicine in the United States is the prospect of a period of time to market the medicine without generic or biosimilar competition. During the exclusivity period, the medicine may face modest price competition from other treatments for the same condition, but it does not face competition from duplicates approved on the basis of abbreviated applications. The manufacturer can use this time to recover its investment in developing the medicine as well as any failed prospects and perhaps medicines that have not been commercially successful.90 Policymakers face a difficult tradeoff when using an exclusive marketing period as the incentive for development of new medicines. The price of a new medicine during the exclusivity period can be prohibitively expensive for some payers or some patients, even those with private health insurance provided by their employers. Earlier approval of copies will hasten access to more affordable versions of new medicines, but it means a shorter exclusivity period—and thus either even higher prices or a reduced incentive for medical innovation. Policymakers in the United States and Europe have struggled 88  Brady Huggett, How Tysabri Survived, 27 Nature Biotech. 986 (2009). 89  See, e.g., Erika Lietzan, The Drug Innovation Paradox, 83 Mo. L. Rev. 39 (2018). 90  Henry Grabowski et al., The Roles of Patents and Research and Development Incentives in Biopharmaceutical Innovation, 34 Health Affairs 302 (2015); Mark A. Lemley, The Regulatory Turn in IP, 36 Harv. J. L. & Pub. Pol’y 109, 114 (2013).

710   erika lietzan and patricia j. zettler for decades with striking the right balance between access and affordability, on the one hand, and adequate incentive for innovation, on the other hand. Like the European Union, the United States ensures a period of exclusive marketing for new medicines through both data exclusivity (also known as regulatory data protection) and protection of patents. The US approach differs fundamentally, however, by having separate schemes with differing rules for drugs and biologics. Like European law, US law permits the patenting of any new and useful process or composition of matter provided various other requirements of the Patent Act are satisfied.91 A medicine manufacturer typically holds or exclusively licenses a patent claiming the medicine’s active ingredient. It may also hold patents claiming the formulation or composition of the particular finished product or an approved method of using or administering the product, the manufacturing process, or even an intermediate chemical entity used during the manufacturing process or a metabolite of the active ingredient.92 Generally speaking, US courts will prohibit a company from marketing an infringing generic drug or biosimilar medicine during the patent term, which today lasts for 20 years from the date that the patent application was filed.93 Ordinarily, making and using a patented product during the patent term would constitute patent infringement.94 Like European law, however, US patent law exempts activities that are designed to generate information supporting approval of a new medicine.95 As a result, generic and biosimilar companies may manufacture and test their infringing products during the patent term. The FDA will also review the application during the patent term. Unlike European law, US law links regulatory approval of generic and biosimilar applications with the patents held by the corresponding innovator, in two ways. First, the drug and biologics statutes require a generic applicant and biosimilar applicant, respectively, to take a formal position on the relevant patents owned by the innovator— as to each patent, either (1) that the patent is no impediment because its product does not infringe the patent or the patent is invalid, or (2) that the patent is valid and that it therefore intends to launch after patent expiry.96 If the generic or biosimilar applicant asserts non-infringement or patent invalidity, US patent law gives federal courts jurisdiction to hear the case even though the product has not launched.97 Second, the drug statute prohibits the FDA from approving a generic application if the applicant concedes it must wait for patent expiry, and it usually delays FDA approval for 30 months if there is patent infringement litigation.98 The biologics statute is different: there is no link

91  35 USC. § 101. 92  21 C.F.R. § 314.53(b); see generally John R. Thomas, Pharmaceutical Patent Law 46–64 (3d ed. 2015). 93  35 USC. § 154. If the application contained a specific reference to an earlier filed application, the patent lasts for 20 years from the date on which the earlier application was filed. 94  35 USC. § 271(a). 95  35 USC. § 271(e)(1). 96  21 USC. §355(j)(2)(A)(vii); 42 USC. § 262(l)(3)(B)(ii). A generic drug applicant is not required to address patents that claim a method of manufacturing the innovative medicine. 97  35 USC. § 271(e)(2). 98  21 USC. § 355(j)(5)(F)(ii).

Regulating Medicines in the United States   711 between the timing of biosimilar licensure and the existence of or disputes over patents claiming the innovator’s medicine. The period for marketing exclusivity also derives from data exclusivity. Data exclusivity refers to the period after an innovative medicine’s approval before any abbreviated application, seeking to rely on the innovator’s research, may be submitted or, in some cases, approved. Data exclusivity is a regulatory concept. It is separate from patent protection (and, indeed, applies regardless of whether the innovator holds patents), and it runs in parallel with any patents that the innovator might hold. The term varies. When the FDA approves a new drug containing a new chemical entity, generic applications cannot be submitted for 5 years.99 This period drops to 4 years if the generic applicant takes the position that its product does not infringe the innovator’s patent or that the patent is invalid.100 In contrast, a biosimilar application may not be submitted until 4  years after first license of a biological product, and the FDA may not approve the biosimilar application until 12 years after first licensure of the biological product.101 US law also provides incentives for companies to continue studying their approved medicines. New conditions of use—such new indications, formulations, or dosage forms developed after approval—may benefit from patent protection. In the case of a new drug, these may also enjoy 3 years of regulatory exclusivity.102 This exclusivity protects only the new condition of use, however, which means that the FDA may approve a generic drug for the remaining unprotected conditions of use. For example, a manufacturer might develop a drug for breast cancer and later for colon cancer. If the manufacturer holds 3-year exclusivity for the colon cancer indication, then the FDA cannot approve generic copies for colon cancer—but it can approve them for breast cancer (when the patents and exclusivity associated with the active ingredient and breast cancer indication expire). This approach differs from the European approach of extending the basic (core) data exclusivity term by 1 year for a new indication that brings a significant clinical benefit.103 And it is available only for new drugs—not for biological medicines. In recent years policymakers have also experimented with more targeted exclusivities.104 Finally, US law provides additional and more specific incentives for particular types of innovation. First, just as in Europe, a medicine to treat a rare (“orphan”) disease benefits from a special type of exclusivity: the FDA cannot approve any other application— abbreviated or full—for the same medicine for the same disease for 7 years.105 Second, if a manufacturer performs pediatric research pursuant to a written request from the FDA, all regulatory exclusivities are extended for 6 months (including, in the case of drugs, the prohibition on FDA approval until patent expiry).106 Third, approved applications 99  21 USC. § 355(j)(5)(F)(ii). 100 Id. 101  42 USC. § 262(k)(7). 102  21 USC. § 355(j)(5)(F)(iv). 103  See Mahalatchimy, “Regulating Medicines in the European Union,” in this volume. 104  Policymakers have experimented with various other ways of encouraging medical innovation, including by rewarding desired research with transferrable vouchers for priority review of marketing applications. 21 USC. §§ 360n, 360n-1. 105  21 USC. § 360cc. 106  21 USC. § 355a.

712   erika lietzan and patricia j. zettler for certain types of antibiotics enjoy a 5-year extension of all applicable regulatory exclusivities.107

4.2  Competition and Advertising 4.2.1 Brand-to-Brand The FDA uses its rules concerning the advertising and promotion of medicines as one tool to both manage the risks of medicines and encourage fair competition in the marketplace between brand drugs.108 Thus, the agency generally aims to ensure that manufacturer-provided information about prescription medicines is “truthful, balanced, and accurately communicated.”109 To that end, the FDA imposes numerous requirements on manufacturer’s advertising and promotional materials, including the requirement that any given piece provide a “brief summary” of adverse effects, contraindications, and effectiveness, as well as “fair balance” between benefit and risk information.110 The agency considers material false, lacking in fair balance, or misleading if—among other things— it claims that one drug is safer or more effective than another when this has not been shown by substantial evidence or substantial clinical experience, if it makes claims about effectiveness on the basis of a study but fails to explain the study’s limitations, or if it relies on favorable information that is now outdated and that has been superseded by contrary and more credible recent information.111 The agency also has long taken the position that, although healthcare professionals may generally prescribe medicines for uses that the FDA has not approved, manufacturers may not promote their medicines for these uses. The FDA views this prohibition on off-label promotion as necessary to a number of regulatory goals.112 For instance, the agency has argued that this prohibition promotes public health and safety by encouraging manufacturers to undertake the scientific studies and perform the other steps necessary to obtain approval for off-label uses. The agency also reasons that requiring premarket review of every promoted use will help to prevent harm to the public by protecting against fraud and misrepresentations and prevent the diversion of healthcare resources toward ineffective treatment. The FDA’s rules governing advertising, however, also account for the strong US culture of respecting autonomy, including patients’ desire for information about their med-

107  21 USC. § 355f. 108  A different federal agency, the Federal Trade Commission, regulates the advertising of OTC medicines. 109  US Food & Drug Admin., The Office of Prescription Drug Promotion, Mission, https://www.fda. gov/aboutfda/centersoffices/officeofmedicalproductsandtobacco/cder/ucm090142.htm. 110  21 C.F.R. § 202.1. 111  Id. § 202.1(e)(6). 112  E.g., FDA, Memorandum: Public Health Interests and First Amendment Considerations Related to Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products (Jan. 2017).

Regulating Medicines in the United States   713 icines and, more importantly, respect for constitutional protections for speech.113 Since the 1970s, US courts have concluded that the US Constitution’s protections for freedom of speech extend to commercial speech, including the advertising and promotion of medicines. This may be one reason that, despite concerns that direct-to-consumer (DTC) promotion of prescription medicines will lead to inappropriate use of medicines, the FDA generally permits this promotion. Indeed, the United States is one of only two countries (the other being New Zealand) that permits DTC promotion of prescription medicines. But it is the FDA’s prohibition on off-label promotion, in particular, that has been one of the most high-profile areas in which the tension between protections for speech and FDA policies have come to light. Recent court decisions have called into question the constitutionality of the agency’s policies on off-label promotion. Perhaps most notably, in 2012, a federal appeals court concluded that the prosecution of a sales representative marketing sodium oxybate (Xyrem) for off-label uses violated his First Amendment rights.114 There were some unusual circumstances in the case, such as that, for procedural reasons, the court assumed that the representative’s statements—including statements that Xyrem, a drug associated with a risk of death, was “very safe”—were truthful and not misleading.115 Nevertheless, some manufacturers have since had success challenging the agency’s off-label promotion policies, and the agency is currently reexamining its policies.116 This development has been welcomed by some who view the prohibition as restricting the flow of potentially useful information to healthcare professionals who are permitted to prescribe medicines off-label, while others share the FDA’s concerns that widespread off-label promotion will increase inappropriate prescribing, resulting in public health harms and its view that the prohibition is important to ensure companies perform the high-quality research needed to earn approval of these unapproved uses.117

4.2.2 Substitution Unlike innovator manufacturers, generic manufacturers generally do not advertise or promote their medicines except to insurers. Typically, when the FDA approves a generic drug, it issues a “therapeutic equivalence” rating, signaling that the generic can be substituted for a prescribed brand product, with the expectation that the two will have the

113  See, e.g., Lewis A. Grossman, FDA and the Rise of the Empowered Consumer, 66 Admin. L. Rev. 627, 662 (2014). 114  United States v. Caronia, 703 F.3d 149 (2d Cir. 2012). 115  See, e.g., Patricia J. Zettler, The Indirect Consequences of Expanded Off-Label Promotion, 78 Ohio State L.J. 1053 (2017). 116  Public Hearing, 81 Fed. Reg. 60,299 (Sept. 1, 2016). 117  Although regulating off-label prescribing rather than manufacturers’ promotional activities may seem like the obvious means for addressing concerns about inappropriate off-label use, there may be legitimate reasons for the federal government to be cautious about such regulation, particularly in light of the long-standing convention that states are the primary regulators of medical practice. See, e.g., David Orentlicher, Off-Label Drug Marketing, the First Amendment, and Federalism, 50 Wash. U. J.L. & Pol’y 89 (2016); Zettler, Indirect Consequences, supra note 115.

714   erika lietzan and patricia j. zettler same clinical effect.118 The generic approval and therapeutic equivalence rating, in turn, lead to automatic substitution under state law. That is, state law will either require or permit a pharmacist to substitute an equivalent generic drug when a patient presents a prescription for the corresponding brand product.119 In some states the patient is notified in advance and may decline the substitution, but in other states the substitution happens as a matter of course. Every state’s medical practice law allows the healthcare professional to specify, at the outset, that the pharmacist may not substitute a generic equivalent. This might be important to the healthcare professional if the drug has a narrow therapeutic index (meaning that the lowest effective dose is very close to the highest safe dose), and it can be essential if the patient has an allergy to an inactive ingredient used in the generic drug. But automatic substitution is more common, and it allows generic drugs to penetrate the market quickly.120 Although the FDA usually deems generic drugs therapeutically equivalent to their brand counterparts at the time of approval, the agency cannot deem a biosimilar biologic substitutable for its reference product without additional findings.121 Congress specified a supplemental showing due to concerns about the potential for immunogenicity if patients switch between proteins that are similar but not identical.122 The biosimilar applicant must show that the two products can be expected to produce the same clinical result in any given patient, and it must show that the risk of switching between the products is no greater than the risk of remaining on the original brand product.123 The FDA has indicated that it would like to see the results of clinical switching studies,124 but as the agency gains greater comfort with the safety of biosimilars, the requirements may evolve. In addition, it remains unclear whether interchangeability determinations will be important for biosimilar competition. Most biologics are administered by healthcare professionals or in hospitals, and, in these situations, payers decide which product the patient will receive. Automatic substitution may not play a role.

4.3  Controversies and Challenges 4.3.1  Incremental Innovation Versus Evergreening Innovation does not stop when the FDA approves a new medicine. A company might seek to improve the clinical profile of its product, for instance by developing a formula118 FDA, Approved Drug Products with Therapeutic Equivalence Evaluations (38th ed. 2018) at vii-viii. 119  New York v. Actavis, 787 F.3d 638, 645 (2d. Cir. 2015). 120  Murray L. Aitken et al., The Regulation of Prescription Drug Competition and Market Responses: Patterns in Prices and Sales Following Loss of Exclusivity 6–7 (Nat’l Bureau of Econ. Research, Working Paper No. 19487 2013), http://www.nber.org/papers/w19487.pdf. 121  42 USC. § 262(k)(4). 122  See Carver, Elikan, and Lietzan, supra note 53, at 733–34, 737–38, 751–52, 809. 123 Id. 124 FDA, Guidance for Industry: Considerations in Demonstrating Interchangeability with a Reference Product (draft) (Jan. 2017).

Regulating Medicines in the United States   715 tion or route of administration with a better safety profile or improved effectiveness. Some incremental innovations can lead to therapeutic options for previously untreated subpopulations; for instance, a new dosage form might be suitable for geriatric patients. For example, a patch or liquid might be helpful for patients unable to swallow tablets and capsules, and a monthly injection might be helpful for patients unable to remember daily oral medicines. Others might improve the convenience of the product: for instance, an extended-release formulation may reduce the frequency of dosing, which can improve patient compliance and, in turn, health outcomes. Antitrust regulators and others have raised concerns, however, about the potential for this incremental innovation to “evergreen” a new medicine’s exclusive position in the marketplace. Typically, the newer versions are protected by patents and regulatory exclusivity, which will preclude approval of generic copies of the newer versions. In theory, healthcare professionals could still write prescriptions for the older version of the innovative product, in which case pharmacies would generally dispense generic copies of that product. And they could in theory simply write prescriptions for the generic copies in the first instance. But instead healthcare professionals and patients might select the newer version of the innovative product. In this case, because there is no generic substitute for the newer product, the innovator will continue to enjoy exclusive sales. It may be appropriate for a company to enjoy a certain amount of continued exclusivity in the marketplace (i.e., to enjoy most of the sales) if it discovers and introduces newer and better versions of its products that payers and prescribers prefer to the older version and its generics.125 And US law generally assumes that prescribers and payers can make rational decisions about whether product differences justify the resulting increase in price. But, in some cases, the courts have found that an innovator inappropriately blocked generic competition (e.g., by withdrawing the first version of its product from the market before approval of any generic versions, which effectively defeated automatic substitution).126 Finding a solution that addresses anticompetitive behavior without stifling the incentive for incremental innovation remains an important challenge for US policymakers. Allegations of improper “evergreening” have been a fixture of US policy debate relating to medicine innovation for decades. The scope of the problem is disputed, but most policymakers believe that at least some improper behavior occurs. It remains to be seen whether antitrust law will be sufficient to address these situations. Congress has also considered acting, and some have argued that the FDA should take steps. Any solution will need to ensure continued incentives for incremental innovation that would benefit patients.

4.3.2  Lost Opportunities: New Uses of Established Medicines Policymakers face a different challenge with respect to another type of incremental innovation: the development of new uses for already approved medicines. Many in the 125  See, e.g., Dmitry Karshtedt, The More Things Change: Improvement Patents, Drug Modifications, and the FDA, 104 Iowa L. Rev. 1129 (2019). 126  New York v. Actavis, 787 F.3d 638 (2d. Cir. 2015).

716   erika lietzan and patricia j. zettler scientific and public health communities believe that medicines already on the market are likely to be safe and effective treatments for other diseases. The classic example is interferon: the FDA initially approved the medicine for treatment of hairy cell leukemia, but, 15 years later, a modified form (with polyethylene glycol attached) was, combined with ribavirin, the standard of care for hepatitis C.127 US law may not adequately incentivize new-use research, however, because of a basic conflict between the incentives, on the one hand, and prescribing and dispensing rules and practices, on the other hand. The “problem of new uses,” as it has been called, has vexed US policymakers for decades.128 Although the investment required for a new use is generally less than the investment required to develop a new molecular entity from scratch, it is still substantial. It can take 3–6 years and between $100 and $300 million to conduct the phase 2 and 3 trials needed to secure FDA approval of a new use.129 A company might be willing to invest this money in the years just after FDA approval of its new medicine. But as the patent and regulatory exclusivity on the initial approval draw to a close, the incentive to perform the research dwindles. A new use may be entitled to its own patent and regulatory exclusivity, but these protections apply only to the new use. The FDA will still approve a generic copy of the original product for the original uses. It will also designate the generic drug therapeutically equivalent to the innovative drug. As a result, generally, if a prescriber selects the innovative drug for the new use, a pharmacist will dispense the generic drug— despite the patent and regulatory exclusivity. Thus, once a new medicine is available in generic form for any use, the innovator has little incentive to study the compound any further. In theory another company could study the drug for new uses. But it typically faces the same problem. Although it may earn a new-use patent or regulatory exclusivity, healthcare professionals in the United States will be free to prescribe another product for the condition—including the original innovator’s off-patent product. And if generic substitutes for that product are available, pharmacists will dispense those products. The second innovator may never enjoy the sale. The social cost of these overlooked research opportunities may be enormous. Some research suggests that hundreds of already approved drugs have therapeutic potential that has never been explored.130 Moreover, because these medicines are no longer protected by patent or exclusivity and their initial research and development costs were long ago recouped by their initial developers, they could be considerably more cost127  Erika Lietzan, Paper Promises for Drug Innovation, 26 Geo. Mason L. Rev. 168, 176 (2018). 128  See generally Rebecca S. Eisenberg, The Problem of New Uses, 5 Yale J. Health Pol’y L. & Ethics 717 (2005). 129  Henry Grabowski & Jeffrey Moe, Impact of Economic, Regulatory, and Patent Policies on Innovation in Cancer Chemoprevention, Cancer Prev. Res. 82, 85 (July 2008); Benjamin N. Roin, Solving the Problem of New Uses (draft Oct. 14, 2016) (unpublished manuscript available at http://www. bu.edu/law/files/2016/10/Solving-the-Problem-of-New-Uses-Ben-n.-Roin.pdf) at 24 n.169. 130 Roin, supra 129, at 43–44 and n. 274; see also Sam F. Halabi, The Drug Repurposing Ecosystem: Intellectual Property Incentives, Market Exclusivity, and the Future of “New” Medicines, 21 Yale J. L. & Tech. 1 (2018).

Regulating Medicines in the United States   717 effective for patients and payers than new molecular entities for the same uses. Some have suggested better targeted incentives for new-use research. Others have suggested public funding for the research. In addition, if the US healthcare system transitioned to indication-based prescribing, dispensing, and pricing, it might be possible for new-use innovators to recover their investments—which would solve the problem.

5  Challenges to the Traditional Paradigm The FDCA and PHSA date to the early part of the 20th century, and the basic definitions of “drug” and “biological product” have changed little since that time. These concepts date to an era in which a medicine was a discrete tangible item—tablets or a liquid for ingestion or perhaps injection—that changed hands, typically from a pharmacist to a patient. Scientific progress since that time has changed medical practice and the concept of “medicine” so fundamentally that the FDA and the courts sometime struggle with how to classify new interventions and even with whether the FDA may (or should) regulate them. We discuss three examples here: gene therapy products, cell- and ­tissue-based products, and microbiome-based products.

5.1  Gene Therapy Products More than 20 years ago, the FDA defined the phrase “gene therapy” as the administration of genetic material to modify or manipulate the expression of a gene product or to alter the biological properties of living cells for therapeutic purposes.131 The agency has applied this definition to a variety of medical treatments involving use of genetic technology. For instance, it includes voretigene neparcvovec-rzyl (Luxtuma), approved for treatment of an inherited form of vision loss that stems from mutation of one gene. Luxturna works by delivering a normal copy of the gene directly to retinal cells, which then produce a normal protein that converts light to an electrical signal, thus restoring the patient’s vision.132 The concept also includes tisagenlecleucel (Kymriah), which instead genetically modifies a patient’s healthy cytotoxic T cells, a type of white blood cell that destroys tumor cells in the body, so that the T cells will recognize and destroy tumors with specified cell surface antigens.133 Thus, at the FDA, “gene therapy” includes not only addressing existing genetic mutations but genetically modifying normal cells to treat disease. 131  58 Fed. Reg. 53247, 53249 (Oct. 14, 1993). 132  FDA News Release, FDA approves novel gene therapy to treat patients with a rare form of inherited vision loss (Dec. 19, 2017). 133  See generally, Chimeric Antigen Receptor (CAR) T-Cell Therapy, 3 JAMA Oncology 1595 (2017).

718   erika lietzan and patricia j. zettler Every gene therapy product requires an approved BLA. But it may not be possible or appropriate for the manufacturer to conduct a traditional premarket research program. These products are often intended for rare diseases, which limits the number of subjects that can be enrolled in clinical trials. If the treatment effect is both large and self-evident, however, that can eliminate the need for large numbers of subjects and may also allow the manufacturer to dispense with controls or blinding. To date, the approved products have had comparatively small premarket clinical programs. For instance, the FDA approved axicabtagene ciloleucel (Yescarta), another immunotherapy product that genetically modifies a patient’s T cells, on the basis of a single study in 111 patients that (a) had the features and goals of both a phase 1 study and a phase 2 study, but not a phase 3 study, and (b) was “open label” (everyone participating was aware what the patients were receiving) and did not have a control arm.134 Approvals like this illustrate that the statutory standard for approval of a new medicine can be flexible when needed, but the agency’s use of its flexibility can often draw criticism—particularly if patients feel the flexibility has been unfairly withheld for a promising new treatment or that it has been used to let an ineffective product on the market.

5.2  Cell- and Tissue-Based Products Many medical therapies contain or consist of human cells and tissue. Blood, for instance, consists of human cells. Tissues from cadavers intended for transplantation—ligaments, heart valves, and corneas, for example—consist of human cells. Vascularized organs for transplantation consist of human cells. The gene therapy product discussed in the preceding paragraph is cellular. Another example is treatment with a patient’s own stem cells after the cells have been removed and modified in some fashion. Other examples include epithelial cells grown on a synthetic matrix, and semen and other reproductive tissue. The FDA’s approach to cellular and tissue-based treatments varies. Some, such as gene therapy products, require approved marketing applications, as just described. Others, such as vascularized human organs for transplantation, are not regulated by the FDA at all. The FDA has also created a special category—human cells, tissues, and cellular and tissue-based products (HCTPs)—that are subject to regulatory oversight but do not require premarket approval. Only certain cell- and tissue-based products qualify for regulation as HCTPs.135 To qualify, a product must have been minimally manipulated—thus, for instance, processing cannot alter its relevant biological characteristics. It must also be intended for homologous use, meaning that the tissue or cell must be intended to perform the same basic function in the recipient (patient) as in the donor. The manufacturer cannot combine the cells or tissues with other items or substances (with limited exceptions, such as water and preserving agents). Most HCTPs—such as ligaments from cadavers—do not 134  Summary Basis for Regulatory Action, BLA 125643 (Oct. 18, 2017). 135  21 C.F.R. § 1271.10(a).

Regulating Medicines in the United States   719 have a systemic effect and are not dependent on the metabolic activity of living cells. If the HCTP has a systemic effect or relies on metabolism, then it must be taken from and returned to the same person or provided to certain close relatives unless it is intended for reproductive use (e.g., it is an donated embryo). If a product qualifies as an HCTP, the manufacturer must register and list the product with the FDA, screen and test donors, prepare and follow written procedures for the prevention of the spread of communicable diseases, and maintain records.136 A product that does not qualify as an HCTP generally requires an approved BLA.

5.3  Microbiome-Based Products Microbiome-based products, currently in their infancy, are likely to present a wide range of challenges for policymakers.137 In the past decade, fecal microbiota transplantation (FMT)—which involves transplanting a filtered stool preparation from a healthy donor to the lower gastrointestinal tract of the patient—has emerged as a potential solution for recurrent Clostridium difficile infections, which can be life-threatening. The procedure is widely used, the cure rate may significantly outstrip even the best antibiotics, and increasing resistance to approved antibiotics makes the prospect of microbiome-based therapy particularly important. Many healthcare providers perform the procedure with a friend-or-family donor, but several “stool banks” now ship fecal microbiota for clinical use. The FDA takes the position that the material transplanted into the patient qualifies as a medicine subject to its jurisdiction, requiring either an effective IND or an approved marketing application. The stool banks generally ship their products under INDs, although they have no plans to perform pivotal trials and seek marketing authorization. The agency exercises enforcement discretion with respect to other transplants (e.g., friend-or-family donors) if intended for patients with C. difficile infection not responding to standard therapy.138 Meanwhile, several pharmaceutical companies are developing fecal microbiota products in conventional pharmaceutical dosage form, with a view to completing traditional phase 3 trials and securing market authorization. Complicating the policymaking landscape considerably, patient support groups and online forums provide detailed do-it-yourself instructions online for patients and family caregivers. Although fecal microbiota transplantation has been more exhaustively researched and is reasonably well known in clinical practice, other microbiome-based treatments are also under investigation. The human body has many distinct microbiomes—in the mouth and in the vagina, for instance, as well as on the skin—and many diseases may be 136  21 C.F.R. Parts 1270 and 1271. 137  Rachel Sachs & Carolyn Edelstein, Ensuring the Safe and Effective FDA Regulation of Fecal Microbiota Transplantation, 2 J. Law & Biosciences 396 (2015). 138 FDA, Guidance for Industry: Enforcement Policy Regarding Investigational New Drug Requirements for Use of Fecal Microbiota for Transplantation to Treat Clostridium difficile Infection Not Responsive to Standard Therapies; (March 2014).

720   erika lietzan and patricia j. zettler associated with dysbiosis or productively treated with transplants. This field of inquiry presents significant challenges for federal policymakers, however, for several reasons. First, microbiome-based treatments may be discovered, developed, and used first by healthcare professionals in clinical practice. Although the FDA takes the view that ­generally it may still regulate the treatments as medicines, asserting authority over healthcare professionals performing friends-and-family procedures is likely to be controversial. Second, if microbiome-based treatments are quickly understood to be highly effective and if they seem safe, then patients and caregivers are likely to engage in selfhelp. They might do so simply to avoid the cost (or, depending on the procedure, the embarrassment) of an office procedure or product. If the FDA’s assertion of authority in this area leads to regulatory barriers and thus cost barriers, patients will be more likely to resort to self-help. Third, microbiome banks may also emerge for the same reasons, particularly with the support of patient advocacy groups. The storage and interstate shipment of microbiome samples and the use of microbiomes from strangers increases safety concerns, including concerns about communicable diseases. But these entities are unlikely to pursue conventional clinical trials and marketing approval, which presents a quandary about the appropriate regulatory framework to apply as well as a question about whether they should be allowed to operate indefinitely. Fourth, if the FDA permits healthcare professionals and stool banks to proceed with minimal regulation, then the public health community will obtain rigorous clinical safety and effectiveness data on the use of microbiota to treat disease only if some other entity pays for and performs the trials. But it is not clear that US law provides an incentive for anyone to do so. And we may not benefit from more conventionally designed pharmaceutical products unless the companies considering this research believe they will enjoy exclusivity in the marketplace after approval. In turn this may mean taking enforcement action against healthcare professionals who have operated without oversight, and it may mean revoking any authority for microbiome banks to ship products. This, in turn, would be highly controversial because it would reduce patient access to affordable effective medicines, and it could drive more patients to less safe self-help procedures. Medical research may continue to move away from the traditional pills and liquids of the past century and toward novel and highly personalized interventions that are arguably as much a medical service as a tangible product. In the end, the decision whether to assert jurisdiction over microbiome-based interventions, autologous stem cell interventions, and similarly personalized service-like treatments requires a difficult judgment call about the balance between protecting consumers and incentivizing research, on the one hand, and giving healthcare providers sufficient autonomy to practice medicine as they see fit for the individual needs of their patients, on the other. And, if they choose to exercise jurisdiction, policymakers in the United States will face a wide range of regulatory challenges ranging from the suitability and feasibility of the traditional three-phase premarket research and development program to the meaningfulness of research incentives that focus on exclusivity in the marketplace.

chapter 35

R egu l ati ng M edici n es i n the Eu rope a n U n ion Aurélie Mahalatchimy

This chapter describes the framework for regulation of medicines in the European Union (EU). The overall goal of this framework is to ensure that medicines are of good quality, safe, and effective. This is accomplished through extensive regulation of medicines throughout their lifecycle, from early research and development through to marketing authorization. In European contexts, medicines regulation interfaces in important ways with authorization for reimbursement within national health systems. In practice, without such reimbursement authorization, patients will not be able to access medicines even if those medicines have formal marketing authorization for the EU. Reimbursement decisions consider relative therapeutic benefits, whereas marketing authorizations consider only quality, safety, and efficacy.1 Although the overall goals of the European and US medicines’ frameworks are the same, the structure of the EU framework is complicated by the fact that both EU-­level and Member State-­level authorities have responsibility for regulation of medicines. For instance, some marketing applications must be reviewed centrally by the European Medicines Agency (EMA) (and, if they are approvable, will be approved by the European Commission). Overall, EU law is dominant from clinical research to marketing au­thor­i­za­tion while national law is dominant for healthcare services and reimbursement issues.2 At the EU level, the EMA is the European counterpart to the US Food and Drug Administration (FDA) although, unlike the FDA, the EMA is not competent for medical devices. Each Member State also has its own medicines regulatory authority, known as a 1  Space precludes a full discussion of reimbursement laws, which are nationally determined in Europe. 2  Reimbursement issues are discussed in the chapter by den Exter and Syrett, ‘Access to Health Care in Europe,’ in this volume.

722   Aurélie Mahalatchimy national competent authority. For example, the French government includes L’agence nationale de securité du médicament et des produits de santé (ANSM) and the Italian government includes Agenzia italiana del farmaco (AIFA). As this chapter illustrates, responsibility for some aspects of the broader medicines regulatory framework falls to these national authorities.

1  Defining Medicines As a threshold matter, the regulatory framework described in this chapter applies to medicinal products—here, ‘medicines’—intended for human use. The scope of the term ‘medicines’ has evolved over time, through both jurisprudence (judicial decisions) and revisions of the key European legislation—Directive 2001/83/EC.3 These changes have taken into account the emergence of new therapies as well as the challenges presented by ‘borderline products’—those that have characteristics of both medicines and other regulated products. For instance, a product with the claim ‘promotes hair growth’ is usually considered a medicine, especially when it contains a substance that is prohibited as a cosmetic ingredient, such as minoxidil. However, a product with the claim ‘reduces hair loss’ is usually considered a cosmetic product. Another example could be dental pastes: some are considered as medicines while other are medical devices and most of them are cosmetics, depending on their mode of action. Under EU law, a product will be classified as a medicine by virtue of either its presentation or its function. Both national and EU authorities can make this determination. First, a medicine is ‘any substance or combination of substances presented as having properties for treating or preventing disease in human beings.’ This generally entails considering the looks of the product, including the claims made about the product. The claims may be explicit or implicit, including, for example, testimonials from medical practitioners commending the qualities of the product or even simply references to research by pharmaceutical laboratories or to methods or substances developed by medical practitioners. Other evidence, including the product’s form and packaging could give rise to classification as a medicine.4 For instance, in France, the French medicines agency (ANSM) has considered injectable skin lightening products medicines by presentation because their sales presentation online (high-­risk administration presented online: ‘injection into a vein or muscle or under the skin’) can lead the average consumer to think it is a medicine. Second, a medicine includes substances used to restore, correct, or modify physiological functions or to make a medical diagnosis. This entails considering the product’s function, including its use. Thus, the national competent authority considers all of the characteristics of the product, including its composition, 3  Directive 2001/83/EC on the Community code relating to medicinal products for human use, OJ L 311/67-­128. This chapter will discuss different parts of the Directive throughout. 4  See ECJ, 21 March 1991, Delattre, aff. C-­369/88, ECR I-­1487, ¶ 41.

Regulating Medicines in the European Union   723 its pharmacological properties, the manner in which it is used, the extent of its distribution, its familiarity to consumers, and the risks associated with its use. For instance, in France, the ANSM has considered two massage oils (HL2 oléokinum and TRH5) not as cosmetics but as medicines by function because they are very concentrated in essential oils that give them pharmacological properties. Within the broad category of medicines, EU law also defines various specific types of medicine, such as biological medicines,5 blood- or plasma-­derived medicinal products,6 immunological medicinal products,7 radiopharmaceuticals,8 homeopathic medicinal products,9 and herbal medicinal products.10 Some specially defined categories are subject to their own regulatory frameworks: orphan medicines (discussed in section 4.2),11 medicines for paediatric use (discussed in section 4.2), and advanced therapy medicinal products12 (discussed in section 5). But all of these are ‘medicines’ in the first instance, either because of their presentation or because of their function. In some cases, a product may have features of both a medicinal product and another type of regulated product, such as a food, novel food,13 cosmetic, medical device, in vitro diagnostic medical device,14 or biocidal products.15 EU law refers to this as a ‘bord­ er­line product.’ In the first instance, national authorities or European agencies determine the appropriate categorization in borderline cases, applying the definitions of the different product categories, subject to supervision by the courts (national and European). The European Commission has published several guidance documents to 5  A biological medicinal product has an active ‘biological substance,’ which in turn is ‘produced by or extracted from a biological source.’ Immunological medicines and medicines derived from human blood and human plasma are also considered biological medicinal products. 6  These products include albumin, coagulating factors, and immunoglobulins of human origin. 7  An immunological medicinal product is one consisting of vaccines, toxins, serums, or allergen products. 8  A radiopharmaceutical is a medicinal product that, when ready for use, contains one or more radionuclides (radioactive isotopes) included for a medicinal purpose. 9  A homeopathic medicinal product is one prepared from a substance known as a ‘homeopathic stock’ in accordance with a homeopathic manufacturing procedure described by the European Pharmacopoeia or a pharmacopoeia used officially in the Member States. 10  An herbal medicinal product is one containing, as active ingredients, only herbal substances, herbal preparations, or a combination of the two. 11  These are medicines that are intended to treat rare diseases or which are not otherwise expected to generate sufficient return in the marketplace to justify their development. 12  These medicines are based on genes, cells, or tissues of human or animal origin. 13  ‘Novel food’ is food not very often used for human consumption in the EU before 15 May 1997. A wide range of products is covered, such as food with a new or modified molecular structure, food using a new food production process (bread treated with UV-­light to increase vitamin D content), or food produced from microorganisms, fungi, or algae. Regulation (EU) 2015/2283 on novel foods OJ L 327/1–22. 14  They are devices used in vitro for the examination of human biological elements such as blood or tissue, or for providing information, such as on a physiological or pathological process or state or on a congenital physical or mental impairment. This covers a wide range of devices such as self-­tests for pregnancy or tests for highly transmissible substances using specimens taken from the human body. 15  These products, such as household disinfectants, insecticides, and other chemicals, are used to protect materials or to suppress pests (such as parasites, fungi, or bacteria).

724   Aurélie Mahalatchimy facilitate this determination and to help manufacturers determine which rules will apply. It can also be helpful for companies to have early contact with regulators, especially the European Medicines Agency (EMA). EU law also specifies that, in cases of doubt, the rules governing medicinal products should apply. In that context, the broad interpretation of the ‘medicine by presentation’ protects consumers from products which do not have the effectiveness that they are entitled to expect. On the other hand, the definition of ‘medicine by function’ covers products ‘the pharmacological properties of which have been scientifically observed and which are genuinely designed to make a  medical diagnosis or to restore, correct or modify physiological functions.’16 Consequently, this ‘rule of doubt’ does not apply if it is merely possible that the product could be classified as a medicinal product, if it has not been ‘scientifically established’ that the product is a medicinal product by function—in other words, if it has not been established that the product is capable of restoring, correcting, or modifying physiological functions or can be used to make a medical diagnosis.17 Nevertheless, the appropriate classification of some borderline products remains unresolved. For instance, it is not yet clear how microbiome-­based products should be classified. These products contain live bacteria and include fecal microbiota, currently used to treat recurrent Clostridium difficile infections. Although the US FDA has concluded that microbiota are biological products and drugs when intended to treat illness, there is no consensus yet in the European Union. They might be classified as medicinal products, but they might also be classified as medical devices or food supplements depending on their mechanism of action, characteristics, or claimed intended treatment.18

2  Bringing Medicines to the Market 2.1  Standard Approval Process No medicine may be lawfully marketed in the European Union without marketing authorization from either the national competent authority or the European Commission. There are four possible procedures. One is national: authorization only at the individual Member State level. Three are European: authorization that directly operates Europe-­wide (the ‘centralized’ authorization procedure which is mandatory for some medicines), and two types of authorization that involve Member States

16  See Case C-­319/05 Commission v Germany [2007] ECR I-­9811, ¶ 61. 17  See Case C-­140/07 Hecht-­Pharma GmbH v Staatliches Gewerbeaufsichtsamt Luneburg [2009] ECR I-­00041, ¶ 29. 18  Nino Mihokovic, ‘EMA Experience and Perspective, Regulatory Challenges in the Drug-­Food Continuum,’ IMI Stakeholder Forum 2017, Microbiome forum 2017 (powerpoint slides) (https://www. imi.europa.eu/sites/default/files/events/SF2017/microbiome_Mihokovic.pdf).

Regulating Medicines in the European Union   725 r­ elying on each other’s assessments (the ‘decentralized’ procedure and the ‘mutual recognition’ procedure). First, certain products must—and others may—be reviewed under the centralized authorization procedure, governed by Regulation 726/2004.19 Under this procedure, a company submits a single market authorization application to the EMA. The application is reviewed by the relevant scientific committees within the EMA, typically the Committee for Medicinal Products for Human Use (CHMP). The CHMP may consult other committees, and one of its Scientific Advisory Groups (SAG), groups of external experts comparable to FDA advisory committees in the United States, to provide advice.20 The EMA’s evaluation should ordinarily take 210 days, following which it provides a recommendation (or ‘opinion’) on the approval decision.21 The European Commission makes the final decision, generally following the opinion of the EMA, and its decision is binding throughout Europe. In other words, medicinal products authorized under the centralized procedure may be marketed in all EU Member States as well as European Economic Area (EEA) and European Free Trade Association (EFTA) States—thus also Iceland, Liechtenstein, Norway, and Switzerland. The centralized procedure is mandatory for some medicines: medicines derived from biotechnology (including notably medicines based on recombinant DNA technology), orphan medicines, advanced therapy medicinal products, and biosimilar medicines. It is also mandatory for any medicine that contains a new chemical, biological, or radiopharmaceutical active substance and that is intended for the treatment of AIDS, cancer,22 a neurodegenerative disorder,23 diabetes,24 an autoimmune disease or other immune dysfunction,25 or a viral disease.26 The centralized procedure is optional for 19  OJ L 136/1–33. 20  As in the United States, these expert consulting groups are classified by therapeutic area: cardiovascular issues, anti-­infectives, diabetes and endocrinology, HIV and viral diseases, neurology, oncology, psychiatry, and vaccines. 21  In 2018, EMA’s recommendations for marketing authorizations included 84 positive opinions and 5 negative opinions. EMA, ‘Human Medicines Highlights,’ 2018 (https://www.ema.europa.eu/en/ documents/report/human-­medicines-­highlights-­2018_en.pdf). 22  For instance, telotristat ethyl (Xermelo) was authorized on 17 September 2017 in the European Union. It contains the active substance telotristat. It is used to treat adults with severe diarrhoea associated with a carcinoid syndrome occurring when certain tumours produce and release too much serotonin into the blood. 23  For instance, opicapone (Ongentys) was authorized on 24 June 2016 in the European Union. It contains the active substance opicapone. It is used to treat adults with Parkinson’s disease, a progressive brain disorder causing shaking, muscle stiffness, and slow movement. 24  For instance, dapagliflozin/metformin (Xigduo) was authorized on 16 January 2014 in the European Union. It contains the active substances metformin hydrochloride and dapagliflozin propanediol monohydrate. It is used together with diet and exercise to control blood glucose (sugar) levels in adults with type 2 diabetes. 25  For instance, alemtuzumab (Lemtrada) was authorized on 12 September 2013 in the European Union. It contains the active substance alemtuzumab. It is used to treat adults with relapsing-­remitting multiple sclerosis, a disease where inflammation destroys the protective sheath surrounding the nerve cells. 26  For instance, elbasvir/grazoprevir (Zepatier) was authorized on 22 July 2016 in the European Union. It contains the active substances elbasvir and grazoprevir. It is used to treat adults with chronic hepatitis C (infectious disease affecting the liver) caused by the hepatitis C virus.

726   Aurélie Mahalatchimy medicines containing a new active substance for a different indication, as well as for other medicines that constitute a significant therapeutic, scientific, or technical innovation, and medicines for which EU-­level authorization is in the interests of public health. The logic behind this distinction between compulsory and optional centralized procedure can be considered multiple. First, biotechnological medicines were accepted and disseminated in Europe thanks to a legal common approach and a high level of expertise in the EU (legal harmonization in this field began in the 1990’s). Moreover, their cost of development as well as the rarity of the targeted disease for orphan drugs do not allow that they are restricted to few national markets. Thus, specific attention and interest is given to biotechnological medicines. Furthermore, the optional pathway allows, on the one hand, other medicines which are not specifically named and framed and that constitute therapeutic innovation to benefit from the centralized procedure if their manufacturers consider it to be helpful, as for the manufacturers of biotechnological medicines. On the other hand, it exists to give Member States the possibility of authorizing at national level the generic form of medicines authorized at the European level, as long as each EU Member State has its own national policy to govern the generics sector, especially their pricing and reimbursement. Second, the decentralized procedure is available for medicines that have not been authorized in any individual Member States and for which the centralized procedure is not compulsory. The applicant submits identical applications to the competent national authorities in several Member States, seeking access to their markets. It identifies one Member State to act as the ‘reference Member State’ and thus take the lead in the evaluation. The theory, in other words, is that the Member States can rely on the scientific evaluations and judgments of each other’s national competent authorities. The other Member States will consider the marketing application, but the reference Member State has the primary responsibility for assessment. If the reference Member State issues a positive assessment and there is disagreement between the Member States, a Coordination Group for Mutual Recognition and Decentralized Procedures for human medicines (CMDh) considers the issue.27 Disagreement must be based on a ‘potential serious risk to public health.’ If the CMDh in turn cannot reach consensus within 60 days, the matter goes to the EMA (specifically, the CHMP) for arbitration.28 The CHMP adopts an opinion within 60 days, and the European Commission takes the final binding decision. Third, the mutual recognition procedure is available if a medicinal product has already been authorized in one EU Member State. The theory here is essentially the same: the Member States rely on each other’s assessments and judgments. The first Member State to authorize the medicine is considered the ‘reference Member State’ and 27  This group is composed of one representative per Member States, plus Norway, Iceland, and Liechtenstein. In 2018, only 5 out of 1,023 decentralized procedures (0.5%) were referred to CMDh. CMDh Statistics 2018, op. cit. 28  In 2018, half of the procedures (6/12) referred to the CMDh went to the EMA for CHMP arbitration. However, it is not specified whether they were related to decentralized procedures, mutual recognition procedures, works sharing, or repeat use. CMDh Statistics 2018, op. cit.

Regulating Medicines in the European Union   727 provides its evaluation of the application to the other Member States selected by the applicant. As is true when an applicant selects the decentralized procedure, if a country refuses to recognize the original national authorization (on the ground of a potential serious risk to public health), the issue is taken to the CMDh.29 If the CMDh cannot reach consensus, the matter goes to the EMA for arbitration. Finally, it is possible to secure approval simply at the Member State level by applying to the national competent authority of each Member State. In practice, this really means 27 national procedures for 27 national markets. Medicines manufacturers will use the national procedure when medicines are not in the scope of the centralized procedure. It is particularly used for generics given the impact of their pricing on national healthcare systems, an area that is regulated at the national rather than the EU level. The national procedure can be combined with the mutual recognition procedure.30 Although the four licensing procedures are different, the substantive requirements for approval are similar. In each case, as in the United States, approval is based on showing, during an open assessment, that the medicine meets the three criteria of quality, safety, and efficacy, provides a positive benefit/risk ratio, and complies with benchmarks: the Good Manufacturing Practices (GMP)31 and the pharmacopeia.32 As in the United States, the marketing application contains preclinical and clinical data, as well as detailed ‘quality’ information about the chemistry and manufacturing of the medicine. Indeed, the substantive requirements for marketing authorization in Europe are largely harmonized with the drug and biologic approval requirements in the United States. An applicant in Europe organizes its application in accordance with the Common Technical Document (CTD), which was designed to provide a common format for marketing applications in Europe, the United States, and Japan. Generating the safety and efficacy data necessary for marketing authorization is a multistep process that begins with laboratory studies and culminates in large randomized controlled clinical trials (phase 3 trials) in patients. The EMA and national competent agencies in the Member States regulate preclinical testing (laboratory and animal

29  In 2018, only 4 out of 291 mutual recognition procedures (1.4%) were referred to CMDh. CMDh Statistics 2018, op. cit. 30  For instance, in France in 2017, the French National agency for medicines (ANSM) delivered 955 marketing authorizations, including 801 for generics, 305 in accordance with national procedures, 606 in accordance with decentralized procedures, and 44 in accordance with mutual recognitions procedures. ANSM, Rapport d’activité 2017, 105 (https://www.ladocumentationfrancaise.fr/var/storage/ rapports-­publics/184000668.pdf). 31  GMPs are a benchmark of various principles and guidelines to be respected in order to guarantee the highest standards of quality, as a prerequisite of safety and efficacy in any process that involves the manufacture of medicines. They include basic requirements for a pharmaceutical quality system, personnel, premise and equipment, documentation, production, quality control, outsourced activities, complaints and product recall, and inspection, as well as specific guidelines on the manufacturing of particular medicines, such as radiopharmaceuticals, biological active substances, sterile medicines, etc. 32  It provides a legal and scientific reference for the quality control of medicines, including, notably, a set of appropriate tests to confirm their identity and purity and ascertain the strength of their active substance and their performance characteristics.

728   Aurélie Mahalatchimy testing) through good laboratory practices (GLP).33 In addition, testing in animals must comply with European law on the protection of animals used for scientific purposes. After preclinical testing indicates that it is safe to proceed with testing in humans, the manufacturer applies for permission to conduct clinical trials. Clinical trial regulation in Europe is in a state of transition. At the time of writing, clinical trials were governed by the EU Clinical Trial Directive, with national legislation put in place to implement the directive.34 A new Clinical Trial Regulation, however, will soon fully take effect.35 Once it takes effect, manufacturers will apply for permission to start clinical trials using a single point of entry: an EU clinical trial portal. The new Regulation also introduces an assessment procedure leading to a single decision for the EU Member States, as well as rules on the protection of subjects and informed consent and transparency requirements. The Regulation will take effect 6 months after the European Commission publishes confirmation that the portal and the EU-­associated database on clinical trials are fully functional. Technological challenges relating to this portal, the relocation of the EMA from London to Amsterdam due to Brexit, and the Covid-­19 pandemic have caused some delay, but the EMA has recently confirmed that the system is on track to go live by 31 January 2022. As a ‘regulation,’ the Clinical Trial Regulation will apply immediately throughout the EU without the need for transposition into national laws. Even after the Clinical Trial Regulation takes effect, though, the individual Member States will have responsibility for authorizing and overseeing clinical trials within their borders. In accordance with the Clinical Trial Regulation, if a company intends to perform its trial in more than one Member State, it will designate a ‘reporting Member State.’ This country has responsibility for preparing a Part I assessment of the clinical trial application, focusing on the scientific and medical aspects of the trial. This assessment considers, for example, the information known about the investigational medicine and the risks to trial subjects. The reporting Member State circulates the Part I assessment to the other involved countries for a coordinated review and final consolidation by the reporting Member State. Every clinical trial is also subject to ethical review by an ethics committee in accordance with the law of the Member State in which it will occur. Thus, each involved Member State also conducts its own Part II assessment of the ethical 33  GLPs are a benchmark of various principles and guidelines to be respected in preclinical testing through analytical, toxicological, and pharmacological experiments in order to promote the quality and validity of the data generated in the testing of medicines. 34  Directive 2001/20/EC on good clinical practice in the conduct of clinical trials, OJ L 121/34–44). The Clinical Trial Directive does not apply to non-interventional trials, and the new Clinical Regulation will not apply to non-interventional studies. A non-interventional trial is defined as ‘a study where the medicinal product(s) is (are) prescribed in the usual manner in accordance with the terms of the marketing authorization. The assignment of the patient to a particular therapeutic strategy is not decided in advance by a trial protocol but falls within current practice and the prescription of the medicine is clearly separated from the decision to include the patient in the study. No additional diagnostic or monitoring procedures shall be applied to the patients and epidemiological methods shall be used for the analysis of collected data’ (Directive 2001/20/EC, art. 2c). Non-interventional study means ‘a clinical study other than a clinical trial’ (Regulation (EU) 536/2014, art. 2(4)). 35  Regulation (EU) 536/2014 on clinical trials, OJ L 158/1–76.

Regulating Medicines in the European Union   729 aspects of the trial. Following assessment of the clinical trial application, the sponsor is notified by way of a single decision sent through the EU portal. Ordinarily, this decision must be made within 60 days of the submission of the application. As in the United States, clinical trials must be conducted in compliance with good clinical practices (GCP). This means that the trial’s sponsor and investigators must comply with a variety of rules designed to ensure the protection of human subjects and the validity and reliability of the clinical trial data. EU law requires an investigator to obtain informed consent from research participants, for example, and to report adverse events to the sponsor. A national regulatory agency may suspend or terminate a clinical trial if the criteria for proceeding are no longer met (e.g., if research participants face unreasonable risks).36 The Clinical Trial Regulation also created a new category—the low-­ intervention clinical trials—which pose minimal risk to subjects compared to normal clinical trials and for which modified risk procedures are appropriate.37 A new marketing authorization is valid for 5 years. After the first 5-­year period, the authorization can be renewed indefinitely or—if appropriate, given safety concerns— renewed for another 5-­year period. Authorization can be suspended or withdrawn by the competent authorities at any time if the legal requirements are no longer satisfied.

2.2  Expediting Access The standard process for securing marketing authorization is time-­consuming and expensive, which can significantly delay patient access to new medicines. As a result, several mechanisms to expedite access to new medicines have been developed. They fall in three categories: formal legal alternatives to the standard marketing authorization procedure, programs sponsored by the EMA to speed the development process, and mechanisms allowing patients access to medicines before marketing authorization. All of them involve difficult decisions as they are characterised by uncertainty regarding the benefit/risk ratio.

2.2.1  Alternative Marketing Authorization Procedures Regulation (EC) 726/2004 establishes three additional legal procedures to accelerate access to the market for products subject to centralized authorization. Individual

36  Regulation (EU) 536/2014. 37  A low-­intervention clinical trial studies an already authorized medicine, the use of which follows the terms of the marketing authorizations or published scientific evidence. Any additional procedures must not pose more than minimal additional risk or burden to the safety of the participants compared to normal clinical practice. Regulation (EU) 536/2014, Art. 2.3. In these cases, a risk-­proportionate approach to clinical trial practice may be warranted. See ‘Risk Proportionate Approaches in Clinical Trials: Recommendations of the Expert Group on Clinical Trials for the Implementation of Regulation (EU) 536/2014,’ 25 April 2017 (https://ec.europa.eu/health/sites/health/files/files/eudralex/ vol-­10/2017_04_25_risk_proportionate_approaches_in_ct.pdf).

730   Aurélie Mahalatchimy Member States also have mechanisms for accelerating the process for national marketing authorizations, which are not described here. First, in some cases conditional marketing authorization may be available. This procedure is available for medicines intended to treat, prevent, or diagnose seriously debilitating or life-­threatening diseases, as well as ‘orphan medicinal products’—those intended to treat rare disease. In the first 10 years after the conditional marketing authorization regulation was adopted, the European Commission granted 30 out of 52 requests for conditional marketing authorization. Of these, 11 were converted into standard au­thor­i­ za­tions, 2 were withdrawn for commercial reasons, and 17 are still conditional. In 2018, only one medicine obtained a conditional marketing authorization: rucaparib (Rubraca) to treat relapsed or progressive ovarian cancer. Conditional marketing authorization is also available in emergency situations and  in response to public health threats. Pandemic influenza vaccines have been approved via this pathway. In these cases, although the applicant lacks the comprehensive safety and effectiveness data ordinarily required for marketing authorization, the European Commission will grant authorization for renewable 1-­year terms. The European Commission will grant this authorization if (1) the medicine’s benefit-­risk ­balance is positive, (2) it is likely the applicant will be able to provide comprehensive clinical data, (3) the medicine will fulfil unmet medical needs, and (4) the benefit to the public health of the medicine’s immediate availability outweighs the risk inherent in the fact that additional data are still required. A conditional authorization can be converted to a standard authorization once the applicant submits the data that were lacking. Conditional marketing authorization has no equivalent in the United States, although the US ‘accelerated approval’ pathway plays a similar role for similar drugs.38 For instance, brentuximab vedotin (Adcetris), a lymphoma drug, was approved under conditional marketing authorization by the EMA and through the accelerated approval pathway by the FDA. Second, the European Commission can grant marketing authorization under exceptional circumstances. This procedure—which allows for marketing authorization without comprehensive data and without any expectation of comprehensive data in the future—has no equivalent in US law. An applicant may be eligible for marketing under exceptional circumstances if (1) the intended condition is so rare that gathering the data will not be possible,39 (2) comprehensive information cannot be provided given the current state of scientific knowledge,40 or (3) the collection of full information would be 38  See chapter by Zettler and Lietzan, ‘Regulating Medicines in the United States,’ in this volume. 39  For instance, velmanase alfa (Lamzede) was approved through this path. It is a medicine used for patients with mild to moderate alpha-­mannosidosis, a rare, inherited disease with features that include notably learning disability and difficulty controlling movement. 40  For instance, vestronidase alfa (Mepsevii) is a medicine to treat mucopolysaccharidosis type VII (also known as Sly syndrome), an inherited disease caused by a lack of an enzyme needed to break down complex carbohydrates known as glycosaminoglycans; it has been authorized under ‘exceptional circumstances’ both because the disease is rare and because in the present state of scientific knowledge, comprehensive information could not be provided.

Regulating Medicines in the European Union   731 unethical.41 A marketing authorization under exceptional circumstances is valid for 5 years and can be renewed after 5 years for an unlimited period, but the medicine’s benefit-­risk balance is reassessed annually. Unlike the conditional marketing au­thor­i­za­ tion, marketing authorization under exceptional circumstances does normally not lead to the completion of a full dossier or become a standard marketing authorization. There were six opinions recommending marketing authorization under exceptional circumstances between 2015 and 2018. Third, some marketing authorization applications are eligible for accelerated assessment at the EMA. In these cases, assessment occurs within 150 days instead of the usual 210 days. This accelerated review, which is equivalent to the ‘priority review’ in the United States,42 is available for medicines that are ‘of major interest from the point of view of public health and in particular from the viewpoint of therapeutic innovation.’ Applicants must request accelerated assessment, however. For instance, the CHMP agreed to the applicant’s request for an accelerated assessment of emicizumab (Hemlibra), a medicine used to prevent or reduce bleeding in patients with hemophilia A, as it was considered to be of major public health interest. Between 2013 and 2017, 58  requests for accelerated assessment were accepted, and 32 were rejected.

2.2.2  EMA Regulatory Programs The EMA has established two regulatory programs intended to expedite access to medicines in Europe by assisting with the research and development process. First, in March 2016, it launched PRIority MEdines (PRIME) to provide supplementary support for certain priority medicines. A medicine is eligible for PRIME if early clinical data show its potential to benefit patients with unmet medical needs, which in turn means it might offer a major therapeutic advantage over existing treatments or benefit patients without treatment options. Once a medicine has been selected for PRIME, the developer benefits from early dialogue and regular exchanges with the EMA. The goal is to help the company optimize its clinical trial designs to speed the development plan and ensure the production of quality data that can be submitted for marketing authorization. The marketing authorization application, in turn, should generally also be eligible for accelerated assessment. By 12 November 2018, the EMA had received 211 requests for PRIME eligibility; of these, it granted 46. More than one-­quarter of the medicines found eligible are intended for treatment of cancer. Second, the EMA has developed an ‘adaptive pathways’ approach to medicine development and data generation. This involves iterative development: an applicant may either (1) obtain (conditional) marketing authorization based on early data using

41  For instance, dinutuximab beta (Qarziba), a cancer medicine used to treat cancer of nerve cells (neuroblastoma) in patients over 1 year of age, has been authorized under ‘exceptional circumstances’ because obtaining complete information implied giving placebo to some patients during a trial, which would have been not ethical—the active substance of Qarziba, dinutuximab beta, is a recommended treatment for high-­risk neuroblastoma. 42  See chapter by Zettler and Lietzan, ‘Regulating Medicines in the United States,’ in this volume.

732   Aurélie Mahalatchimy ‘­surrogate endpoints’43 (or early time points, or a smaller population sample) and expanding later with additional clinical data to reduce uncertainty, or (2) obtain approval in stages, beginning with a narrow patient population and expanding later to a broader patient population. The first option corresponds to ‘accelerated approval’ in the United States as far as surrogate endpoints are concerned, in the sense that it allows patients access to important new medicines before phase 3 trials have confirmed clinical benefit.44 Both options involve iterative phases of evidence gathering and progressive licensing adaptations as uncertainties about the medicine’s safety and efficacy are progressively reduced and/or the targeted population is expanded. The adaptive pathways approach also allows an applicant to supplement clinical trial data with evidence gathered through real-­life use of the medicine. Although the adaptive pathways approach responds to patient needs for earlier access to innovative medicines, it necessarily involves a tradeoff between early access and certainty about a medicine’s actual risks and benefits. Between March 2014 and August 2016, the EMA ran an adaptive pathways pilot to explore the practical implications of the concept with medicines under development. Among the 62 applications received, 18 proposals were selected for face-­to-­face meetings. At the end of this pilot project, six of the applicants received parallel advice from EMA and Health Technology Assessment (HTA) bodies and one benefited from EMA scientific advice. The EMA has considered that this concept needs to be further explored, and this is being done in the context of parallel scientific advice with HTA bodies, with the inclusion of additional stakeholders such as patients and payer organizations. Indeed, patients and HTA bodies are also involved in the discussion of individual product development programs using the adaptive pathways. HTA bodies provide recommendations on new medicines that might be financed or subject to reimbursement by the relevant national or regional healthcare system; in countries where they operate, their assessments are used to determine reimbursement status, among other things. The involvement of HTA bodies in the new adaptive pathways approach is part of a new European trend of promoting collaboration between regulators and HTA bodies in order to narrow the gap between regulatory and HTA body expectations and to promote harmonization among the differing European HTA bodies. This should be in the interests of both patients and companies by eliminating duplication of work and allowing earlier decisions on pricing and reimbursement, which in turn may allow earlier access to medicines. However, this new approach has been criticized for its defects in evidentiary support. Drugs approved on early, and therefore limited, evidence may lead to false conclusions about efficacy and safety, thereby harming patients. In addition, realistic management strategies are still needed to obtain reliable real-­world data, and post-­ marketing confirmatory studies are often slow to be completed. As a result, medicines 43  Rather than reflecting an actual improvement in health such as longer survival, surrogate endpoints reflect measures that correlate with improvements in health, such as shrinkage in the size of a cancerous tumor. 44  See chapter by Zettler and Lietzan, ‘Regulating Medicines in the United States,’ in this volume.

Regulating Medicines in the European Union   733 with uncertain benefit-­risk ratios may be prescribed to patients for an extended period of time. Moreover, many HTA bodies and payers are afraid to pay for weakly tested yet expensive treatments, while manufacturers have to address uncertainty regarding assessment by HTA bodies and payers as well as possible sustainable price and managed-­ entry agreements.45 Although the latter are growing in number, they are neither successful nor available in every EU Member State or commonly used for products with a conditional marketing authorization or authorized under exceptional circumstances. Ultimately, patient access to medicines is not achieved until products have gone through pricing and reimbursement national systems, which are regulated differently in each different EU Member State.

2.2.3  Named Patients and Compassionate Use As in US law, EU law allows patients access to medicines outside of the marketing au­thor­i­za­tion pathway, by exception in certain situations. First, EU law contains what is known as a ‘named patient’ exception to the marketing authorization requirement. The primary EU medicines directive authorizes individual Member States to exempt medicines from the marketing authorization and other requirements in the directive in order to fulfil special needs and apart from any clinical trial. Exempted medicines must be supplied in response to a bona fide unsolicited order, formulated in accordance with the specifications of an authorized healthcare professional, and intended for use by an individual patient under the direct personal responsibility of the prescribing physician.46 Second, EU Member States may make certain medicines available for ‘compassionate use.’47 This is appropriate for patients with a life-­threatening, chronically, or seriously 45  Managed Entry Agreements are agreements between manufacturers and payers to share financial risk due to uncertainty with the provision of new innovative drugs in order to enable and foster access to medicines. 46  For instance, French law provides for named provisional authorizations (autorisation temporaires d’utilisation nominatives). See Article L5121-­12 2° of the French Public Health Code (named). Alpelisib, administered plus fluvestrant, to treat a genomic subgroup of breast cancer patients, or crizotinib for infants and children with lung cancer, have obtained named provisional authorizations in France. The United Kingdom permits the supply of unlicensed medicinal products (‘specials’) in order to fulfil the special needs of a single patient. Human Medicines Regulations 2012, Reg. 167; MHRA, The supply of unlicensed medicinal products ‘specials’ MHRA guidance note 14, 2014 (https://www.gov.uk/ government/uploads/system/uploads/attachment_data/file/373505/The_supply_of_unlicensed_ medicinal_products__specials_.pdf). 47  Regulation 726/2004. For instance, French law provides for group provisional authorizations (autorisation temporaires d’utilisation de cohorte). See Article L5121-­12 1° of the French Public Health Code (group). Larotrectinib for the treatment of glioma, which has obtained a marketing authorization in the United States and for which a marketing authorization application under the European centralized procedure is ongoing, has obtained a group provisional authorization in France. The UK also launched an ‘Early Access to Medicines Scheme’ (EAMS) in April 2014; this program gives patients with life-­threatening or seriously debilitating conditions access to medicines that are not yet authorized. See Implementation of Article 5(1) of directive 2001/83/EC; Office for Life Sciences, Early access to medicines scheme (EAMS): task group and principles, 2016 (https://www.gov.uk/government/ publications/early-­access-­to-­medicines-­scheme-­eams-­how-­the-­scheme-­works/early-­access-­to-­ medicines-­scheme-­eams-­task-­group-­and-­principles) and Office for Life Sciences, The Early Access to Medicines Scheme (EAMS): Operational Guidance, 2016.

734   Aurélie Mahalatchimy debilitating disease who cannot be treated satisfactorily by an authorized medicinal product. The medicine in question must be the subject of a pending marketing au­thor­i­ za­tion application or ongoing clinical trials. The Member State must notify the EMA when it permits compassionate use, and the CHMP may adopt opinions on the conditions for use and distribution and on the patients targeted. As long as this is an area of EU medicines law where there is not a single EU approach, the individual Member States take differing approaches to named patient access and compassionate use. Consequently, many but not all EU countries benefit from a compassionate use programme, and patients can access them at different times depending on where they live in the EU.

2.3  Other Routes to Market Specific medicines may also reach the market in Europe without approval of full marketing authorization applications containing data from safety and effectiveness testing. First, as in US law, European law allows the authorization of abbreviated applications for generic medicines. A generic application is exempt from the requirement to contain data showing safety and effectiveness. Instead it makes a comparative showing to an earlier authorized medicine and relies on the research performed to support that medicine’s market entry. A generic medicine must have the same qualitative and quantitative composition in active substances as the reference medicine, as well as the same pharmaceutical form, and it must be bioequivalent to the reference medicine. It can be prepared more quickly and cheaply than a full application. The resulting medicine is therefore less expensive than the original (reference) medicine, which provides savings for patients and healthcare providers and contributes to the sustainability of healthcare financing in Europe. Approval of a generic medicine means that the manufacturer of the original reference medicine will lose its exclusive position in the market, and, typically, its market share and price will drop dramatically. This means it must recover its research and development costs before generic market entry, and it is generally assumed that continued development of innovative medicines depends on those medicines having a sufficient period of exclusivity in the market. EU law aims to balance the need for a continuing supply of innovative medicines, on the one hand, and the need for prompt access to inexpensive generic medicines, on the other hand. As in US law, EU law therefore specifies a period of time before generic applications may be submitted and authorized. For medicines authorized through the central procedure and those approved at the Member State level, generic applications cannot be accepted until 8 years after approval of the reference product, and the generic drug cannot be marketed until 10 years after this approval. This 10-­year period of marketing exclusivity is extended to 11 years if, in the first 8 years, the marketing authorization holder of the reference medicine obtains approval of a new indication that brings a ‘significant clinical benefit in comparison with existing therapies.’ EU law permits a generic

Regulating Medicines in the European Union   735 company to conduct the studies and trials needed for a generic application during these periods, notwithstanding any patent protection on the reference medicine. Second, in some cases, approval of an abbreviated application is not possible: because the strict legal definition of ‘generic medicinal product’ is not satisfied, bioequivalence cannot be demonstrated through bioavailability studies or changes have been made to the active substance, therapeutic indication, strength, pharmaceutical form, or route of administration. Such medicine is so-­called hybrid medicine. For instance, vigabatrin (Kigabeq), for treating epilepsy in children between 1 month and 7 years of age, is similar to a reference medicine (Sabril 500 mg granules) containing the same active substance (vigabatrin) but Kigabeq is available in a different form and different strengths. In this case, the applicant must submit the results of appropriate preclinical tests or clinical trials. This application relies in part on the research that supported the reference medicine and in part on new research. It can be prepared more quickly and cheaply than a full application, and it can be expected to produce cost savings for patients and healthcare systems. The same 8-, 10-, and 11-­year exclusivity rules apply. Third, EU law permits the approval of ‘similar biological medicinal products,’ also known as ‘biosimilars.’ Like generic medicines, biosimilars are authorized on the basis of abbreviated marketing applications, but they are copies of biological medicines, and the marketing applications are therefore different. Biological medicines include recombinant proteins, monoclonal antibodies, medicinal products derived from human blood and human plasma, immunological medicinal products, and advanced therapy medicinal products. If a biological medicine cannot satisfy the conditions for generic approval due to differences in raw materials or manufacturing processes from its reference product, the applicant must submit results from preclinical tests or clinical trials. As in the United States, the applicant must illustrate similarity in terms of quality characteristics, biological activity, safety, and effectiveness based on a comprehensive comparability exercise. The EMA has issued detailed guidelines on the type and quantity of supplementary data needed, including numerous guidelines tailored to specific product classes. Use of this centralized procedure is mandatory for copies of biotechnology-­ derived medicines and certain other biological medicines, but (as is true of reference biologics as well) other biosimilars may be authorized at the national level. Fourth, the EU medicines regulatory framework does not apply to certain medicines prepared by pharmacies for patients. There are two possibilities. A pharmacy might prepare a medicine for an individual patient in accordance with a medical prescription for that patient, for instance because the patient is allergic to an ingredient in the commercial preparation, or more often because the dosage is not adapted to the specific need of the patient (e.g., adult or child). This medicine is known as a magistral formula and corresponds to a compounded medicine in the United States. Individual Member State laws governing compounding vary as there is no single EU-­wide regulatory approach here. For example, some Member States permit third parties to compound medicines for pharmacies and others do not. In the alternative, a pharmacy may prepare medicines in accordance with a pharmacopeia and supply those medicines directly to its patients. This medicine is known as an officinal formula, which is quantitatively very used, and

736   Aurélie Mahalatchimy has no equivalent in the United States. The specific entity of a formally recognized pharmacist is a key element of the European regulatory landscape. Hence, we have these kinds of approaches regarding the officinal formula but also ‘behind the counter medicines’ (see later discussion), where the pharmacist has a central role.

3  Risk Management All medicines have both desirable beneficial effects and undesirable adverse effects. Indeed, the dual nature of medicines is embodied in the etymology of the words ‘medicine’ and ‘pharmaceutical.’ They derive from the Latin medicamentum and the Greek pharmakon, both of which mean both ‘remedy’ and ‘poison.’ Because medicines have this dual nature, regulators must consider both when assessing marketing applications, and authorization is granted only if a medicine’s benefits outweigh its risks. Risk in this context can mean any risk relating to the quality, safety, or efficacy of the medicine or relating to patient health or public health, as well as to any risk of undesirable effects on the environment. EU law also requires that risk be appropriately managed all along the therapeutic chain. Nevertheless, this occurs particularly at the time of marketing au­thor­i­za­tion and after authorization.

3.1  At the Time of Marketing Authorization The regulatory authorities—whether EMA and the European Commission or national competent authorities—assess a medicine’s risks at the time of marketing authorization, permitting market entry only if the good quality medicine has a positive benefit-­risk balance. The authorization decision thus involves both risk assessment and risk management. More precisely, for medicines subject to the centralized procedure, the EMA assesses the risk when it reviews the application, and the European Commission manages the risk when it makes an authorization decision on the basis of the EMA’s opinion. These decisions consider both the product and its associated information, meaning the Summary of Product Characteristics (SmPC), which is technical labelling for healthcare professionals, and the package leaflet, which is intended for patients. Such associated information is fully part of the marketing authorization dossier. Marketing authorization applications must also include risk management plans (RMPs), and the EMA’s assessment of the application includes its assessment of the planned risk management. An RMP has three elements. First, it must identify or characterize the medicine’s safety profile, focusing on important risks, missing information, and safety concerns that need to be managed proactively or studied further. Second, it must plan ‘pharmacovigilance’ activities to characterize and quantify clinically relevant risks and identify new adverse reactions. As discussed in the next section, this includes routine pharmacovigilance (a passive surveillance system) but may include proactive

Regulating Medicines in the European Union   737 activities such as nonclinical studies or even clinical trials. Third, it must describe the planning and implementation of risk management (or ‘minimization’) measures. These always include the SmPC and package leaflet, the content of which is controlled and which use a standardized format; these are the primary means by which a marketing authorization informs healthcare professionals and patients about the risks associated with its medicine. Risk management measures may also include controlled access programs, such as a requirement that prescribers and dispensers document their understanding of the risks of the product or that limits the dispensing to certain registered pharmacies. When granting a marketing authorization, the competent authority must determine whether the medicine will be available only by prescription. A medicine must be classified as prescription-­only if (1) it is likely to present a danger if used without medical supervision, (2) it is frequently and often used incorrectly and as a result is likely to endanger health, (3) it contains substances that need further safety assessment, or (4) it is administered parenterally. Competent authorities may subdivide the prescription-­only category into three further subdivisions: medicines on renewable or nonrenewable prescription, medicines subject to special medical prescription (typically for narcotics and other drugs that can be abused), and medicines on restricted medical prescription, reserved for use in certain specialized areas (e.g., treatments that can only be used in a hospital). Some Member States also create subcategories of nonprescription medicines, so that some medicines are available in general retail settings, others only in pharmacies, and others only after consultation with a pharmacist. European law relating to the legal status of medicines differs profoundly from US law, and the differences affect patient experiences in the healthcare system. US law generally provides for two general categories of medicine: prescription medicines (including controlled substances and other prescriptions drugs) and nonprescription medicines, and the latter can be purchased in any retail setting.48 In the EU, Good Distribution Practices (GDPs) apply to medicines and they are taken into account at the time of marketing authorization (as well as GMPs that are a prerequisite to obtaine marketing authorization). They describe the minimum standards to be met from the site of manufacture to the pharmacy or person authorized or entitled to supply medicines to the public to ensure the quality is maintained throughout all stages of the supply chain. Pharmacies and pharmacists in particular, whose special status has also been recognized by the European Court of Justice (ECJ), play a fundamentally different role in Europe than they do in the United States. In Europe, they may be the primary healthcare professionals with whom a patient interacts. They diagnose conditions, recommend treatments, and counsel their patients. If a medicine is placed behind the counter in Europe, a patient must have a sustained conversation with a pharmacist about his or her condition before purchasing this medicine. And if it is available only in pharmacies, this conversation may still occur. In addition, Member States can have higher 48  See the chapter by Zettler and Lietzan ‘Regulating Medicines in the United States,’ in this volume.

738   Aurélie Mahalatchimy requirements regarding distribution. For instance, in France, the entire chain of development of medicines is under the so-­called pharmaceutical monopoly: medicines are distributed in authorized pharmaceutical establishments under the responsibility of pharmacists only. Consequently, the same medicine may well be available at a grocery store in one Member State, as it may be in the United States, while it may be available only in a pharmacy in another EU Member State. These types of rules have been challenged before domestic courts and the ECJ because they could have the effect of impeding trade, but, in general, the ECJ recognizes the public interest inherent in maintaining the special position of pharmacists and pharmacies in European societies. Notwithstanding formal regulation, even in Member States that do not have a third behind-­the-­counter status, pharmacists play a role providing medical advice (and thus managing the risks of medicines) that they do not play in the United States.

3.2  After Marketing Authorization Various rules applicable after marketing authorization also ensure that the risks associated with medicinal products are continually assessed and minimized as much as possible. Broadly speaking, these fall in two categories: risks inherent to the medicine itself and risks associated with the manufacture and distribution of a medicine.

3.2.1  Risks Inherent to the Medicine Additional information about the overall clinical profile of a medicine continues to emerge after marketing authorization. EU authorities have repeatedly strengthened EU laws governing the safety monitoring of medicines, or pharmacovigilance, including through significant legislative changes in 2010 and 2012. Every participant in the healthcare system shares responsibility for pharmacovigilance. Traditionally, the marketing authorization holder (the company or other legal entity having the authorization to market a medicine), national competent authorities, the EMA (especially its Pharmacovigilance Risk Assessment Committee), and the European Commission play a role. Healthcare professionals and patients now also play a key role. The marketing authorization holder must operate a pharmacovigilance system. Among other things, the marketing authorization holder must report adverse reactions to the EMA; this includes both expedited (15-­day) reports of serious adverse reactions and Periodic Safety Update Reports (every 6 months for the first 2 years after market entry, then once a year for 2 years, and then every 3 years thereafter). The EMA in turn operates EudraVigilance, an electronic system for managing and analyzing suspected adverse reactions to medicines that are authorized or in clinical trials in the EEA. This system allows electronic exchange of reports among the EMA, national competent authorities, marketing authorization holders, and clinical trial

Regulating Medicines in the European Union   739 sponsors. It also includes a large pharmacovigilance database, with search and tracking functions, containing information on suspected adverse reactions arising from uses of medicines within or outside the terms of the marketing authorization as well as those occurring during post-­authorization studies. The data in this database are publicly available in an aggregated format only. The Member States also operate a pharmacovigilance system and are required to evaluate the information, consider options for risk minimization and prevention, and take appropriate regulatory action concerning marketing authorization. For instance, in the case of a centrally authorized medicine, a Member State could recommend that au­thor­i­ za­tion should be modified, suspended, or revoked. In this case, it would inform the EMA, other Member States, and the marketing authorization holder. The European Commission would make a final decision, after consulting the EMA. A Member State may also, on its own initiative or at the Commission’s request, suspend the use of medicines in its territory where urgent action is essential to protect human health or the environment. Individual Member States must also encourage doctors, pharmacists, nurses, other healthcare professionals, and patients to report suspected adverse reactions to the competent authorities. Certain medicines subject to additional monitoring by regulatory authorities now have black inverted triangles in their package leaflets and SmPCs, such as risankizumab (Skyrizi)49 or fluciclovine (Axumin).50 This includes any medicine containing a new active substance first authorized after January 1, 2011; any biological medicine au­thor­ ized after that date; medicines with conditional approval; and medicines authorized under exceptional circumstances. The inverted triangle allows patients and healthcare professionals to identify medicines that require additional monitoring and encourages them to report unexpected adverse reactions through national reporting systems. In 2018, 351 medicines were subject to additional monitoring (4% more compared to 2017). EMA publishes a list of medicines under additional monitoring that have been au­thor­ ized centrally or nationally.

3.2.2  Risks Associated with Manufacture and Distribution Compliance with current GMPs and GDPs is also ensured after the marketing au­thor­i­ za­tion. Apart from licensing the manufacturer and wholesale distributor of a medicine, the competent authority of each applicable Member State must ensure that the legal requirements governing medicinal products are complied with after marketing au­thor­i­za­tion. Regarding manufacturing surveillance, this includes repeated and unannounced controls and inspections of manufacturing facilities to ensure compliance with quality requirements through the implementation of GMPs. Regarding distribution 49  A medicine intended for the treatment of moderate to severe psoriasis, authorized on 26 April 2019. 50  A diagnostic medicine used with a body scan to check whether or not prostate cancer has returned, authorized on 21 May 2017.

740   Aurélie Mahalatchimy surveillance, the marketing authorization holder is expected to audit the compliance of any active substance manufacturers and distributors on which it relies. However, as previously said, individual Member States may impose additional requirements on the distribution of medicines. The marketing authorization holder also has an ongoing responsibility of ‘supervision’ with respect to its approved medicinal products. For instance, it must continually ensure that its medicines are manufactured and controlled according to generally accepted scientific methods, which entails introducing changes as appropriate (such as substantial changes to the manufacturing process of the active substance, which may have a significant impact on the quality, safety, or efficacy of the medicinal product, or changes in the testing procedures for active substances or the starting materials, reagents, or intermediates used in the manufacturing process of the active substance) to take account of scientific and technical progress. This may entail securing approval of a ‘variation’ to the marketing authorization, which corresponds to a ‘supplement’ in the United States. In addition, the marketing authorization holder has an ongoing responsibility to provide the EMA, the Commission, and the Member States with any new information that might require amending the particulars or documents relevant to its permission to market its medicine. This would include, for instance, new information relevant to the benefits and risks of the product, including results for clinical trials and data on the use of the medicine (including data from uses that are outside the scope of the approval). Finally, as in the United States,51 EU authorities have taken steps to address the presence of ‘falsified medicines,’ such as for Fentanyl (a synthetic opioid originally used to treat severe pain, especially for patients with cancer) and its derivatives, in the supply chain.52 A falsified medicine is any medicine accompanied by a false representation of identity, source, or history. Various measures are intended to reduce the risk of falsified medicines in the EU, including GMP requirements, GDP requirements for active substances, inspection principles, safety features for the packaging of medicinal products, and information campaigns to raise awareness of the dangers of falsified medicines. Perhaps most significantly, beginning in February 2019, most medicines in the EU will have a unique identifier (in the form of a unique sequence carried by a two-­dimensional barcode) on their packaging to allow their identification and authentication at any time on the market, as well as a device to allow verification that packaging has not been tampered with. These safety features are intended to minimize the risk of falsified medicines circulating undetected for lengthy periods of time.53

51  See the chapter by Zettler and Lietzan, ‘Regulating Medicines in the United States,’ in this volume. 52  Directive 2011/62/EU on falsified medicines, OJ L174/74-­87. 53  Commission Delegated Regulation (EU) 2016/161 laying down detailed rules for the safety features appearing on the packaging of medicines for human use, OJ L 32/1–27.

Regulating Medicines in the European Union   741

4  Innovation and Competition European law relating to innovation and competition in medicines takes into account both economic policy and public health policy, as described in the following subsections.

4.1  Economic Policy While medicinal products raise special risks, they are nevertheless considered as goods. As such, they are covered by the provisions of the EU’s Treaty on the Functioning of the EU relating to the free movement of goods within the EU’s internal market (similar to the US Commerce Clause). That is why the ECJ always controls whether any legal restraints on trade constitute quantitative import or export restrictions or other measures with equivalent effect. These kinds of rules are prohibited on trade between Member States. There is, however, an exception for prohibitions and restrictions justified on public health grounds, provided those measures are proportionate to their goals.54 Moreover, five areas of EU economic policy are relevant to medicines innovation and competition in the medicines market. First, EU intellectual property policy plays a key role in innovation and competition in medicines. In addition to brands (trademarking), which is not developed here due to limited space, EU law provides one main incentive for the development of new medicines: patents. As in the United States, branding of a medicinal product can be renewed indefinitely, but patents are limited in duration to 20 years. Patents are regulated at multiple levels in Europe. Each Member State has a national patent office that issues patents under its own patent law, which in turn is harmonized with the patent law of other Member States through an EU Directive.55 Disputes arising out of national patents are handled by national Courts and by the ECJ. In addition, and separate from the EU structures, since the 1970s, it has been possible to apply for a European patent from the European Patent Office (EPO). The EPO was created as part of the European Patent Convention (EPC), which involves some—not all—EU Member States, as well as ten additional countries. An applicant applies to the EPO for a European patent. The resulting patent confers protection in all the contracting States designated by the applicant as long as it has been validated through complex and costly procedures by their national patent offices (e.g., implementation of national requirements such as payment of designation fee and translation of the patent into the national language). 54  Consolidated Version of the Treaty on the Functioning of the European Union, art. 34–36, 2012, OJ C 326/47-­390. 55  Directive 98/44/EC on the legal protection of biotechnological inventions OJ 1998 L 213/13-­21.

742   Aurélie Mahalatchimy In 2012, the EU Member States (except Spain, Poland, and Croatia) agreed on a ‘patent package,’ a legislative initiative based on enhanced cooperation and intended to lay the ground for unitary patent protection in Europe. This package includes a regulation creating a European patent with unitary effect, a regulation on the language regime applicable to the unitary patent, and an agreement to create a single Unified Patent Court. The patent package will enter into force once the Unified Patent Court agreement has been ratified by thirteen Member States including France and Germany. Such unitary patent will make possible to obtain patent protection in up to 26 Member States by submitting a single request to the EPO, thus removing the need for national validation procedures existing for European patents. The EU patent eligibility requirements are generally harmonized with those of the United States, with one significant exception. Any new, useful, and nonobvious invention can be patented provided it is concrete and the claims are adequately enabled in a clear, written description.56 But there are several general exceptions, including one relevant to medicines: ‘methods for treatment of the human or animal body by surgery or therapy and diagnostic methods practiced on the human or animal body.’57 While a substance or composition may be patented based on the use in question,58 the method of treatment itself is not patentable. In the United States, methods of treatment can be patented. Since 1992, EU law has also provided for a Supplementary Protection Certificate (SPC), which extends the term of patents associated with certain medicines in order to compensate in part for the loss of patent life during premarket research and development.59 One SPC is available per product, and it is available only if (1) the patent has not expired, (2) the product is authorized, and (3) the authorization is the first authorization to place the product (meaning, active ingredient) on the market as a medicine. Although the language in the Regulation is simple, application of the ‘one SPC per product’ and ‘first authorization of the active ingredient’ rules has proved complex, for instance 56  Article 52(1) of the European Patent Convention. The following are not considered inventions within the scope of Article 52(1): (a) discoveries, scientific theories. and mathematical methods; (b) aesthetic creations; (c) schemes, rules, and methods for performing mental acts, playing games, or doing business, and programs for computers; (d) presentations of information. Id. Art. 52(2). In addition to these patentability requirements, the full scope of a claim must be adequately enabled by disclosing methods of practicing the invention in the specification. This requires both ‘enablement’ and a ‘clear written description,’ although the difference between the two is not always distinct. 57  Article 53(c) of the European Patent Convention. Moreover, inventions the commercial exploitation of which would be contrary to ‘ordre public’ or morality are also excluded. Id. Art. 53(a). European patent law provides a nonexhaustive list of these inventions: processes for cloning human beings; processes for modifying the germ line genetic identity of human beings; and uses of human embryos for industrial or commercial purposes, processes for modifying the genetic identity of animals which are likely to cause them suffering without any substantial medical benefit to man or animal, and also animals resulting from such processes. Rule 28 of the implementing regulations to the European Patent Convention and Article 6(2) of Directive 98/44/EC. 58  European Patent Convention and Recital (35) of Directive 98/44/EC, Article 53(c). 59  Council Regulation (EEC) 1768/92 (OJ 1992 L 182/1–5) and Regulation (EC) 469/2009 on the supplementary protection certificate for medicinal products, OJ 2009 L 152/1-­10.

Regulating Medicines in the European Union   743 with respect to ingredients that may or may not be active ingredients and with respect to  combination products that include both new and previously approved active ingredients.60 Furthermore, Regulation (EU) 2019/933 introduced an exception, known as the ‘manufacturing waiver.’ It allows companies to manufacture generics and biosimilars during the term of the SPC for the purpose of export to non-­EU markets where patent or SPC protection has expired or never existed, but also for the making (no earlier than 6 months before the expiry of the certificate) of a product and storing it in the Member State of making for the purpose of entering the market of any Member State upon expiry of the corresponding certificate. As ‘[t]he timely entry of generics and biosimilars into the Union market is important, particularly in order to increase competition, to reduce prices and to ensure that national healthcare systems are sustainable and that patients in the Union have better access to affordable medicines,’ Regulation (EU) 2019/933 aims to correct the significant competitive disadvantage for makers of generics and biosimilars established in the EU in comparison with makers based in third countries that offer less or no SPC protection and who could enter the Union market immediately after expiry of the certificate. This Regulation also aims to ‘strike a balance between restoring a level playing field between those makers and ensuring that the essence of the exclusive rights of holders of certificates . . . is guaranteed in relation to the Union market.’ As such, it is likely to have a positive impact that leads to more manufacturing activity in the EU. Most importantly, it will allow generics to reach the market in non-­EU countries, and in EU Member States after SPC expiry, as a relevant step for quicker patient access to medicines with a more affordable price. Second, EU competition policy focuses on four topics relevant to the regulation of medicines: concerted practices, abuse of a dominant position, generic medicines, and parallel trade. Only parallel trade will be briefly mentioned here. Parallel trade—also called ‘parallel import’—refers to the transfer of a supply of medicine from one EU country to another EU country outside of the distribution network established by the medicine’s manufacturer. For example, a wholesaler buys medicinal products in Spain and sells them in Germany. Parallel trade can be explained by the variation in medicine prices throughout the EU since individual Member States reach different negotiated prices for their national health systems with the pharmaceutical industry. Under the EU principle of free movement of goods within the EU, parallel trade gives patients greater access to lower prices for their medicines. Parallel trade is permitted where the concerned product is the same or very similar61 to a product already authorized for sale in 60  Case C-­322/10 Medeva BV v Comptroller General of Patents, Designs and Trade Marks [2011] ECR I-­12051; Case C-­493/12 Eli Lilly and Company [2013] EU:C:2013:835; Case C-­443/2012 Actavis et al. v Sanofi et al [2013] ECLI:EU:C:2013:833; Case C-­577/13 Actavis Group PTC and Actavis UK v Boehringer Ingelheim Pharma GmbH & Co., KG [2015] EU:C:2015:165. 61  The medicines have a common origin (i.e., they are both manufactured pursuant to agreements concluded with the same licensor) and the same composition (without having to be identical in all respects, they are manufactured according to the same formulation, using the same active ingredient, and also have the same therapeutic effects). Case C-­201/94 Smith & Nephew Pharmaceuticals and Primecrown v The Medicine Control Agency [1996] ECR I-­5819.

744   Aurélie Mahalatchimy the destination Member State. The European Commission’s position is that parallel imports increase price competition and benefit consumers, but the practice has critics. Medicines tend to flow from countries with lower prices to countries with higher prices, and some believe it contributes to shortages in the lower income countries. Finally, manufacturers object to parallel trade on the grounds that price discrimination is more economically efficient than unitary pricing. Third, as a matter of industrial policy, the EU aims to boost the competitiveness of the European pharmaceutical industry, ‘one of the cornerstones of a knowledge-­based economy’ and ‘important for European public health, economic growth, trade, and science.’62 The European Commission has therefore promoted activities designed to create incentives and rewards for a sustainable and competitive pharmaceutical sector while prioritizing and improving patient access to medicines, and it has identified numerous areas for future development. It supports cooperation between Member States in the implementation of Health Technology Assessments, for instance, while also strengthening industry competitiveness with special legislation to encourage the development of paediatric and orphan medicinal products. It aims to foster transparency and ethical behavior in the pharmaceutical sector, to improve access to medicines worldwide, and to reinforce the presence of the European pharmaceutical industry in the global market. As part of this initiative, the Commission has identified public–private cooperation as critical to the long-­term sustainability of the pharmaceutical sector. It established a Joint Technology Initiative (JTI) for innovative medicine. A JTI combines private-­sector investment with European public funding and fosters transnational cooperation through the creation and management of a common Strategic Research Agenda. The current Innovative Medicines Initiative 2 (IMI 2) has an operating budget of 3.3 billion euros for the 10-­year period from 2014 to 2024,63 and it is viewed as implementing the world’s largest public–private partnership in the field of life sciences. The EU also aims to support small to medium-­sized enterprises (SMEs), ‘the backbone of Europe’s economy’ as the vast majority of business in the EU (99%) are SMEs, and these companies employ two-­thirds of the private sector. SMEs play an important role in the development of new medicines, accounting for 10–15% of all applications, more than half of which are for new active substances. The EMA provides administrative, regulatory, and financial support to SMEs through an SME office with dedicated personnel. Its current action plan includes expanded outreach to newly created companies (e.g., to increase awareness of SME support services), strengthened training of SMEs, and promoting the use of regulatory development support tools (such as the PRIME scheme). Fourth, the European Commission has numerous policy initiatives and practices designed to support European research and innovation. The current Framework 62  Commission Staff Working Document, Pharmaceutical Industry: A Strategic Sector for the European Economy (SWD(2014)216/F1), 26 June 2014 (http://ec.europa.eu/DocsRoom/ documents/7649). 63  IMI, ‘Introducing IMI, The Budget’ (http://www.imi.europa.eu/content/mission).

Regulating Medicines in the European Union   745 Programme for Research and Innovation (2014–2020) (also known as Horizon 2020) notes that biotechnology is poised to deliver a ‘stream of new applications’ and thus identifies strategic investment in biotechnology research and development as a priority. Some of this work overlaps with the policy initiatives intended to boost the competitiveness of the pharmaceutical industry. For example, the Commission identifies support of SMEs and the Innovative Medicines Initiative as among its policies relating to health research and innovation. The Horizon 2020 work programme for health, demographic change, and well-­being solicits proposals in numerous areas of medicine, such as proposals relating to use of the human microbiome in treatment of disease, proposals focusing on the development of regenerative medicine, and proposals exploiting the full potential of in-­silico medicine research for personalized diagnostics and therapeutics. Finally, the EU’s external policy includes various undertakings that directly affect the pharmaceutical sector. These include, for instance, the Agreement on Trade-­Related Aspects of Intellectual Property Rights (TRIPS) and the Council of Europe’s Convention on the Elaboration of a European Pharmacopeia. The goal of the European Pharmacopeia was to harmonize specifications for medicines to enable free circulation of high-­quality medicines throughout Europe.

4.2  Public Health Policy EU public health policy also affects innovation and competition in medicines. Broadly speaking EU public health policy aims to protect and improve the health of EU citizens, support the modernization of health infrastructure, and improve the efficiency of Europe’s health systems. The EU Health Programme outlines the EU’s strategy for ensuring good health and healthcare. The current programme, applicable for the years 2014 to 2020, has four specific objectives: health promotion, protection from serious health cross-­border health threats, capacity building, and access to improved healthcare. The special regulatory frameworks for three types of medicine—orphan medicines, paediatric medicines (discussed below), and advanced therapy medicinal products (discussed in section 5)—fall within the broad scope of this programme. As in US law, EU law offers incentives for the development of orphan medicines— drugs that are intended to treat rare diseases or which are not otherwise expected to generate sufficient return in the marketplace to justify their development.64 To receive orphan designation the company also must show that there is no satisfactory alternative medicine authorized for the market or that the proposed orphan medicine will be of 64  Regulation (EC) 141/2000, op. cit. A medicine is eligible for orphan designation if it is (1) intended for the diagnosis, prevention, or treatment of a life-­threatening or chronically debilitating condition affecting not more than 5 in 10,000 persons in the Community; or (2) it is intended for the diagnosis, prevention, or treatment of a life-­threatening, seriously debilitating, or serious and chronic condition in the Community, and without incentives it is unlikely that the marketing of the medicinal product in the Community would generate sufficient return to justify the necessary investment. For instance, Mepsevii, a medicine to treat the Sly syndrome that has been authorized under ‘exceptional circumstances’ 23 August 2018 as mentioned earlier, was designated an orphan medicine.

746   Aurélie Mahalatchimy significant benefit. Orphan medicines must be authorized through the centralized marketing authorization procedure. Companies developing orphan drugs benefit from protocol assistance and may be exempted from marketing authorization fees. After approval, an orphan medicine enjoys 10 (rather than 8) years of market exclusivity, during which neither the Commission nor an EU Member State may approve a ‘similar’ medicinal product for the ‘same’ therapeutic indication. This period may be reduced to 6 years if the designation criteria are no longer met. Between 2000 and 2015, the Commission granted 1,544 orphan designations and granted marketing authorization to 117 orphan medicines.65 Orphan medicines can be very expensive and present a challenge to the sustainability of the national health systems, as noted regularly by Eurordis (the voice for rare disease patients in Europe), and, for this reason, orphan exclusivity remains very controversial in Europe, just as in the United States, despite its apparent success in incentivizing development of these medicines. The regulation on paediatric medicine was adopted in 2006, to improve the health of  children under 18 years of age in the European Union through the promotion of research, development, and marketing authorization of medicinal products for paediatric use.66 As in US law, EU law aims to increase paediatric research through a combination of obligations and incentives. Today every new medicine in Europe must also be studied in children, and this research is grounded in a paediatric investigation plan (PIP) that must be developed early in the preauthorization clinical development program. The PIP, in turn, is ‘a research and development programme aimed at ensuring that the necessary data are generated determining the conditions in which a medicinal product may be authorized to treat the paediatric population.’ Paediatric medicines benefit from strengthened incentives. The Paediatric Regulation also created a new type of marketing authorization, the Paediatric Use Marketing Authorization (PUMA), for a medicine to be developed exclusively for use by children (following a PIP approved by the PDCO). For instance, Slenyto has obtained a PUMA on 20 September 2018 for the treatment of insomnia in children and adolescents with autism spectrum disorder or Smith-­Magenis syndrome. If there is a corresponding product authorized for adults that is no longer covered by intellectual property rights, the manufacturer may pursue a Paediatric Use Marketing Authorization, using the same brand name in order to capitalize on existing brand recognition while benefiting from the ‘8 plus 2’ years of exclusivity normally associated with new marketing au­thor­i­za­tions. If there is a corresponding product authorized for adults covered by intellectual property rights, the holder of the patent or supplementary protection certificate is entitled to a 6-­month extension of the SPC’s duration (except in the case of an orphan drug). If the paediatric medicine is also an orphan drug, the 10-­year period of orphan market exclusivity is extended to 12 years. For 65  European Commission, Inventory of Union and Member State incentives to support research into, and the development and availability of, orphan medicinal products, State of play 2015, SWD (2015) 13 final. 66  Regulation (EC) 1901/2006, op. cit., and Regulation (EC) 1902/2006, op. cit.

Regulating Medicines in the European Union   747 instance, glibenclamide (Amglidia) to treat newborns and children with neonatal ­diabetes is both an orphan and a paediatric medicine.

4.3  Advertising and Promotion Unlike US law, European law uses the term ‘advertising’ to refer broadly to any kind of promotion of a medicinal product—thus not only advertising to the general public and advertising to prescribers, but also visits by medical sales representatives to prescribers, supplying of drug samples, sponsorships of promotional meetings attended by prescribers, and sponsorship of scientific congresses attended by prescribers.67 As in US law, EU law prohibits advertising of any medicine that has not yet been authorized for the market. Also unlike US law, EU law prohibits direct-­to-­consumer advertising of prescription medicines. This prohibition raises particular difficulties where advertising is through the internet so consumers can also access it. Indeed, the dissemination on a website of information relating to medicines available on medical prescription only, where that information goes beyond ‘the literal and complete reproduction of the package leaflet or the summary of the product’s characteristics, which have been approved by the [competent] authorities’ is prohibited, as is information ‘selected or rewritten by the manufacturer, which can be explained only by an advertising purpose.’68 Direct-­to-­consumer advertising of nonprescription medicines that contain narcotics or psychotropic substances is also prohibited. There is an exception for vaccination campaigns approved by the competent authorities of the Member States. EU law similarly prohibits industry from distributing medicinal products directly to the public for promotional purposes. It further permits Member States to ban the advertising of any medicinal product subject to reimbursement, which several have done, including France.69 It does, however, expressly authorize advertising of medicines ‘which, by virtue of their composition and purpose, are intended and designed for use without the intervention of a medical practitioner for diagnostic purposes or for the prescription or monitoring of treatment, with the advice of the pharmacist, if necessary.’70 For instance, this is the case in France for cough or cold medicines that are not submitted to prescription and not reimbursed by national health insurance, such as Actifed or NurofenFlash. This advertising is subject to various requirements imposed under EU law. For instance, it must be clear that the message is an advertisement and that the product in question is a medicinal product, and the advertising must include the information necessary for correct use of the product. Some additional principles are similar to those applied by the US FDA, including the rule that advertising must not 67  Directive 2001/83/EC, Arts. 86 et seq. 68  Case C-­316/09 MSD Sharp & Dohme GmbH v Merckle GmbH [2011] ECLI:EU:C:2011:275. 69  Article L5122-­6 of the French Public Health Code. 70  Directive 2001/83/EC, Art. 88.2.

748   Aurélie Mahalatchimy suggest that the effects of a medicine are guaranteed or that they are unaccompanied by adverse reactions, but others are unique to Europe, such as the rule that advertising may not suggest that the health of a person can be enhanced by taking the medicine and may not give the impression that a medical consultation is unnecessary. In contrast to direct-­to-­consumer advertising, advertising to persons qualified to prescribe or supply medicines is generally permitted under EU law. Two basic rules apply and are broadly speaking analogous to rules that apply in the United States. First, all parts of the advertising of a medicinal product must comply with the particulars listed in the product’s summary of product characteristics. Second, the advertising must present the medicine objectively, without exaggerating its properties and without being misleading. In addition, advertising to prescribers and suppliers must include essential information compatible with the summary of product characteristics and the supply classification (nonprescription vs. prescription) of the medicinal product. Member States may add supplementary requirements such as that the advertising must state the selling price. Finally, a Member State may instead decide that the advertising of a medicinal product to persons qualified to prescribe or supply the product may include only the name of the medicinal product or its international nonproprietary name, where this exists, or the trademark, if it is intended solely as a reminder.

5  A Special Framework for Advanced Therapy Medicinal Products In 2007, the European Union established a specific legal framework for medicines based on genes, cells, or tissues of human or animal origin, also known as advanced therapy medicinal products (ATMPs).71 ATMPs include four categories of product: gene therapy medicinal products, cell therapy medicinal products, tissue engineered products, and the combination of any of these with medical devices as combined ATMPs. Many of  these products, like stem cell therapies, would be considered ‘regenerative medicines’ in the United States. Cell and tissue products that are minimally manipulated or intended for homologous use (i.e., for the same purpose in the recipient as in the donor) are not considered ATMPs.72 A classical blood transfusion would not constitute an ATMP. ATMPs tend, in fact, to be developed for the purpose of preventing, treating, or even curing rare diseases, acquired diseases such as cancers, central nervous system degenerative conditions (such as Alzheimer’s and Parkinson’s disease), and genetic diseases. For instance, Strimvelis is a gene therapy ATMP authorized from 26 May 2016 to treat severe combined immunodeficiency due to an enzyme (adenosine deaminase) deficiency (ADA-­SCID), a rare inherited condition. Such enzyme deficiency leads to an 71  Regulation (EC) 1394/2007 on advanced therapy medicinal products, OJ L324/121-­137. 72  They are covered instead by the Directive 2004/23/EC on tissues and cells (OJ L102/48-­58) or the Directive 2002/98/EC on human blood and blood components (OJ L33/30-­40).

Regulating Medicines in the European Union   749 immune system deficiency where patients rarely survive more than 2 years without effective treatment. Broadly speaking, the ATMP framework is intended to foster the competitiveness of European pharmaceutical companies in the field, facilitate the flow of goods throughout the European market, and protect public health. In addition to the primary Directive (2001/83/EC) that applies to all medicinal products in the European Union, these medicines are subject to a Regulation specifically relating to ATMPs (as well as many guidelines). Together these documents create a stricter legal regime for ATMPs than apply to other medicines in the EU and, indeed, than apply to tissue-­based products in the United States. ATMPs must be authorized under the centralized procedure discussed in section 2.1. Within the EMA, the Committee for Advanced Therapies (CAT) has responsibility for every issue relating to ATMPs, including recommending their classification and assessing the resulting marketing authorization applications. Small to medium-­sized enterprises (SMEs) frequently perform the earliest work relating to new ATMPs, such as proof of concept and feasibility studies. These companies do not necessarily have the means to carry out the clinical trials needed for authorization. The Regulation on ATMPs therefore creates a procedure for the EMA to review and certify the quality and nonclinical information generated by an SME, which in theory will facilitate a partnership between the SME and a larger company with the resources to carry out the expensive and risky clinical trials needed for centralized authorization.73 This procedure has been used very little. There is some discussion of extending it to academics who also develop feasibility studies. ATMPs are subject to additional regulatory requirements— beyond those applicable to other medicines—relating to clinical trials, manufacturing, pharmacovigilance, and traceability. In a 2014 report on the ATMP regulation, the European Commission highlighted that few—only four—ATMPs have received marketing authorization. It also suggested clarification of the legislation in order to help the translation of research into medicines available to patients in the European Union. Nevertheless, it also reported an increase in the activity of the CAT, revealing the dynamics of this sector. Indeed, since adoption of the Regulation, the Commission has finally granted 17 marketing authorizations for ATMPs.74 However, the four ATMPs that were first authorized were later withdrawn by their marketing authorization holders for commercial reasons.75 Such ATMP market failures may be because they are very costly and they may not be reimbursed by national health insurance systems. In addition to the high cost of ATMPs and general lack of health resources, the reimbursement challenges may reflect the limited evidence of long-­term cost and effectiveness to date and the lack of 73  The Regulation also allows SMEs a 90% reduction in the fees (and other applicants a 65% reduction in the fee) that would otherwise be payable to the EMA for scientific advice. 74  The 15 authorized products are: ChondroCelect, Glybera, MACI, Provenge, Holoclar, Imlygic, Strimvelis, Zalmoxis, Spherox, Alofisel, Yescarta, Kymriah, Luxturna, Zynteglo, Zolgensma, Libmeldy, and Tecartus. 75  Marketing authorizations for Provenge, Chondrocelect, MACI, and Glybera were withdrawn at the request of the marketing authorization holders.

750   Aurélie Mahalatchimy flexibility in reimbursement regulatory pathways. Moreover, there are differing views of the value of ATMPs among stakeholders. There may be ways to address some of these challenges. For instance, the iterative adaptive pathways approach to market (discussed in section 2.2.2), already used for some ATMPs, was established in part to fill the gap between entering the market and being reimbursed by national health insurances, given that these two aspects are discussed both earlier and together with the various stakeholders involved. In addition, the European Commission has adopted a proposal for a Regulation on health technology assessment especially to improve the availability of innovative health technologies for EU patients, to ensure efficient use of resources and strengthen the quality of HTA across the European Union, and to improve business predictability. Various regulatory flexibilities described in this chapter have been used—or could be used—for ATMPs. For instance, the PRIME scheme aims to support medicines development, including ATMPs, and it has been used for Yescarta, Kymriah, Zynteglo, and Zolgensma. Yescarta and Kymriah for the treatment of blood cancer as well as Zynteglo for the treatment of beta thalassaemia, were reviewed under the accelerated assessment scheme, discussed in section 2.2.1, which allows for faster review at the EMA. The development of ATMPs involves also other challenges. First, the scope of the regulation remains uncertain. The Regulation on ATMPs applies only to medicines that are ‘intended to be placed on the market in Member States and either prepared industrially or manufactured by a method involving an industrial process.’76 Consequently, this regulation, as for all European pharmaceutical, does not apply to magistral or officinal formula. Furthermore, ATMPs are not covered if they are ‘prepared on a non-­routine basis according to specific quality standards, and used within the same Member State in a hospital under the exclusive responsibility of a medical practitioner, in order to comply with an individual medical prescription for a custom-­made product for an individual patient.’77 Instead, they are regulated at the national level. The individual Member States interpret this ‘hospital exemption’ differently, especially the notion that the exemption applies to ATMPs prepared on a ‘non-­routine basis.’ Some will authorize an ATMP under the hospital exemption only if it is provided to a single patient or a single patient at a time, while others will authorize ATMPs intended for several patients (with different upper limits varying more or less between 5 and 100 according to unofficial discussions with various national competent authorities). As a result, there is competition, which was not anticipated, between ATMPs authorized under the European centralized marketing authorization and ATMPs that are authorized nationally under an exemption for hospitals. Clarification of the scope of the hospital exemption would be helpful. In addition, the definition of ‘ATMP’ itself raises concerns: for example, regarding the distinction between gene therapy medicinal products (which are included) and vaccines

76  Directive 2001/83/EC, Art. 2(1).

77  Regulation (EC) 1394/2007, Art. 28(2).

Regulating Medicines in the European Union   751 (which are not).78 In addition, as noted, the Regulation covers only cells and tissues that have been ‘substantially manipulated’79 or that are not intended to be used for the same essential function or functions in the recipient as in the donor. These criteria have proved difficult to interpret, and it can be hard to predict a product’s classification in advance. Another challenge relates to differentiating between combined ATMPs covered by the ATMPs framework, on the one hand, and medical devices containing tissues and cells covered by the medical devices legislation, on the other hand. The CAT provides recommendations on the classification of ATMPs and has adopted various guidelines on the topic, which have been very helpful, but the challenge remains. Therefore, it seems necessary to accept the definition status will always rely on case-­by-­case analysis. Second, the biological nature of the ATMPs and the complexity of their processes make manufacturing requirements very specific and costly. In addition, special ­qualifications and training are often required for the clinical delivery of ATMPs. This has led to the development of new structural arrangements in the European Union for the manufacture and delivery of these products. For instance, some large manufacturing centers specializing in ATMPs provide their services to multiple ATMPs developers as well as specific treatment centers targeting specific diseases. They sometimes also combine with manufacturing centers especially for autologous products, those in which the patient’s own cells are used.

6 Conclusion EU medicines law is trying to find the best balance between risks of (new) medicines and patients’ access to them, costs to national health systems of novel medicines and treating conditions for which there is no cure, incentivising innovation and creating a single European market while ensuring a high level of protection for patients in the European Union. This is to be done taking into account three contextual challenges. First, as new kinds of medical treatment are developed, it is not always clear where they fit into existing legal categories, as with microbiome-­based products. This is especially true for medicines that have to be delimited from other health products covered by other legal regimes (such as medical devices) and that also include many subdivisions (e.g., orphan drugs, paediatrics, ATMP), often complementary and potentially cumulative. As a result, it becomes difficult to ensure that innovations are subject to the appropriate regulations to protect patients’ welfare, as well as whether they receive 78  For instance, live recombinant viral vectors intended to treat pathologies caused by infection (e.g., malignancies) in oncology should be gene therapy medicinal products, while similar products intended as prophylactics against infectious disease should be vaccines. 79  Annex I of Regulation (EC) 1394/2007 provides a nonexhaustive list of manipulation that are not substantial: cutting, grinding, shaping, centrifugation, soaking in antibiotic or antimicrobial solutions, sterilization, irradiation, cell separation, concentration or purification, filtering, lyophilization, freezing, cryopreservation, vitrification.

752   Aurélie Mahalatchimy appropriate incentives or regulatory support from health agencies to properly promote their development. This is part of the complexity of the EU system. Second, the particularity of Europe is to combine both national law and EU law. For medicines, this challenge is emphasised as long as management of healthcare systems and their resources are regulated at the national levels. Consequently, despite strong EU law regarding the creation of a single European market for medicines, the promotion of research and innovation, and the limitation of unfair practices, the last steps (i.e., directly supplying medicines to patients or making them accessible as a matter of practical reality to patients within the healthcare systems) rely on heterogeneous national systems with different resources and different approaches. Third, the wider context needs to be taken into account, and especially the promotion of the European pharmaceutical sector worldwide. Of course, the internal limits mentioned earlier have consequences here as well. For instance, it is impossible to promote a European system that is not one but the sum of various national systems regarding healthcare management. Nevertheless, the pharmaceutical sector is also regulated to compete globally as a key European industry. The EU approach has clearly focused on the medicines’ access to the market, including as part of a public health objective of patients’ health protection. As such, the legal system has been increasingly strengthened regarding safety (notably pharmacovigilance legislation) and removal of barriers for common and profitable mechanisms (e.g., new regulation on clinical trials). Nevertheless, such approach by stratum has led to a system that is quite fragmented (regarding the different medicines or the different steps of medicines development [specifically] regulated) and lacks a global view on the entire lifecycle of medicines as a continuous, although long, process. Particularly, and despite several attempts, it lacks a coherent part on patients’ access to medicines, while it has been quite successful in fostering medicines’ access to the market. One main reason for that is to protect Member States competency regarding the organization of their national health systems. One example here is the Commission’s proposal of a Regulation on HTA that would contribute to a more efficient use of resources and to a homogeneous and, to a more or less wide extent according to the Member States experiences in this field, to a higher quality of HTA within the European Union. Unfortunately, the current rise of nationalism within the European Union, added to constraints over austerity-­limited budgets, are not good parameters for these problems to be solved.

Bibliography Abraham J, Lewis G. Regulating Medicines in Europe: Competition, Expertise and Public Health (Routledge 2000). Abraham J, Lewis G. ‘Harmonising and Competing for Medicines Regulation: How healthy are the EU’s Systems of Drug Approval’ (1999) 48 Social Science and Medicine 1655. Allensteller C, Permanand  G. ‘EU Regulation of Medical Devices and Pharmaceuticals in Comparative Perspective’ (2007) 24(5) Review of Policy Research 385. Balasubramanian G et al. ‘An overview of Compassionate Use Programs in the European Union Member States,’ (2016) 5(4) Intractable Rare Disease Research 244.

Regulating Medicines in the European Union   753 Bister  S. ‘La sécurité des médicaments et des dispositifs médicaux, quelles améliorations depuis le traité de Lisbonne,’ in De Grove-Valdeyron N, eds., Les nouveaux enjeux de la politique pharmaceutique européenne: pour des produits de santé sûrs, innovants et accessibles (Editions Clément Juglar 2019) 11. Blanquet M, De Grove-Valdeyron N. Etudes de droit communautaire de la santé et du médicament (Presses de l’Université des Sciences Sociales de Toulouse 2009). Bouvy JC et al. ‘Managed Entry Agreements for Pharmaceuticals in the Context of Adaptive Pathways in Europe’ (2018) 9 Frontiers in Pharmacology doi:10.3389/fphar.2018.00280/full. Brosset E. Droit européen et protection de la santé: bilan et perspectives (Bruylant 2015). Brosset E, Mahalatchimy A. ‘EU Law and Policy on New Health Technologies,’ in Hervey TK, Young CA, Bishop LE, eds., Research Handbook in EU Health Law and Policy (Edward Elgar 2017) 197. Chabannon C et al. ‘Les unités de thérapie cellulaire à l’épreuve de la règlementation sur les médicaments de thérapie innovante,’ (2014) 30 (5) Médecine/Sciences 576. Courtney D et al. ‘‘Adaptive Pathways’ to Drug Authorization: Adapting to Industry?’ (2016) 354 BMJ i4437. De Grove-Valdeyron  N. (ed.) La Sécurité des Produits de Santé dans l’Union Européenne (Presses de l’Université des Sciences Sociales de Toulouse 2014). De Grove-Valdeyron N. Droit européen de la santé (LGDJ 2018). De Grove-Valdeyron N. (ed.) Les nouveaux enjeux de la politique pharmaceutique européenne: pour des produits de santé sûrs, innovants et accessibles (Editions Clément Juglar 2019). De Grove-Valdeyron N. Médicaments. Répertoire de droit communautaire Dalloz, juin 2011. Den Exter A. (ed.) European Health Law, (Maklu 2017). Ermisch M et al. ‘Payers’ views of the changes arising through the possible adoption of Adaptive Pathways’ (2016) 7 Frontiers of Pharmacology 305. 10.3389/fphar.2016.00305. Eurordis. Rare diseases Europe. The Voice of Rare Disease Patients in Europe. ‘Early Access to Medicines in Europe: Compassionate Use to Become a Reality,’ 2017 (https://www.eurordis. org/publication/early-access-medicines-europe-compassionate-use-become-reality). Faulkner A, Mahalatchimy A. ‘The institutional and methodological politics of valuation and payment for regenerative medicine products in the UK’ (2018) 37(3) New Genetics and Society 227. Feldschreiber P. The Law and Regulation of Medicines (Oxford University Press 2008). Ferrario A, Kanavos P. ‘Dealing with uncertainty and high prices of new medicines: a comparative analysis of the use of managed entry agreements in Belgium, England, the Netherlands and Sweden’ (2015) 124 Social Science and Medicine 39. 10.1016/j. socscimed.2014.11.003. Ferrario A, Kanavos P, ‘Managed Entry Agreements for Pharmaceuticals: The European Experience’ (2013) EMiNet, Brussels, Belgium (http://eprints.lse.ac.uk/50513/). Flear ML, Farrell AM, Hervey TK, Murphy T (eds.), European Law and New Health Technologies (Oxford University Press 2013). Gardner J et al. ‘Are There Specific Translational Challenges in Regenerative Medicine? Lessons from Other Fields’ (2015) 10(7) Regenerative Medicine 885. Goldberg RS. Medicinal Product Liability and Regulation (Hart 2013). Goldberg R, Lonbay J (eds.), Pharmaceutical Medicine, Biotechnology and European Law (Cambridge University Press 2001). Hancher L ‘The EU Pharmaceutical Markets: Parameters and Pathways,’ in E Mossialos et al., eds., Health Systems Governance in Europe: the role of EU Law and Policy (Cambridge University Press 2010) 635.

754   Aurélie Mahalatchimy Hervey TK, McHale JV. European Union Health Law (Cambridge University Press 2015). Hervey TK, Young CA, Bishop LE. Research Handbook in EU Health Law and Policy (Edward Elgar 2017). Jackson E. Law and Regulation of Medicines (Hart 2012). Leca A, Maurain C, Moine- Dupuis I, Rousset G. Le monopole pharmaceutique et son avenir (Cahier de Droit de la Santé juridiques, historiques et prospectifs 2017). Liberatore F. ‘UK Calls for Ban of Parallel Trade of Prescription Medicines: What Are the EU Competition Law Implications?’ (2013) 34(4) European Competition Law Review 189. Mahalatchimy A. ‘La promotion de l’innovation en matière de santé: quelles logiques à l’œuvre dans l’Union européenne?’ (2017) 4 Revue des Affaires Européennes 627. Mahalatchimy A. ‘Reimbursement of Cell-Based Regenerative Therapy in the UK and France,’ (2016) 24(2) Medical Law Review 234. Mahalatchimy A et al. ‘The Legal Landscape for Advanced Therapy Medicinal Products: Material and Institutional Implementation of European Union Rules’ (2012) 39(1) Journal of Law and Society 131. Mahalatchimy A, Rial-Sebbag E, De Grove-Valdeyron N, Tournay V, Cambon-Thomsen A, Duguet AM, Taboulet  F. ‘The European Medicines Agency: A Public Health European Agency?’ (2012) 31(1) Medicine and Law Journal 25. Mendoza-Caminade A. Médicament et Droit, Droit français et européen (Larcier 2018). Mihokovic  N. EMA Experience and Perspective, Regulatory Challenges in the Drug-Food Continuum. IMI Stakeholder Forum 2017, Microbiome Forum 2017 (powerpoint slides) https://www.imi.europa.eu/sites/default/files/events/SF2017/microbiome_Mihokovic.pdf. Moore TJ, Furberg CD ‘Electronic Health Data for Postmarket Surveillance: A Vision Not Realized’(2015) 38 Drug Safety 601–10. doi:10.1007/s40264-015-0305-9 pmid:26025018. Mossialos E et al. (eds), Health Systems Governance in Europe: The Role of European Union Law and Policy (Cambridge University Press 2010). Permanand G, Mossialos E. ‘Constitutional Asymmetry and Pharmaceutical Policy-Making in the European Union’ (2005) 12(4) Journal of European Public Policy 687. Permanand G, Mossialos E, McKee  M. ‘Regulating Medicines in Europe: The European Medicines Agency, Marketing Authorization, Transparency and Pharmacovigilance’ (2006) 6 Clinical Medicine 87. Pilgerstorfer M. ‘EU Law and Policy on Pharmaceuticals Marketing and Post-Market Control Including Product Liability,’ in Hervey TK, Young CA, Bishop LE, eds., Research Handbook in EU Health Law and Policy (Edward Elgar 2017) 156. Tsouloufas  G. ‘Limiting Pharmaceutical Parallel Trade in the European Union: Regulatory and Economic Justifications’ (2011) 36(3) European Law Review 385.

chapter 36

R egu l ati ng M edica l Dev ices i n the U n ited States Patricia J. Zettler and Erika Lietzan

The goal of the US regulatory framework governing medical devices is the same as the goal of the framework governing medicines. US law aims to ensure that medical devices are safe and effective for their intended uses, that they become available for patients promptly, and that manufacturers provide truthful, non-­misleading, and complete information about the products. The US Food and Drug Administration (FDA) accomplishes this mission by regulating medical devices throughout their lifecycle, just as it does medicines. Often this means a medical device manufacturer must test its product before entering the market, and sometimes it means the company must obtain approval from the FDA. US medical device law is different from US medicines law in many ways, however, perhaps most notably because most marketed devices do not require pre-­market approval. In addition, US medical device law has historically differed substantially from European medical device law by involving a regulator throughout the device lifecycle.1 The law in both jurisdictions has been evolving, however, which may lead to more ­harmonization. Finally, in the United States, individual states play a role in regulating medical devices, just as they do medicines, for instance through establishing rules for product liability.2 1  For example, the current European regulatory framework for medical devices, which includes stricter pre­-market controls and post-­market surveillance than had been required historically, was not established until 2017. For further discussion of the European approach to regulating medical devices, see the chapter by Mahalatchimy,“Regulating Medical Devices in the European Union,” in this volume. 2  See, e.g., Patricia J. Zettler, Pharmaceutical Federalism, 92 Ind. L.J. 845 (2017). In some situations, however, state law may not impose liability on medical device manufacturers. The US Constitution’s “supremacy clause” makes clear that when state and federal law conflict, federal law preempts state law (i.e., the state law has no effect). US Const. Art. VI, cl. 2. The US Supreme Court has held that the FDCA expressly preempts certain products liability claims against device manufacturers. 21 USC. § 360k; Riegel v. Medtronic, Inc., 552 US 312 (2008).

756   patricia j. zettler and erika lietzan The FDA’s mission with respect to medical devices—and the laws that authorize the FDA’s regulatory scheme—often requires the agency to balance goals and values that can be in tension. This chapter explores how the FDA seeks to accomplish its mission and balance competing goals and values through its implementation of its medical device authorities. It starts by explaining what constitutes a medical device and by describing how the FDA classifies medical devices by risk level. The chapter then discusses how medical devices reach the market, the FDA’s risk management tools, and the rules and incentives for innovation and competition. It concludes by exploring case studies of innovative medical technologies that challenge the traditional US regulatory scheme to consider the future of medical device regulation.

1  Basic Concepts For the FDA’s regulatory scheme for medical devices to apply to an item, it must meet the definition of a “device” in US federal law.3 A “device” is an “instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article” that is “intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease” or “intended to affect the structure or any function of the body.”4 Like the definition of medicine in US law, the definition of device focuses on the intended use of the item. “Intended use” refers to the objective intent of the entity responsible for the item’s labeling—typically the manufacturer—and is most frequently, but not always, evinced by the manufacturer’s statements about its product.5 And the intended uses that make an item a device are very similar to those that make an item a medicine. There is one significant distinction: devices include items intended to diagnose both diseases and conditions that are not diseases (such as pregnancy). Otherwise, the focus is the same: devices are items intended to address disease (such as pacemakers that treat heart arrhythmias) and items intended to affect the structure or function of the body. US law distinguishes between medicines and devices in two ways. First, whether an item is a device depends on the item’s intended use and the type of item at issue. A device is an instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article. This broad list of categories includes items commonly understood to be devices (such as replacement heart valves) as well as, according to the  FDA, certain kinds of genetic tests and some software programs. By capturing “machines,” “contrivances,” and similar articles that are intended to affect the structure 3  The item also must travel (or have a component that travels) in interstate commerce. 21 USC. § 331. As a practical matter, however, because of the nature of the industry and modern supply chains, this limitation on the FDA’s jurisdiction rarely matters in the device context. 4  21 USC. § 321(h). 5  21 C.F.R. § 801.4.

Regulating Medical Devices in the United States   757 or function of the body, the device definition also arguably could include a vast array of consumer items. The FDA has generally taken a common sense approach to this aspect of the device definition. For example, the agency will regulate exercise equipment as a device only when the equipment is intended for medical purposes, although all exercise equipment is clearly intended to affect the structure or function of the body.6 The FDA’s assertion of device jurisdiction at times has been controversial in the past, however, leading to litigation and sometimes even statutory changes.7 Second, a medical device does not achieve its primary intended purposes through metabolism in the body or through chemical action within or on the body.8 Such an item would instead be classified as a medicine, which would require compliance with the often more burdensome medicine regulatory framework. As a result, the FDA’s application of this rule has similarly led to litigation in the past.9 Some products include both medicines and medical device components. That is, they are neither solely medicines nor solely medical devices. For example, a pre-­filled epinephrine autoinjector intended for treating life-­threatening allergic reactions, such as EpiPen, comprises both a medicine (the epinephrine) and a medical device (the self-­ injectable syringe that delivers the epinephrine). The FDA considers this a “combination product.” For such products, the agency determines whether the medicine component or the device component makes the greatest contribution to the item’s overall therapeutic effect. This “primary mode of action” generally dictates which regulatory authorities the agency will apply.10 Thus the FDA regulates EpiPen under its regulatory scheme for medicines: epinephrine delivers the product’s most important therapeutic action by stopping the allergic response, while the device component (the autoinjector) simply delivers the epinephrine. At the same time, the FDA regulates cardiovascular drug-­ eluting stents under its scheme for devices: the stent delivers the product’s most important therapeutic action by physically keeping open an artery, while the drug component only augments that therapeutic action by helping to prevent the artery from narrowing again. When US lawmakers established the modern scheme for regulating devices in the Medical Devices Amendments of 1976, they recognized that not every device requires the same level of regulation.11 A tongue depressor intended to aid in a physician’s examination of a patient’s mouth and throat, for example, poses little risk and need not be regulated the same way as an implanted brain stimulator that treats refractory epilepsy. Congress therefore directed the FDA to place every device type into one of three classes 6  21 C.F.R. pt. 890. 7  For instance, in 2005, the US Congress amended the FDCA to ensure that the FDA would regulate noncorrective colored contact lenses as medical devices. 21 USC. § 360j(n). In contrast, in 2016, the legislature amended the statute to ensure that FDA would not exercise device authority over software intended solely for “maintaining or encouraging a healthy lifestyle” without an intent related to “to the diagnosis, cure, mitigation, prevention, or treatment of a disease or condition.” 21 USC. § 360j(o)(1)(B). 8  21 USC. § 321(h). 9  E.g., Prevor v. FDA, 67 F.Supp.2d 125 (D.D.C. 2014). 10  21 USC. § 353(g); 21 C.F.R. § 3.2. 11 Richard A. Merrill, The Architecture of Government Regulation of Medical Products, 82 Va. L. Rev. 1753, 1762 (1996).

758   patricia j. zettler and erika lietzan according to the level of risk it presents and the kinds of regulatory controls necessary to manage that risk. Class III devices present the greatest risk to patients (and include devices such as implantable brain stimulators) and are subject to the strictest requirements, including, typically, pre-­market approval. Class II devices present moderate risks that can be managed through tools known as “special controls,” and they often require some pre-­market authorization from the FDA, although not pre-­market approval. Class I devices (such as tongue depressors) present low risks, can be marketed without seeking permission from the FDA, and are subject to the least regulation. Using this scheme, the FDA categorized all medical devices on the market before the 1976 statute by type (e.g., a “blood pressure cuff ” or a “stethoscope”) and placed each type in the appropriate class (e.g., a blood pressure cuff falls in class II). In addition, after placing a device type in Class III, the agency specified (or will specify) a date by which manufacturers of the device type were to submit pre-­market approval applications (PMAs). Federal law specifies that novel devices that enter the market after the 1976 statute—and are unlike any marketed devices—are automatically placed in class III unless they undergo the reclassification process described in the following section.12

2  Access to the Market A medical device reaches the market in the United States in one of four ways: approval of a PMA comparable to a marketing application for a new medicine, clearance of a much simpler pre-­market notification known as a “510(k),” grant of a petition to reclassify a device to a lower class, and (in the case of low-­risk devices) simply by registration of the company and device with the FDA. We discuss the first three paths in the subsections that follow. Regardless of a device’s classification and path to market, federal law requires the FDA to take the least burdensome approach to pre-­market evaluation permitted under the relevant statutory standard for authorization.13 This rule is meant to further the goal of ensuring medical devices are safe and effective without unduly delaying patient access.14

2.1  The Pre-­Market Approval Application Generally, the manufacturer of a class III medical device may not market the device unless the FDA has approved a PMA. Some medical devices in class III may be marketed on the basis of a pre-­market 510(k) notification, discussed in the next subsection, 12  21 USC. § 360c(f). 13  FDA, Draft Guidance: Least Burdensome Provisions: Concept and Principles (2017), https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/ GuidanceDocuments/UCM588914.pdf. 14  21 USC. § 393.

Regulating Medical Devices in the United States   759 because the particular device type was marketed before the 1976 Medical Device Amendments and the FDA has not yet called for the submission of PMAs. Eventually, however, the agency will do so, and all class III devices will be subject to the PMA requirement. In addition, any new type of device never before marketed in the United States falls automatically in class III and must be the subject of a PMA15—although the manufacturer may persuade the agency to place it in a less regulated class. Like a medicine, a medical device subject to the requirement for a PMA must be shown safe and effective under the conditions of use described in its labeling (i.e., that the device’s benefits outweigh its risks).16 The manufacturer’s application must provide a “reasonable assurance” of safety and effectiveness. US law does not explicitly require data from testing in humans, but it does require full reports from any investigations that have been performed, and, as a practical matter, the agency generally requires clinical data from randomized, controlled clinical trials. A manufacturer must comply with good clinical practices, including human subject protection regulations, when conducting those trials, and, if the medical device puts subjects at “significant risk,” the manufacturer must submit an investigational device exemption (IDE) to the FDA.17 Once the FDA has approved the PMA, the company may market the device described in the application. If the company wishes to make changes that affect the device’s safety or effectiveness, it must supplement its application and, generally, wait for the agency’s approval.18 Some changes, including the addition of new warnings to the labeling, can be implemented before the FDA has approved the supplement in question, as with medicines. Many manufacturers follow the traditional approach of completing a pre-­market research and development program and then preparing and submitting a complete marketing application. The FDA has, however, developed two alternative approaches that are designed to allow earlier collaboration with a manufacturer. In theory, these earlier interactions could speed approval, which in turn may reduce development costs and give patients access to the device sooner. First, a company may submit its PMA for FDA review one module at a time.19 For instance, it might submit a module of preclinical laboratory studies, then a module of manufacturing information, and then a module of clinical data. The result is a “modular PMA”—a PMA comprising different sections submitted at different times that together become a complete application. This approach can speed the approval process, for instance by allowing the manufacturer to resolve deficiencies with respect to each module as the FDA reviews it. Second, the company may follow a product development protocol (PDP), essentially a contract with the agency laying out the work that must be done to provide reasonable assurance of the safety and 15  21 USC. § 360c(f). 16  21 USC. § 360e(d)(2). 17  21 C.F.R. §§ 812.1, 812.2. The protections for human subjects, and IDE requirements, in FDA regulations are very similar to those for medicines, discussed in the chapter by Lietzan and Zettler, “Regulating Medicines in the United States,” in this volume. 18  21 C.F.R. § 814.39. 19 FDA, Guidance for Industry: Pre-­market Approval Application Modular Review (Nov. 2003).

760   patricia j. zettler and erika lietzan effectiveness of a device.20 It typically describes the development activities in detail as well as the outcomes that will be measured and the results needed for approval. The company may perform the protocol at its own pace, submitting milestone reports as specified in the protocol, and, when the FDA finds the PDP complete, the company is deemed to have an approved PMA.

2.2  The 510(k) Notification Only a small percentage of medical devices reach the market through the PMA pathway. In fact, most do not go through any pre-­market process at all. According to the US Government Accountability Office, for instance, 67% of the medical devices that entered the market from 2003 to 2007 were exempt from any pre-­market review.21 Only 1% entered the market on the basis of an approved PMA; examples include the implantable brain stimulator mentioned earlier as well as carotid stents and temporary total artificial heart systems. The remaining medical devices reached the market pursuant to a pre-­ market notification (PMN), sometimes called a 510(k) after the key provision of the Federal Food, Drug, and Cosmetic Act (FDCA) governing the process.22 The 510(k) is thus the most common type of pre-­market submission for a medical device in the United States. A 510(k) notification claims that the medical device is substantially equivalent to another medical device already on the market (its “predicate” device).23 Although this submission involves a comparison to another manufacturer’s product, it is not analogous to an abbreviated application for a generic or biosimilar medicine. An abbreviated application for a new medicine omits safety and effectiveness data because it relies on the safety and effectiveness data in an earlier application approved by the FDA. But for a medical device pre-­market notification, by design, the agency has never seen a full application with safety and effectiveness data. The pre-­market notification is designed for low- and moderate-­risk devices that are similar to other low- and moderate-­risk devices, none of which has pre-­market approval. To be substantially equivalent to a predicate device, the new device must have the same intended use as the predicate device. In addition, the technological characteristics of the new and predicate devices must be the same, although differences are permitted if they are shown not to raise new questions of safety and effectiveness.24 As a practical matter, this can require the manufacturer to include clinical data in its submission. As would be the case if it were generating clinical data for a PMA, the manufacturer must comply with good clinical practices, although generally it will not need to submit an IDE, because the device in question should not put subjects at significant risk. 20  21 USC. § 360e(f); C.F.R. § 814.19. 21 GAO, Medical devices: FDA should take steps to ensure that high-­risk device types are approved through the most stringent pre-­market review process (GAO-­09-­190) (2009). 22  21 USC. § 360(k). This provision is section 510(k) of the FDCA. 23  21 USC. § 360c(f)(1). 24  21 USC. § 360c(i).

Regulating Medical Devices in the United States   761 The FDA does not approve pre-­market notifications. But a manufacturer must wait 90 days to market the device, and the agency may decide that it needs more time or more information. Ultimately the FDA notifies the manufacturer of its decision on the device, and a device found substantially equivalent is said to be “cleared” by the agency. This decision does not represent a finding that the device is safe and effective for its intended use. And the decision is limited to the device described in the pre-­market notification. There is no such thing as a 510(k) “supplement.” If the manufacturer makes a change (including a change to the device’s intended use) that could significantly affect the device’s safety or effectiveness, the existing 510(k) simply does not apply—and the manufacturer will need to submit a new 510(k) or, if the changes are so significant that there is no suitable predicate device, a PMA.25 It may make less significant changes without a new pre-­market notification provided it documents them appropriately. The pre-­market notification program is controversial for several reasons. First, predicate devices have not received much scrutiny by the FDA. Many were marketed prior to 1976 and simply placed in class II by the agency after Congress added the classification scheme to the statute. Most are moderate- to low-­risk devices, to be sure, but the agency has never looked at their safety or effectiveness. Second, a surprising number of implanted devices—such as hip implants and vascular embolization devices—fall in class II and reach the market on the basis of pre-­market notifications. The agency’s failure to examine the safety and effectiveness of these devices, and their predicates, may be concerning. In some cases these devices are marketed on the basis of predicates later recalled for safety reasons, and some seem to differ meaningfully from their predicates in terms of materials and where they are used in the body.26 Indeed, this relates to a third, broader concern about the 510(k) process: that some devices cleared today look and operate nothing like the original devices that FDA placed in class II. A device may be cleared through the 510(k) process even with potentially substantial differences from the predicate in technological characteristics as long as the manufacturer proves that the changes raise no new issues of safety and effectiveness.27 Another company could submit a 510(k) notification citing the second product and similarly make potentially substantial differences. Many devices marketed today are supported by a “daisy chain” of predicates dating back more than 40 years.28 As an Institute of Medicine committee wrote in 2011, “Device A might be found substantially equivalent to Device B, which had been found equivalent to Device C, which had been found equivalent to Device D, and so on.”29 In such circumstances, even if only minor changes are made on each subsequent device, the end device might be quite different from the original device that started the chain. Finally, inadequacies in other aspects of the agency’s oversight of 25  21 C.F.R. § 807.81. 26  Diana Zuckerman et al., Lack of Publicly Available Scientific Evidence on the Safety and Effectiveness of Implanted Medical Devices, 174 JAMA 1781 (2014). 27  21 USC. § 360c(i). 28  Institute of Medicine, Medical Devices and the Public’s Health: The FDA 510(k) Clearance Process at 35 Years (2011), at 91. 29  Id. at 33.

762   patricia j. zettler and erika lietzan medical devices, such as its post-­market surveillance system, may make it impossible to be confident about the performance of these medical devices.30 A more rigorous pre-­market process for class II devices, however, is also controversial. Most devices subject to 510(k) notification present only moderate risk, and pre-­ market notification is more economical, faster, and less burdensome than pre-­market approval. The burden and expense of the PMA process might deter some manufacturers from developing new medical devices, which could deprive patients of the benefits of improvements in medical technology and eliminate a source of price competition in the market. The companies that did pursue PMAs would need to recoup their investments, which could lead to an overall increase in costs to the healthcare system. Moreover, a more rigorous pre-­market process is not the only option—policymakers also might consider greater use of performance standards for devices that are on the market (described in more detail in Section 3, “Post-­Market Risk Management”). The FDA has the legal authority to require that class II devices comply with these standards, which could include the obligation to submit some effectiveness data after marketing. In the alternative, policymakers might consider the recommendation of the Institute of Medicine—that the 510(k) process be scrapped in its entirety and replaced with an integrated pre-­market and post-­market framework that provides a reasonable assurance of safety and effectiveness throughout a device’s lifecycle.31 Policymakers considering these alternatives must consider whether these devices present sufficient risk to justify a more onerous pre-­market process that may ultimately delay access and increase prices for patients.

2.3  De Novo Classification As noted earlier, if a manufacturer’s new device is nothing like another device on the market—if it is a novel device type—then federal law places it automatically in class III and an approved PMA is needed to market the device.32 In some cases, however, a novel device may be a low- or moderate-­risk device. In these cases, the pre-­market safety and effectiveness testing and formal FDA approval process will impose costs—delay in access and higher prices—that may not be justified. Federal law therefore permits a manufacturer to submit a “de novo classification” petition explaining why the device should instead be placed directly in class I or class II.33 If the FDA grants the petition, the company may market its device immediately. As discussed in Section 5.1, the FDA used this process recently to permit the marketing of several direct-­to-­consumer (DTC) genetic tests. The data requirements for de novo classification petitions may present a difficult business choice for some manufacturers. A classification petition is a considerable undertaking and generally must be supported by extensive safety and effectiveness 30  Id. at 9–10. 31  Id. at 8. 32  21 USC. § 360c(f). 33  21 USC. § 360c(f)(2).

Regulating Medical Devices in the United States   763 information.34 A device marketed on the basis of a reclassification petition may serve as the predicate for another manufacturer’s 510(k) notification, however, which means that the first manufacturer’s investment could directly benefit its competitors. As a result, if the data requirements for de novo classification are too burdensome, the process might be underused. A company might prefer to file a PMA, which would benefit from a period of data exclusivity (see Section 3.1) and would not clear a pathway to market for the manufacturer’s competitors. This, in turn, could lead to fewer medical devices competing in the marketplace and higher prices for the devices that are available. In its review of a classification petition, FDA focuses appropriately on whether the manufacturer proved that the device presents low or moderate risk. But policymakers may ultimately need to face the task of motivating manufacturers to choose the de novo pathway rather than the PMA pathway.

2.4  The Tradeoff Between Wanting Evidence and Wanting Access As with medicines, pre-­market approval of medical devices takes time and money. The process can take 3–7 years from concept to market,35 and the cost for higher risk devices can approach $100 million.36 Devices that are subject to pre-­market approval are generally more expensive and they reach the market more slowly than other devices. The amount and type of data that the FDA requires in any particular type of pre-­market application can affect this cost and delay. As a result, in setting the data requirements for any particular class III device, the agency must balance the social cost of additional research with the benefit of greater certainty about the device’s risks and benefits. In some cases, patients will want access to a medical device before the manufacturer has generated the data that the FDA has concluded are necessary to illustrate a positive benefit-­to-­risk ratio. In other cases, the tradeoff between wanting data or certainty, on the one hand, and wanting to provide prompt access to an affordable device, on the other hand, has led policymakers to an entirely different approach to market entry.

2.4.1  Expanded Access The classic situation—“expanded access”—involves a patient seeking access to a medical device that is still in clinical trials. Such expanded access can be available when a patient has a serious or life-­threatening condition that cannot be adequately addressed with available treatments and does not qualify for a clinical trial (or the trial is no longer 34  See FDA, Guidance: De Novo Classification Process (Evaluation of Automatic Class III Designation) (Oct. 2017). 35  Kyle Fargen, The FDA Approval Process for Medical Devices: An Inherently Flawed System or a Valuable Pathway for Innovation?, 5 J. Neurointerv. Surg. 269 (2013). 36  Josh Makower et al., FDA Impact on US Medical Technology Innovation: A Survey of Over 200 Medical Technology Companies (2010).

764   patricia j. zettler and erika lietzan enrolling patients). For example, one device company provided expanded access to an experimental nerve stimulator for a small group of patients with severe, chronic headaches that had not responded to any available medical therapies. In nonemergency situations, the FDA must authorize the expanded access program before the patient or patients are treated. The FDA grants the vast majority—typically more than 99%—of these requests.37 Typically these are requests for individual or just a few patients, but in rare cases the agency may authorize widespread “treatment use” of an investigational device based on the data emerging in the clinical trials regarding its effectiveness.38 In an emergency situation, on the other hand, a physician may use an investigational device to treat a patient who is not enrolled in the clinical trial or may use an investigational device in a way that is inconsistent with the protocol for the trial without prior FDA authorization. The physician is expected to follow as many patient protection procedures as possible and is required to report the use to the FDA after the fact. The FDA’s regulatory framework governing expanded access to medical products has evolved over the years, largely in response to policymaking efforts initiated by patients seeking expanded access to medicines rather than medical devices. The FDA receives many more requests for expanded access to medicines than to devices: for example, in 2015, the agency received roughly 1,600 requests for medicines compared to 385 for devices.39 After frustration regarding access to investigational medicines led to patient activism and largely unsuccessful lawsuits, policymakers responded in part by enshrining such pre-­approval access in the FDCA in 1997.40 The law applied equally to medicines and devices. Subsequent policymaking, however, has focused more on medicines. Most significantly, the Federal “right to try” law enacted in 2018—which creates a mechanism for nonemergency access without FDA permission—applies only to medicines.41

2.4.2  Humanitarian Use Devices Policymakers in the United States have struck a different balance between ensuring patient access and ensuring robust pre-­market evidence of safety and effectiveness for devices intended to treat or diagnose rare diseases and conditions than for such medicines. The concept of a humanitarian use device (HUD) is similar to the concept of an “orphan” medicine: both are intended for rare diseases and, therefore, small patient populations. A HUD is intended for use in the treatment or diagnosis of a disease or condition affecting no more than 8,000 individuals in the United States per year.42 37  FDA, Expanded Access for Medical Devices (https://www.fda.gov/medical-­devices/ investigational-­device-­exemption-­ide/expanded-­access-­medical-­devices). 38  21 USC. § 360bbb. 39  FDA, Expanded Access for Medical Devices, https://www.fda.gov/MedicalDevices/ DeviceRegulationandGuidance/HowtoMarketYourDevice/InvestigationalDeviceExemptionIDE/ ucm051345.htm; FDA, Expanded Access INDs and Protocols, https://www.fda.gov/Drugs/ DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/ DrugandBiologicApprovalReports/INDActivityReports/ucm373560.htm; FDA, CBER Expanded Access Submission Receipt Reports, https://www.fda.gov/BiologicsBloodVaccines/ucm413041.htm 40  21 USC. § 360bbb. 41  21 USC. § 360bbb-­0. 42  21 USC. § 360j(m).

Regulating Medical Devices in the United States   765 Examples include a bladder stimulation device for the small group of children with spina bifida in the United States and a replacement heart valve for children aged 4 and younger with certain rare heart anomalies. A very small patient population can make it difficult to generate robust evidence of effectiveness in pre-­market trials, and it can also reduce the revenue possible after pre-­market approval. For orphan medicines, policymakers chose to offer an extra incentive for the research in question, allowing a more robust opportunity to recover the investment made.43 In practice, the FDA sometimes approves orphan medicines on the basis of very small trials. Rather than adding an incentive for HUDs, policymakers focused directly on the research needed for approval. The FDA may exempt a HUD manufacturer from the obligation to submit effectiveness data whether in a pre-­market approval application (for a class III device) or pursuant to a performance standard (in the case of a class II device). To obtain this exemption, a company must ask that the device be designated as a HUD and then separately submit a marketing application, known as a humanitarian device exemption (HDE). The FDA will grant an HDE if it concludes that the device does not pose an unreasonable risk to patients and that the probable benefits of the device exceed its probable benefits, considering other devices and treatments available. The tradeoff for manufacturers is that they generally may not sell a HUD for a profit, but there are exceptions, including for devices intended for pediatric populations.

2.4.3  Custom Devices Finally, policymakers have dispensed entirely with pre-­market testing requirements for custom devices, similar to compounded medicines. A custom device is tailored to a specific patient or specific healthcare professional, to address a unique pathology or physiological condition.44 For instance, a manufacturer might modify a hip replacement device to accommodate a patient with skeletal dysplasia that deformed her hip.45 Or a surgical device might be modified to accommodate a deformity in the surgeon’s hand.46 Such a custom device is exempt from pre-­market approval requirements and any performance standards that would otherwise apply. Unlike the HDE, the custom device exemption is not meant as a pathway to market. The custom device exception does not authorize a company to develop and sell a customizable device. Instead, the notion is that a company that sells a medical device may need to customize that device, from time to time, for individuals. As a result, a custom device must be created or modified in response to a specific order from a physician or dentist, and the company cannot produce more than five units of the custom device type in any year.

43  See the chapter by Lietzan and Zettler, “Regulating Medicines in the United States,” in this volume. 44  21 USC. § 360j(b). 45 FDA, Guidance for Industry: Custom Device Exemption (Sept. 2014), at 8. 46  Id. at 9.

766   patricia j. zettler and erika lietzan

3  Post-­A pproval Risk Management Because no device can be perfectly safe (although some class I devices may come close), the FDCA authorizes the FDA to take steps to manage the risks of medical devices that enter the market. The agency’s use of these risk management tools is intended to protect the public health by ensuring that a device is safe and effective when it enters the market and that, as more is learned about the device over time, it remains safe and effective. These goals must be balanced against the need for patients to have access to medical treatment and, in some cases, patient willingness to accept risk. Thus, for instance, if patients are benefitting from a marketed device about which new risk information emerges, the preferred approach is to find a way to manage the newly identified risks rather than removing the device from the market. FDA’s use of its risk management powers also allows room for physicians to exercise their judgment about the care best for their patients. To that end, federal law expressly directs the FDA not to unduly interfere with the ability of physicians to use devices off-­label.47

3.1  Managing Risk at the Time of Market Authorization The FDA exercises many of its risk management authorities when it decides to authorize the marketing of a device, and many of these authorities are similar to those it uses to manage the risks of medicines. The first risk management decision, of course, is the decision whether to allow a particular medical device on the market. As with any other medical product, this decision ultimately reflects a conclusion that—based on the information available—the device is appropriate for marketing when it is used in accordance with its labeling. The agency also makes decisions, at the time of market entry, about risk management tools that will be used while the device is marketed. For instance, as it does with medicines, the FDA decides at the time of authorization whether a device needs a physician’s supervision for safe use and thus requires a prescription or instead may be sold over-­ the-­counter directly to consumers.48 The FDA also requires that individual devices be adequately labeled to ensure their safe and effective use.49 Labeling mandates can include specific requirements for types of devices, such as the obligation to warn about toxic shock syndrome in the labeling of all menstrual tampons.50 The FDA can also use “special controls” to ensure the safety and effectiveness of class II devices.51 Typically the agency issues a guidance document that specifies the special controls that it recommends for a particular type of device. These might include steps to 47  21 USC. § 396. For discussion of what constitutes an “off-­label” use, see the chapter by Lietzan and Zettler, “Regulating Medicines in the United States,” this volume. 48  21 C.F.R. § 801.109. 49  21 C.F.R. pt. 801. 50  21 C.F.R. § 801.430. 51  21 USC. § 360c(a)(1)(B).

Regulating Medical Devices in the United States   767 take before market entry (such as generating specific types of information for the 510(k) submission), recommended design or performance characteristics, or even recommended conditions of marketing (such as special labeling requirements or a patient registry). For example, special controls for digital mammography devices include, among other things, instructions to assure the accuracy of imaging through testing and recommendations in the labeling.52 These guidance documents are not binding, but the theory of a special control is that the device is not safe and effective without adoption of these measures. As a practical matter, therefore, a company must comply with them. The FDA can also use rulemaking to impose special controls known as performance standards for a particular category of device, which in turn are binding.53 These, too, can specify the permitted design and construction of the device, dictate the testing that must be done, specify performance characteristics, or mandate special labeling. For instance, sunlamps, which are regulated as devices, must be designed with a maximum timer interval no greater than the maximum recommended exposure time, among other things.54 The FDA may restrict the sale, distribution, or use of any type of medical device by regulation if doing so is necessary to provide a reasonable assurance of its safety and effectiveness.55 It can also restrict sale, distribution, or use of a class III medical device simply as a condition of approval.56 Restrictions can include a range of measures. For example, for hearing aids, FDA regulations require, among other things, that m ­ anufacturers provide specific information in physician and patient labeling and that dispensers sell hearing aids only to patients whose hearing has been medically evaluated in the previous 6 months.57 As noted later, once FDA restricts the sale, distribution, or use of a device, it has jurisdiction over the device’s advertising.

3.2  Monitoring and Managing the Risks of Marketed Devices The full range of risks associated with a device—particularly a class III device—frequently cannot be understood at the time of market entry. Pre-­approval clinical trials are ­typically conducted in relatively small populations and in tightly controlled settings. Additional risks and new information about previously known risks emerge as devices are used in medical practice in a less controlled setting. For this reason, the FDA requires manufacturers to monitor the risks associated with their devices after they enter the market, just as it requires medicine manufacturers to do. 52  See FDA, Class II Special Controls Guidance Document: Full Field Digital Mammography System (Mar. 2012), https://www.fda.gov/regulatory-­information/search-­fda-­ guidance-­documents/class-­ii-­special-­controls-­guidance-­document-­full-­field-­digital-­mammography­system-­guidance-­industry#6. 53  21 USC. § 360d. 54  21 C.F.R. § 1040.20. 55  21 USC. § 360j(e). 56  21 USC. § 360c(d)(1)(B)(ii). 57  21 C.F.R. §§ 801.420, 801.421. Since 2016, the FDA has chosen not to enforce certain restrictions on hearing aids, including the regulatory requirement for hearing evaluations. See FDA, Guidance for Industry and Food and Drug Administration Staff: Conditions for Sale for Air-­ Conduction Hearing Aids 1 (2016).

768   patricia j. zettler and erika lietzan Device manufacturers and importers, and certain facilities that use devices such as hospitals, must report certain adverse events and device malfunctions to the FDA.58 Others, such as physicians and patients, may voluntarily report adverse events and malfunctions. The adverse event reporting system for devices has the same limitations as that for medicines: namely, required reporters may not become aware of many adverse events, many adverse events are not reported, and those reports that are submitted may not contain adequate information to evaluate the potential risk. Accordingly, another way that the FDA may monitor the safety of marketed devices is by ordering manufacturers of certain class II and class III devices to actively and systematically collect and analyze information about the devices.59 For example, in 2016, the FDA required the manufacturer of Essure, an implantable, permanent birth control device, to conduct a prospective observational study of patients who received the device or underwent tubal sterilization to better understand the device’s risks.60 Additionally, the FDA conducts its own active monitoring of existing databases, such as electronic health records, through its Sentinel Initiative discussed in the US medicines chapter. The FDA is aided in all of these monitoring activities by the unique device identification system which, when fully implemented, will require most devices to have unique identifiers on their labeling.61 One purpose of this system is to improve the accuracy of adverse event reporting by enabling definitive identification of devices. When these surveillance activities identify a problem with a marketed device, the agency possesses various tools to manage the risk. Changing a device’s labeling is often a first step. For example, recently evidence emerged suggesting that a type of uterine cancer is more common than was previously thought in women undergoing surgery for noncancerous uterine growths and that laparoscopic power morcellators used in this surgery can spread undetected cancer beyond the uterus, worsening a patient’s chances for long-­term survival. In response, the FDA recommended changes to the labeling for the devices to include information about the risk of spreading cancer (after which at least one large manufacturer voluntarily withdrew its device from the market altogether). The FDA might also modify special controls in place for the device type or add special controls if none exists, or it could impose new restrictions on the device’s use or distribution. Device manufacturers, likewise, may take steps to address problems associated with devices that are currently on the market. A device manufacturer may physically remove a device from its location to fix, inspect, or destroy it. For example, a manufacturer that made defective lots of a device used for procedures, such as a coronary catheter, might contact hospitals to arrange the devices’ return for destruction. Or the company might “correct” the device—that is, fix, inspect, or destroy the device without physically removing it from its location. For example, if a defibrillator malfunctions due to a soft58  21 C.F.R. pt. 803. 59  21 USC. § 360l. 60  Ultimately, in 2018, the FDA restricted sale and distribution of the device based on safety concerns, and then the manufacturer voluntarily halted sales. FDA, FDA Activities: Essure, https:// www.fda.gov/medical-­devices/essure-­permanent-­birth-­control/fda-­activities-­essure. 61  21 USC. § 360i(f).

Regulating Medical Devices in the United States   769 ware issue, the manufacturer might correct the device by scheduling an update of the software. A manufacturer must notify the FDA if it performs a removal or correction to reduce a health risk.62 The FDA generally refers to these actions—removals and corrections—as “recalls.” Additionally, the FDA may order a recall if there is a “reasonable probability” that the device will “cause serious, adverse health consequences or death.”63 By way of contrast, the agency does not have the power to order a recall of a medicine. Recalls can be complicated for devices. In some cases, devices cannot be easily pulled from the market or fixed. For example, removing implanted devices, such as pacemakers, clearly poses risks to patients that must be compared with the risk of simply leaving the devices in place. Additionally, what constitutes an appropriate risk management strategy depends on the degree of risk associated with the device malfunction or adverse event. Some problems are relatively minor or do not pose a significant health risk, such as a mislabeled test to screen urine for certain drugs. Other problems present a risk of temporary or reversible harm or even a reasonable chance of serious harm, such as a defective component in a defibrillator that causes the device to fail to deliver the necessary shock to a patient. In part for these reasons, recalls do not always involve recovering all stock that has entered the market. Instead, manufacturers develop recall strategies that address various aspects of the recall, including plans for communicating the reasons for the recall and the “depth” of the recall (how far into the distribution chain the recall will extend). One risk management challenge unique to and critical for the future of device regulation is cybersecurity: ensuring the security of the information stored in or transmitted by medical devices.64 Indeed, in 2017, the FDA announced that more than 400,000 pacemakers were vulnerable to hackers unless and until a firmware update was performed at an in-­person visit with a healthcare provider. This correction action highlighted the vulnerability of many devices. Even before this incident, the FDA had begun to address cybersecurity threats to devices, including collaborating with other government agencies and recommending steps that manufacturers can take to identify, assess, and address threats to their devices. But cybersecurity threats are an emerging problem that the FDA and device manufacturers will need to address to capture the benefits of technological progress while appropriately managing the risks of that progress. Finally, the FDA can suspend approval of a class III device if it concludes that continued distribution would cause serious adverse health consequences or death.65 Suspension is followed by formal proceedings to withdraw approval of the PMA. The agency does not have equivalent power with respect to a 510(k) notification, which it does not approve in the first instance.66 The agency may, however, reclassify the device type from class II to class III and set a date for submission of PMAs, which would ultimately preclude the manufacturer from continuing to market on the basis of its 62  21 C.F.R. § 806.10. 63  21 USC. § 360h(e). 64  See, e.g., Scott J. Shackelford et. al., Securing the Internet of Healthcare, 19 Minn. J.L. Sci. & Tech. 405, 406 (2018). 65  21 USC. § 360e(e)(3). 66  Ivy Sports Medicine LLC v. Burwell, 767 F.3d 81 (D.C. Cir. 2014).

770   patricia j. zettler and erika lietzan substantial equivalence showing.67 For example, in 2016, the FDA reclassified surgical mesh for the transvaginal repair of pelvic organ prolapse from class II to class III, requiring manufacturers to submit PMAs by 2018, following reports of serious adverse events including dislodging, infection, and death. In 2019, the agency ultimately decided that the devices lacked a reasonable assurance of safety and effectiveness and could not be marketed. Regardless of the mechanism, however, withdrawing marketing authorization is generally a last resort for risk management. If some patients are benefiting from the devices, and the risks can be managed without disrupting care for those patients, steps short of market withdrawal are generally the preferred approach.

4  Innovation and Competition Like pre-­market testing of new medicines, pre-­marketing testing of medical devices— whether for a PMA or a 510(k) notification—imposes costs on manufacturers that must be recouped in the marketplace. And, just as is true for new medicines, this is made possible primarily through provisions of US law that provide the manufacturer with a period for exclusive sales in the marketplace. But US medical device law differs fundamentally from US medicines law in one basic respect: there is no such thing as a “generic” medical device under US law.68 This affects the design and impact of US incentives for medical device innovation as well as the nature and impact of competition in the medical device marketplace. Incentives for medical device innovation require policymakers to make the same basic tradeoff as do incentives for innovation in medicine. By design, a period for exclusive marketing leads to higher prices for a period of time, which may limit access for some patients and create higher costs for the healthcare system. But it may also lead to the development of some technologies that would not otherwise be developed, including technologies that lower healthcare costs. Thus, the regulatory scheme not only provides a social benefit (by prompting the generation of data about medical devices that might not otherwise be available) but also imposes a social cost (by driving up the price of the medical device while the manufacturer recovers its expenses). This forces policymakers to balance providing and protecting incentives for manufacturers to spend their resources developing new medical devices, on the one hand, and facilitating market launch of lower cost competing devices, on the other hand.

67  21 USC. § 360c(e). See, e.g., 81 Fed. Reg. 364 (Jan. 5, 2016). 68  One company could copy another company’s medical device once the relevant patents have expired, to be sure. But there is not a special—less burdensome—pathway to market for these copies. In contrast, there are special less burdensome pathways to market for generic medicines. For a discussion of how generic medicines are marketed under US law, see the chapter by Lietzan and Zettler, “Regulating Medicines in the United States,” in this volume.

Regulating Medical Devices in the United States   771

4.1  Incentives for Innovation US law generally encourages innovation with the prospect of a period during which to market the resulting products without competition. For medical devices, this incentive mainly takes the form of a patent. The Patent Act permits the patenting of any new and useful process or composition of matter provided various other requirements of the Patent Act are satisfied.69 Medical devices and their uses qualify, and it is often possible to file a medical device patent application earlier than a drug patent application because some requirements of the Patent Act (such as establishing utility) are easier to satisfy. Once granted, the patent will allow the medical device patent owner to prevent any other company from making, using, or selling its invention for a period of time. Federal law also protects the safety and effectiveness data in a medical device pre-­ market approval application for 6 years.70 During this period of “data exclusivity,” the data cannot be used to establish the safety or effectiveness of another device without the consent of the PMA holder. After data exclusivity expires, however, FDA can make the data available and can use the data to approve another medical device. It can also use the data to reclassify the medical device or, if appropriate, another medical device. But there is no such thing as a generic medical device. The paradigm for medicines involves full applications by innovators and—some number of years later—abbreviated applications by their generic competitors. The generic competitors submit comparative data sufficient to justify reliance on the innovator’s data. This bifurcated approach, in which a full application is followed by abbreviated applications after data exclusivity expires, has no parallel in the device world. As a result, the end of a medical device’s data exclusivity period does not mean the market will be flooded by drastically less expensive copies of the device. US law does not provide data exclusivity for the safety and effectiveness data in a 510(k) notification. More importantly, though, unlike the PMA requirements, the 510(k) clearance requirements could themselves discourage innovation that would be in the interests of patients—and the FDA cannot solve this problem without raising serious questions about the safety and effectiveness of cleared devices. As noted earlier, a 510(k) notification must show that the proposed device is substantially equivalent to a predicate device. This may discourage medical device manufacturers from making important innovations to medical devices that would make treatment safer, more effective, or less expensive. Although the FDA has sometimes been fairly flexible about the extent of changes it will permit, meaningful changes attract criticism (as discussed in Section 2.2) in part because the agency permits daisy-­chained 510(k) notifications going back decades and in part because none of the devices is supported by robust pre-­market safety and effectiveness testing.

69  35 USC. § 101. 70  21 USC. § 360j(h)(4); FDA, Guidance on Section 216 of the Food and Drug Administration Modernization Act of 1997 (2000).

772   patricia j. zettler and erika lietzan

4.2  Competition and Advertising Competition in the medical device marketplace is very different from competition in the prescription drug and biologic marketplace in part because there is no such thing as an abbreviated marketing application for a “generic” medical device. The size and scope of medical device pre-­market submissions depends on a device’s classification, which in turn depends on its risk. The 510(k) is abbreviated in the sense that it does not purport to show safety and effectiveness, and it draws a comparison with another device. But it does not rely on the predecessor’s clinical data, and, at least in theory, the two manufacturers faced roughly the same pre-­market burden. In addition, there is no mechanism by which federal or state law requires, or encourages, the substitution of one company’s medical device for another. Medical device manufacturers must compete for sales in the marketplace, for example on the basis of product features or price. In short, unlike competition in the marketplace for medicines, all competition in the medical device marketplace is brand-­to-­brand competition among companies that face similar pre-­market barriers. Two federal agencies have responsibility for regulating advertising and promotion in this marketplace: the FDA and the Federal Trade Commission (FTC). First, the FTC, rather than the FDA, regulates advertising for most medical devices pursuant to its authority under sections 5 and 12 of the Federal Trade Commission Act.71 Generally, claims must be truthful, cannot be deceptive, and must be evidence-­based. The FTC expects companies to have a reasonable basis for any express or implied claim, and claims about effectiveness or safety must be supported by “competent and reliable scientific evidence,” generally meaning the amount of data that experts in the field generally accept as sufficient. The FDA regulates advertising for “restricted” medical devices— devices with special use or distribution restrictions.72 It also regulates the labeling of all medical devices, which includes written, printed, or graphic promotional materials that may not be physically connected to the products.73 As a practical matter, the FDA applies the same principles to regulating the promotion of medical devices whether that promotion takes the form of labeling for a medical device or advertising for a restricted medical devices74 Promotional pieces cannot be false or misleading. Among other things, they must reveal material facts about the product being promoted, and they should present information about effectiveness and information about risk in a balanced manner. A company may not make false or misleading statements about a competitor’s product, and the FDA requires that any comparative claim be substantiated by head-­to-­head studies. In addition, even though a physician generally may prescribe or use a medical device for any use, including a use for which it is not approved,75 the manufacturer may not promote the device for any uses other than 71  15 USC. §§ 45, 52. 72  21 USC. § 352(q). 73  21 USC. § 321(m). 74  E.g., FDA, Guidance for Industry: Presenting Risk Information in Prescription Drug and Medical Device Promotion (May 2009). 75  21 USC. § 396.

Regulating Medical Devices in the United States   773 those covered by its clearance or pre-­market approval. The primary policy justifications for this prohibition are the same as those for prohibiting off-­label promotion of medicines: it helps to manage the risks of medical devices, preserve the integrity of the FDA approval process, and motivate companies to develop information about medical devices through rigorous research. A series of recent court cases has made it clear that the First Amendment to the US Constitution places constraints on the FDA’s ability to prevent manufacturers from speaking truthfully about unapproved uses of their medical devices, and the agency’s policies are evolving.76

5  Challenges to the Traditional Paradigm As noted, the fundamental structure of the US device regulatory scheme was established in 1976. Over the past four decades novel medical technologies have emerged that pose challenges for this regulatory scheme. More of these technologies will inevitably continue to emerge in the future. We discuss three examples of emerging technologies— DTC genetic testing, 3D printing, and novel diagnostic and monitoring technologies that are integrated into everyday life—to highlight some of the questions that innovative devices pose for regulation.

5.1  Direct-­to-­Consumer Genetic Testing Increased interest in precision medicine has also brought an increased interest in accurate and useful genetic testing technologies that diagnose disease in symptomatic patients and assess the risks of future illness in asymptomatic patients.77 The FDA currently regulates as devices disease-­related DTC genetic tests, such as 23andMe’s test for three BRCA1/BRCA2 mutations that are associated with a significantly increased risk for breast and ovarian cancer.78 For these tests, which pose virtually no physical risk to 76  See FDA Memorandum, Public Health Interests and First Amendment Considerations Related to Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products (Jan. 2017). 77  See, e.g., Rachel E. Sachs, Innovation Law and Policy: Preserving the Future of Personalized Medicine, 49 U.C. Davis L. Rev. 1881 (2016). 78  US Food & Drug Admin., FDA authorizes, with special controls, direct-­to-­consumer test that reports three mutations in the BRCA breast cancer genes, https://www.fda.gov/NewsEvents/ Newsroom/PressAnnouncements/ucm599560.htm. Genetic tests ordered through physicians generally fall into a category of diagnostic tests that the FDA has opined are devices but for which it has long chosen not to enforce regulatory requirements for various reasons. US Food & Drug Admin., Laboratory Developed Tests, https://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/ InVitroDiagnostics/LaboratoryDevelopedTests/default.htm.

774   patricia j. zettler and erika lietzan c­ onsumers, the FDA interprets “safe and effective” to mean that the tests are analytically valid—that they accurately identify the presence or absence of a genetic variant—and clinically valid—that the tests accurately describe the genetic variant’s connection to disease. For example, accurately identifying the presence or absence of a BRCA gene mutation would be analytic validity, while accurately reporting that the presence of a detected BRCA gene mutation is associated with an increased risk of breast and ovarian cancer would be clinical validity. DTC genetic testing, like many technologies, poses the challenge of how to assure that consumers are appropriately protected while still allowing for fast-­moving scientific innovation. This balancing may be particularly controversial in the context of DTC genetic testing, however, because the product simply provides information. The potential harms associated with DTC genetic testing are therefore informational: that patients will receive incorrect, misleading, or confusing information (or the patients themselves will misunderstand the information), which will lead them to make decisions with adverse health consequences.79 These kinds of harms differ from the kinds of serious or imminent physical risks that often justify FDA oversight. The benefits, if any, would also derive from providing information, but at this time it is not clear that DTC genetic testing leads to improved health outcomes. Finally, there is a strong culture of autonomy in the United States, including a belief of many consumers that they have a right to their own genetic information without interference from the government. Perhaps because of these tensions, in November 2017, the FDA announced an unusual approach to regulating this technology.80 The FDA exempted from pre-­market review certain tests that screen asymptomatic individuals to determine whether they carry genetic variants that could lead to disease in their children (carrier screening tests). It also proposed to exempt from pre-­market review certain tests that identify genetic health risks (GHR) tests in the individual using the test after the manufacturer had already undergone a one-­time review. The agency’s proposal, however, did not apply to all GHR tests. One notable exclusion was tests that provide individuals information about their own cancer risk if that information may lead to treatments associated with morbidity and mortality, suggesting that the agency views different kinds of GHR tests as posing different levels of risk. Nevertheless, with the agency’s March 2018 decision to authorize the marketing of 23andMe’s BRCA test as a class II device (via a de novo petition), the door is now open for substantially equivalent cancer risk tests to be marketed as well. Despite concerns that 23andMe’s BRCA test may falsely reassure consumers who may not understand that it tests only for genetic variants that account for a small percentage of hereditary cancers, on its face, the FDA’s flexible approach to DTC genetic testing, overall, seems a reasonable one—and a sensible way to fit this technology within the existing regulatory scheme. The approach is, however, quite new and it remains to be 79  See, e.g., Nita A. Farahany & Robert C. Green, The FDA Is Overcautious on Consumer Genomics, 505 Nature 286 (2014). 80  FDA announcement, https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ ucm583885.htm.

Regulating Medical Devices in the United States   775 seen what additional experience with DTC genetic testing technologies will reveal about this regulatory scheme.

5.2  Three-­Dimensional Printing Three-­dimensional printing—also known as additive manufacturing—refers to producing a 3D object of a shape based on a digital model.81 It is, essentially, a method of efficiently manufacturing individualized objects. Although the technology for 3D printing has been available for decades, it has recently become more widespread as 3D printers have become significantly less expensive. Indeed, the FDA has now authorized the marketing of dozens of devices that are 3D printed.82 The 3D-­printed devices themselves are not necessarily new. They include, for example, hearing aids, dental crowns, surgical instruments, and plastic braces. Instead, much of the promise of 3D printing comes from the ability to create patient-­specific devices, such as a prosthetic limb designed specifically for one patient. Accordingly, the FDA treats 3D printing as simply an alternate form of manufacturing, for example by providing recommendations on how 3D printers can satisfy manufacturing standards and how 3D-­printed devices can be tested to determine whether they meet quality standards.83 3D printing—particularly as the technology continues to become less expensive—has the potential to enable entities that have not traditionally manufactured devices to do so, including physicians at the point of care, researchers, and any person who happens to purchase a 3D printer.84 This democratization of device manufacturing may have benefits for patients, for example by potentially lowering costs. Yet it also presents significant challenges for the FDA, similar to those presented by microbiome-­based products that are regulated as medicines.85 The FDA’s enforcement authorities are not well-­suited to regulating do-­it-­yourself (DIY) production of medical products.86 If an individual manufacturers a device for herself, that likely would not fall within the FDA’s authority,

81  US Food & Drug Admin., FDA Goes 3-­D, https://blogs.fda.gov/fdavoice/index.php/2013/08/ fda-­goes-­3-­d/ 82  James M. Beck & Matthew D. Jacobson, 3d Printing: What Could Happen to Products Liability When Users (and Everyone Else in Between) Become Manufacturers, 18 Minn. J. L. Sci. & Tech. 143, 179 (2017). 3D printing is not limited to device manufacturing—it can also be used to manufacture medicines and biomaterials, such as organs. See id. 83  US Food & Drug Admin., Guidance: Technical Considerations for Additive Manufactured Medical Devices (2017), https://www.fda.gov/downloads/MedicalDevices/ DeviceRegulationandGuidance/GuidanceDocuments/UCM499809.pdf. 84  See, e.g., Beck & Jacobson, supra note 82. 85  Lietzan and Zettler, “Regulating Medicines in the United States,” in this volume. 86  See, e.g., Patricia J. Zettler, What Lies Ahead for FDA Regulation of DCS Products? 2 J.L. Biosci. 318 (2016).

776   patricia j. zettler and erika lietzan which generally applies only to devices that are distributed in some way.87 If DIY manufacturers are within FDA jurisdiction (e.g., through small-­scale distribution that intersects with interstate commerce), it will likely be practically difficult for the FDA to identify such manufacturers and reasonably assure the safety and effectiveness of 3D printed devices. Moreover, DIY and small-­scale 3D printing of medical technologies raises a range of social and ethical questions about the amount of freedom individuals should have to create and use such technologies without government intervention. If 3D printing of medical technologies spreads far outside the traditional manufacturing supply chain, policymakers will need to confront such questions.

5.3  Integrating Medical Devices into Everyday Life In recent years, promising new ideas have begun to emerge about ways to add technologies that can monitor and diagnose medical conditions to products and spaces that consumers use in their everyday lives. These include “smart” contact lenses that contain sensors to measure glucose levels in patients with diabetes, recording devices with algorithms that can detect early signs of Alzheimer’s disease in everyday behaviors, and steering wheels with sensors to detect anything from signs of fatigue to the onset of a stroke in drivers. These ideas hold enormous potential by promising to painlessly diagnose potentially serious problems without disrupting individuals’ normal activities and, in some cases, in a more timely fashion than is currently possible These technologies also pose numerous social and regulatory questions, including questions about how such information will be stored and used, protections for privacy, and practical challenges for regulating algorithms within the technologies.88 One important issue for the FDA—particularly because these technologies are largely not yet available—will be ensuring that its regulatory structure sufficiently encourages innovation. This may be particularly difficult because of the novelty of the technologies and the amount of investment in research that will be required to bring them to market. Because all of these technologies have emerged recently, they will automatically be categorized as class III devices requiring pre-­market approval. The risks of at least some of these technologies, however, are likely to be moderate—and thus manufacturers might be able to seek de novo classification instead of pre-­market approval. Yet, regardless of the path to market, demonstrating the risks and benefits of such novel technologies is likely to require significant research. In that context, the pre-­market approval process, while slower than de novo classification, may be a desirable path to market because it offers more protection for a manufacturer’s data. Once a technology reaches the market as a class II device through the de novo process, competitors can submit 510(k)s to seek 87  21 USC. § 331. 3D printing may raise some of the same issues as “custom devices” made on the order of a healthcare provider, discussed supra. Although the FDCA addresses “custom devices,” it does not, however, expressly address the full range of possibilities presented by 3D printing, such as DIY production that patients undertake themselves. 88  See, e.g., W. Nicholson Price II, Regulating Black-­Box Medicine, 116 Mich. L. Rev. 421 (2017).

Regulating Medical Devices in the United States   777 authorization to market substantially equivalent devices, essentially opening the door to competition before the innovator manufacturer may have recovered the costs of its research and development.89 These technologies, thus, may bring to the forefront potential weaknesses in the incentive structures built into the overall scheme for device au­thor­i­za­tion, an issue that policymakers will need to confront.

5.4  Concluding Thoughts As with medicines, the FDA uses a regulatory scheme that covers the entire device lifecycle to help ensure the safety and effectiveness of, and the accuracy of information about, medical devices while also encouraging innovation and enabling patient access. The FDA’s approach to DTC genetic testing, 3D printing, and technologies that would integrate medical device functionality into everyday life show some of the strengths and potential weaknesses of this oversight paradigm. For example, the FDA’s current approach to DTC genetic testing may demonstrate the flexibility under current law to find a path forward for novel devices that balances the need for evidence against patients’ and consumers’ desires for access. At the same time, as novel devices (or means of manufacturing and distributing devices) continue to emerge, regulators may confront new challenges, such as questions about the ability of traditional regulatory approaches to manage newer types of risks, the practical difficulties of asserting jurisdiction over DIY devices, and the need to ensure sufficient incentives for beneficial innovation.

89  Although these 510(k)s notifications may be less expensive and burdensome than the first company’s de novo petition, they are conceptually very different from generic drug applications. Among other things, the predicate device was not the subject of a pre-­market approval, and they do not purport to rely on an FDA finding that the predicate device was safe and effective. For a discussion of how generic medicines reach the market under US law, see the chapter by Lietzan and Zettler, “Regulating Medicines in the United States,” in this volume.

chapter 37

R egu l ati ng M edica l Dev ices i n the Eu ropea n U n ion Aurélie Mahalatchimy

The overall goals of the European regulatory framework for medical devices are the same as the goals of the framework for medicines. The framework aims to protect public health by ensuring that medical devices are of good quality and safe for their intended use. The framework also seeks to promote the good functioning of the internal EU market by treating medical devices as goods governed by the principle of free movement within the EU market. Although, as for medicines, this is accomplished through extensive regulation of medical devices throughout their lifecycle, the regulation of medical devices in Europe is very different from the regulation of medicines, especially in two regards. First, unlike medicines, there is no pre-­market authorization by a regulatory authority for medical devices to lawfully enter the EU market. Similar to the requirements for certain devices in the United States, the manufacturer has to ensure the con­form­ity of its medical devices to essential requirements1 for all medical devices, including upon approval of a notified body (a formally designated conformity assessment body) for certain medical devices. Such procedure, commonly called ‘certification procedure’, leads to the affixing of a ‘CE’ mark to the product concerned. It is less burdensome than the marketing authorization applicable to medicines, which is why companies prefer their products to be categorised as medical devices, rather than medicines, so as to reach the European market more quickly. Second, unlike in the United States where the Food and Drug Administration is primary regulator of devices throughout the nation, the European Union does not have a single regulator, such as the European Medicines 1  Essential requirements are a set of criteria for the safety and performance of the devices, including the establishment of a risk management system and justification for the presence of specific substances which are carcinogenic, mutagenic, or toxic to reproduction. They are specified in the European legislation, mainly Annex I of regulation 2017/745.

780   Aurélie Mahalatchimy Agency, of medical devices. Instead several organizations may be involved, and this is mainly a notified body in specific cases. Moreover, medical devices represent a huge legal category of health products that ranges from plasters to cardiovascular catheters. Artificial joints, heart defibrillators and pacemakers, intrauterine devices (IUDs), and ophthalmoscopes are other ­commonly used devices. This category of medical devices is far wider than medicines and has always been considered different from medicines in European regulatory contexts. Unlike in the United States, where the modern statutory scheme for devices was enacted in 1976, EU-­level medical devices legislation was first enacted in the 1990s, with a set of three Directives distinguishing implantable and diagnostic ­medical devices from other types of medical devices (‘the ’90s Directives’).2 The adoption of these directives has been included within the ‘New’, and then, ‘Global approach’ aiming to limit restrictions on the free movements of goods, here medical devices, through technical harmonization at the EU level. These directives have established safety objectives for medical devices without making it necessary for manufacturers to follow specific manufacturing processes, notably in the ­development of the quality assurance system, to attain these objectives. The ­directives also save manufacturers from specific national approval procedures for Member States. More particularly a harmonization of ‘essential requirements’ (that have replaced national requirements because they are ‘essential’) has been provided. Harmonization has been completed by the standardization of technical specifications established by standardization organizations, such as the European Committee for Standardization for medical devices. However, substantial divergences have emerged between the Member States of the European Union and the European Free Trade Association countries regarding the interpretation and application of the rules (such as national differences in product ­classification). Moreover, certain products have raised uncertainties or fallen under regulatory gaps (e.g., manufactured products based on nonviable human tissues or cells; implantable or other invasive products for cosmetic purposes). The Poly Implant Prothese (PIP) scandal definitively highlighted the limits of the directives. The French health authorities found that PIP, a French manufacturer, had used industrial silicone instead of medical-­grade silicone for the manufacture of breast implants, contrary to the approval issued by the notified body and causing harm to 400’000 women around the world due to the selling of 1 Million of defective prothesis between 2001 and 2010. After that, the European regulatory framework for medical devices was revamped, the main objective being to further strengthen patient safety. Consequently, two new European Regulations on medical devices (‘the 2017 Regulations’) were adopted: one on medical devices including implantable devices and

2  Council Directive 90/385/EEC; Directive 98/79/EC; Council Directive 93/42/EEC.

Regulating Medical Devices in the European Union   781 one on in vitro diagnostic medical devices.3 These Regulations entered into force on 26 May 2017 and would have been applicable from 26 May 2020 for the regulation on medical devices and from 26 May 2022 for the regulation of in vitro medical devices although there are many transitional periods. Nevertheless, given the pandemic of COVID-­19, the application of the regulation on medical devices has been postponed by one year, i.e. it will be applicable from 26 May 2021.4 Thus, the ’90s Directives currently apply, and two legislative frameworks (the ’90s Directives and the 2017 Regulations) will be running in parallel from May 2020 until May 2024, and until May 2027 for few provisions. After May 2027, only the 2017 Regulations will apply. This chapter will therefore focus on the 2017 Regulations. Overall, these new medical devices regulations revise the definition of medical devices, clarify the obligations of economic operators (manufacturer, authorized representative, importer, and distributor) in medical device markets, and strengthen the role of private ‘notified bodies’ (see further discussion later) and their surveillance by national authorities. The Regulations also generalise obligations associated with clinical assessment and clinical investigation and improve medical devices vigilance and traceability. First, this chapter explains what constitutes a medical device and how devices are classified according to their level of risk in the European Union. It then discusses how medical devices reach the market, how their risks are managed all along their lifecycle, and what kinds of incentives are provided for innovation and competition. The chapter also analyses the balance between public and private actors in the regulation of medical devices. It then concludes with case studies of innovative medical technologies that have challenged the traditional European regulatory scheme and that have led to many revisions in the 2017 device regulations.

1  Legal Definitions of Medical Devices in the EU The EU Regulations apply to products that are considered to be medical devices and that are to be marketed or used in the European Union. European law generally defines medical devices similarly to the way US law defines devices. As in the US, these include diagnostic instruments such as stethoscopes, blood sugar monitoring kits, or chemicals used by pathologists to facilitate their examination of biopsy tissue. There are many devices

3  Regulation (EU) 2017/745 and Regulation (EU) 2017/746. 4  Regulation (EU) 2020/561.

782   Aurélie Mahalatchimy used in the treatment of disease as well, including implantable therapies such as cardiac stents, insulin pumps, and metal screws and plates for bone fractures.5 While medicines have a pharmacologic, immunologic, or metabolic mode of action, medical devices generally have a mechanical mode of action. In addition, they must be intended for use in some aspect of healthcare, such as prevention, diagnosis, monitoring, or treatment. Since the 2017 Regulations, the European Union’s definition of medical devices now also includes software that provides physiological or pathological information on the body, such as that used in the monitoring of intensive care patients. When a product, after taking into account all its characteristics, may fall within the definition of either a medicine or a medical device and there is doubt as to which definition applies, the legal regime for medicines applies as a stricter regime more protective of patients’ health.6 In this context, the European Commission has adopted guidance on medical devices (known as the MEDDEVs guidance). It includes a manual that specifically includes guidance on the borderlines between medical devices, medicines, biocides, and cosmetics. Despite these general rules, it may be difficult to know whether a product will be regulated as a medical device or a medicine, especially where both types of products are combined. In that context, similar to the United States, the principal intended mode of action is the main criterion for determining how the product is categorized. For instance, the European Commission concluded that for the group of products depending on proanthocyanidins present in cranberry (Vaccinium macrocarpon) to prevent or treat cystitis, the principal mode of action was pharmacological because the constituents of cranberry ‘exhibit most probably a pharmacological activity’ and ‘a mechanical mode of action . . . is highly unlikely’. Consequently, these products are not medical devices.7 The medical devices sector is far more heterogeneous than the medicines sector and includes many innocuous products, such as tongue depressors. That is why, unlike medicines, medical devices are explicitly classified according to their intended use (to be used alone or in combination with another device, to be used only or principally in a specific body part, to be used temporarily, or in short- or long-­term use) as well as to the potential risks with which they are associated (class I for low-­risk medical devices, class 5  More specifically, medical device ‘means any instrument, apparatus, appliance, software, implant, reagent, material or other article intended by the manufacturer to be used, alone or in combination, for human beings for one or more of the following specific medical purposes: — diagnosis, prevention, monitoring, prediction, prognosis, treatment or alleviation of disease, — diagnosis, monitoring, treatment, alleviation of, or compensation for, an injury or disability, — investigation, replacement or modification of the anatomy or of a physiological or pathological process or state, — providing information by means of in vitro examination of specimens derived from the human body, including organ, blood and tissue donations, and which does not achieve its principal intended action by pharmacological, immunological or metabolic means, in or on the human body, but which may be assisted in its function by such means.’ Article 2.1 of Regulation 2017/745. 6  Article 2.1 of directive 2001/83/EC. 7  European Commission, Decision on the qualification of cranberry products published on 3 October 2017.

Regulating Medical Devices in the European Union   783 IIa for middle-­risk, class IIb for high-­risk, and class III for medical devices raising most risks and being regulated more strictly).8 The rules to be applied vary according to such classification.

2  Access to the Market In the European Union, medical devices’ access to the market depends on a certification process that varies according to the type of devices. To this end, the devices must comply with a set of legislative criteria, and the conformity with these criteria must rely on clinical evidence, a requirement that has been particularly strengthened with the 2017 Regulations. Beyond the general rules applicable to most medical devices, two kind of medical devices have been considered distinctly: in vitro diagnostic medical devices have been regulated specifically given their specific features, and specific rules apply to custom-­made devices in comparison with other devices to find the right balance between the need of evidence and access to medical devices.

2.1  Certification: CE Marking In order to be commercialized, medical devices must follow a certification procedure which aims to ensure their compliance with minimum and essential quality and safety criteria and ends with affixing the ‘CE marking’ for all devices, except custom-­made devices and investigational medical devices.9 It is delivered for 5 years either by the manufacturer (autocertification for low-­risk class I medical devices10) or by a ‘notified body’ chosen by the manufacturer (for higher risk classes IIa, IIb, III, and implantable medical devices). As discussed later, notified bodies are private regulatory entities. Such CE mark shows that a medical device complies with a set of minimum requirements called the ‘general safety and performance requirements’. These include demonstrating the attainment of a number of objectives. For example, companies must ensure the safety of patients, users, and third parties; show an acceptable benefit-­risk balance; and reduce risks related to ergonomic features (usability or human factors) and consideration of the use environment. The safety requirements take into account the technical knowledge, experience, education, training, and use environment; the medical and physical conditions of intended users; the generally acknowledged state of the art for safety principles; and the creation of a risk management system. 8  Article 51.1 of Regulation 2017/745. The United States has a similar system for classifying devices by risk level. See the chapter by Zettler and Lietzan, ‘Regulating Medical Devices in the United States’, in this volume. 9  See later discussion. 10  Except for sterile devices, devices with a measuring function, or reusable surgical instruments.

784   Aurélie Mahalatchimy Companies must also ensure the quality of the medical devices (i.e., the conformity of each produced unit to its claimed nature and intended use). To this end, a quality management system has to be established by the manufacturer and will be controlled via the notified bodies’ audits. Companies also have responsibilities regarding maintenance of the characteristics and intended level of performance of devices during their intended use, transport, and storage. For instance, potential temperature and humidity fluctuations should be considered by manufacturers in transportation and storage restrictions on device labelling. Requirements regarding the information supplied with the device are related to the label and instructions for use for all medical devices. The conformity to essential requirements regarding health and safety of patients, users, and third parties is provided for each category of medical devices. The amount and type of evidence to be given by the manufacturer is mainly in accordance with the level of risk (and a fortiori the classification) raised by the medical devices. Unlike class I medical devices (e.g., simple bandages) which have the lowest perceived risks and as such are self-­certified by manufacturers, notified bodies are involved for every other class of medical devices, with a growing role according to the level of risk raised by the devices. The notified body assessing class IIa medical devices (e.g., syringes for pump infusion) must have regard for the quality assurance system and the clinical evaluation of the product. The notified body must go even further for class III medical devices (e.g., stents) as it must also verify their design. The notified body may also consult an expert panel for the clinical evaluation of certain class IIb medical devices (e.g., anaesthesia machines) and implantable class III medical devices.

2.2  Clinical Evaluation and Investigation One of the biggest changes from the previous regimes of the ’90s Directives is the strengthened requirements for clinical evaluation that include both prior and postcommercialization studies. Clinical data for all classes of medical devices must provide sufficient evidence to confirm conformity with the relevant general safety and performance requirements under normal conditions for the intended use of the device. Data also must confirm the evaluation of any undesirable side effects and the acceptability of the benefit-­risk ratio. To that end, manufacturers are required to plan, conduct, and document an assessment using several possible options for clinical data: to evaluate relevant scientific literature currently available, to conduct clinical investigations, or to use data available from a similar medical device. The demonstration of equivalence with another medical device already on the market, which is not a new concept, is similar to the ‘510(k)’ notification in the United States. However, the new EU medical device regulation has provided a narrowed and refined definition of ‘equivalence’, similar to the US concept of ‘substantial equivalence’. Indeed, according to EU law, the device for which equivalency is claimed must share the same technical, biological, and clinical characteristics to show that there is no clinically sig-

Regulating Medical Devices in the European Union   785 nificant difference in the safety and clinical performance of the device. Moreover, sufficient levels of access to the data (technical documentation) of the equivalent device has to be justified. As a practical matter, this will require explicit agreement between the two manufacturers. The conduct of clinical investigations is the most complicated, being long, risky, and expensive, but it is generally required for implantable devices and class III medical devices.11 These investigations must be conducted under specific conditions, including authorization of the Member State(s) in which the clinical investigation is conducted following an application on the EUDAMED database (see later discussion), positive opinion of an ethics committee, a sponsor or its legal representative being established in the European Union, and protection of vulnerable populations. The company also has to show that the anticipated benefits to the subjects or to public health justify the foreseeable risks and inconveniences, obtain informed consent of the participants or their legal representative,12 ensure protection of the fundamental rights of the participants, and conform the investigational device to the applicable general safety and performance requirements. The regulation of clinical investigations for medical devices thus converges with the regulation of clinical trials for medicines. In both cases, the scientific elements of the applications are common to every EU Member State, as provided by the European regulation, and the ethical elements are specific to national laws. Finally, for all class III and certain class IIb medical devices, the manufacturer also may consult an expert panel prior to its clinical evaluation and/or investigation with the aim of reviewing its intended clinical development strategy and proposals for clinical investigation. As far as implantable devices and class III devices (other than custom-­made or investigational devices) are concerned, a summary of safety and clinical performance is required. It will be validated by the notified body as part of the technical documentation.

2.3  In Vitro Diagnostic Medical Devices Like the earlier directives, the 2017 Regulations include specific rules for in vitro diagnostic medical devices (IVD medical devices), such as pregnancy tests or tests for determining the level of cholesterol or glucose in blood. The new definition of IVD devices now clearly includes genetic tests and companion diagnostics.13 Generally, the supervision 11  Clinical investigations are not required for minor modifications to existing devices. 12  In case of emergency situations, such as ‘the urgency of the situation, caused by a sudden life-­threatening or other sudden serious medical condition’, information can be given and informed consent can be obtained after the decision to include the subject in the clinical investigation. Article 68 of Regulation 2017/745. 13  Companion diagnostics are in vitro diagnostic tests supporting the safe and effective use of specific medicines through the identification of suitable patients to be treated by these medicines.

786   Aurélie Mahalatchimy of notified bodies, risk classification, conformity assessment procedures, performance evaluation and performance studies, and vigilance and market surveillance have been significantly reinforced, and provisions ensuring transparency and traceability have been introduced to improve health and safety.14 The directive on IVD medical devices has provided for two limitative lists (high-­risk and moderate-­risk) of IVD medical devices for which the involvement of a notified body is required. By default, all other IVD devices are ‘low-­risk’, except self-­testing ones. The use of these limitative lists implies that newly developed tests not mentioned in these lists, such as a test for Creutzfeldt-­Jakob disease, do not require scrutiny of a notified body, irrespectively of their risks. That is why Regulation 2017/746 renews the classification of IVD medical devices according to their risks and their final use, from class A for low-­risk IVD medical devices (such as specimen receptacles) to class D for higher risk IVD medical devices (such as HIV tests): a rule-­based classification system as for other medical devices. Indeed, the conformity assessment procedure for lowest risk medical devices (here class A) is under the sole responsibility of the manufacturer, while notified bodies are involved for class B, C, and D medical devices. Examples of factors determining the risk of IVD medical devices are the possible consequences of an incorrect test result or severity of the disease or disorder tested. Based on the work of the Global Harmonization Task Force and its follow-­up initiative, the International Medical Devices Regulators Forum, classification criteria that have been used, such as the impact of the diagnostic test result on the individual and on public health, have contributed to the reclassification of certain IVD medical devices, like genetic tests from a lower class to class C. In this context, a Dutch study carried out by the National Institute for Public Health and the Environment on the impact of the new Regulation 2017/746 on the classification of IVD medical devices has shown that many more IVDs will end up in a higher risk category (84% instead of 7%). It implies that the involvement of notified bodies in the commercialization procedure will considerably increase as will the efforts needed from the manufacturers to comply with the con­form­ ity assessment procedures.

2.4  Navigating the Tradeoff Between Wanting Evidence and Wanting Access: Custom-­Made Devices As with medicines, the standard process for securing commercialization is time-­ consuming and expensive, which can significantly delay patient access to new medical devices. As a result, two mechanisms allow patients access to medical devices under specific conditions that are less strict than those provided by the standard process.

14  Recital 4 of Regulation 2017/746.

Regulating Medical Devices in the European Union   787

2.4.1  Custom-­Made Devices A custom-­made medical device is specifically designed solely to meet the individual needs of a particular patient, such as dental prothesis, orthopedic corsets or insoles. This category does not include mass-­produced devices (such as contact lenses) that are adapted to meet the specific needs of patients even though the devices are supplied for the sole use of a particular patient. As long as custom-­made devices are made for the specific needs of an individual patient, the regulatory requirements for manufacturers are lower than those applicable to other medical devices developed to be widely commercialized in the European Union. For instance, the authorization of a notified body is not required for custom-­ made devices. Nevertheless, the manufacturers must ensure that their devices are safe and perform as intended. To place a custom-­made medical device on the market, manufacturers must comply with the essential (i.e., general safety and performance) requirements and draw up a statement that the device conforms to these requirements. They must also establish, document, implement, maintain, keep up to date, and continually improve a quality management system. They must draw up technical documentation, which is a lighter regulatory burden than that required for classical medical devices in that it does not include a clinical evaluation report, a post-­market surveillance plan, or updates to the periodic safety update report. Manufacturers must also perform a post-­market clinical follow-­up and report any problem coming to light through post-­market vigilance. Finally, it should be noted that Member States may not create obstacles to custom-­ made medical devices being made available on the market in accordance with the European medical devices regulations.

2.4.2  Use of Devices in the Interest of Public Health or Patient Safety or Health Unlike in the United States, there is no concept of ‘humanitarian use devices’. Nevertheless, EU law provides derogation from the assessment procedures with the possible authorization from any national competent authority to place on the market or put into service a device in the interest of public health or patient safety or health. For instance, this procedure has been used in France to allow delivery of non–class III partial replacements of already implanted medical devices for joints and teeth. Indeed, it has notably been considered that, in the interest of the patient, a surgical procedure of partial change is more limited and less traumatic than the one necessary for a total replacement. Such derogative authorization has to be communicated to the European Commission and other Member States when granted for use other than for a single patient. Moreover, in exceptional cases relating to public health or patient safety or health, the European Commission can extend this authorization to the entire EU for a limited period of time.

788   Aurélie Mahalatchimy

3  The Balance Between Public and Private Actors Both public and private actors are involved in the regulatory process applicable to ­medical devices in the European Union. The 2017 Regulations have established a ‘pyramidal system’ with a base constituted of private bodies including economic actors (e.g., manufacturers, importers, and distributors) with wide and strengthened obligations, especially for manufacturers (3.1), on top of which sits the position of notified bodies as the main private regulators (3.2), and above that public oversight, mainly from national authorities in Member States, with the European Commission responsible for coordination (3.3). Consequently, most regulation is done and implemented by private actors that form the large base of the pyramid (economic actors and notified bodies, both being private companies), and public actors only do a small portion of regulation, in ac­cord­ance with the obligations imposed by laws.

3.1  The Obligations of Economic Operators: The Base of the Pyramidal System The 2017 Regulations reinforce the basic legal framework of medical devices, notably through imposing stricter requirements on private companies in the device market. The 2017 Regulations clearly set out who holds those obligations, which range from the manufacturers as the actors having most obligations, to the authorized representatives they have designated, and to importers and distributors. Such private actors are at the base of the pyramidal system.

3.1.1  Manufacturers’ Responsibilities Manufacturers are the main operators in the supply chain for medical devices. They are responsible for bringing into the market medical devices that are fully compliant with legal requirements. It includes ensuring enforcement of the regulations regarding design, manufacture, and technical documentation. For instance, the procedures and techniques for monitoring, verifying, validating, and controlling the design of devices must be adequately described in the documentation of the quality management system. Manufacturers are in charge of the analysis of risks linked to their medical devices and of providing the relevant dossier in compliance with the essential requirements set out in the Regulations. In accordance with these responsibilities, they will decide on the commercialization of their devices, upon authorization of a notified body for other devices than class I medical devices. Manufacturers also append the CE mark after obtaining the conformity certificate (including autocertification for class I medical devices) and informing their national competent authority.

Regulating Medical Devices in the European Union   789 Manufacturers must conclude agreements with other private actors where relevant. In that context, it should be underlined that the definition of ‘manufacturer’ in the 2017 Regulations clarifies the situation regarding the so-­called own brand labeller (OBL). The OBL is a company that buys a finished medical device with CE marking from an original manufacturer to sell it under its own brand name while not necessarily being the entity that actually designs, manufactures, packages, or labels the device. This practice consequently raises the question: Who is legally responsible for enforcing the manufacturer’s obligations? Indeed, it appears that some Member States found, when inspecting OBL sites, that the OBL did not have the required technical files because they had been kept secret by the original manufacturer. While the ’90s Directives were not very clear about the responsibility of an OBL, the European Commission recommended twice, in 2008 and 2013, that an OBL should have the same responsibilities and obligations as the original manufacturer. This clarification is clearly confirmed by the 2017 Regulations in their definition of the ‘manufacturer’. Under the 2017 Regulations, manufacturers are obliged to conduct a clinical evaluation for every medical device in accordance with strengthened requirements, whatever its classification. Previously, medical device manufacturers had frequently used the ‘equivalence procedure’, which in effect established clinical safety through references to the existing scientific literature. When the Regulations enter into force, this procedure will become exceptional, especially for implantable devices and class III medical devices, which will generally require clinical investigations. As is the case for medicines, manufacturers of all classes of medical devices must have a designated person in their organization (or have such a person permanently and continuously at their disposal for micro and small enterprises) who is responsible for regulatory compliance. That person’s qualifications must be demonstrated by diploma or professional experience. The designated person must also ensure compliance with the ‘Unique Device Identifier’ assignments (see later discussion). Manufacturers are the main operators liable for defective medical devices that cause damage.15 Where a manufacturer is not established in the European Union, they must designate an authorized representative within the Union.

3.1.2  Authorized representatives, importers and distributors Regulation 2017/745 provides a list of mandatory obligations for the authorized representative (such as the verification of the drawing up of the EU declaration of conformity and technical documentation) which must, at least, be included in the mandate of designation by the extra-­EU manufacturer. Two elements of the new obligations of the authorized representative can been highlighted. First, the authorized representative must terminate its representation if the manufacturer does not comply with its obligations. Such obligation goes beyond the classical mandate rules, where the mandatory controls the authorized representative. Second, ‘the authorized representative shall be legally liable for defective devices on the same basis as, and jointly and severally with, the 15  Council Directive 85/374/EEC.

790   Aurélie Mahalatchimy manufacturer.’ Such a rule, although justifiable on safety grounds, appears to be contrary to the classical regime applicable for the liability of defective products, where the authorized representative is only obliged to complete operative and administrative tasks on behalf of the manufacturer, inasmuch as the manufacturer is the only economic operator able to modify a device to ensure its conformity with essential requirements under the Regulations. These two obligations that go beyond the classical mandate rules show the specificity of an international situation (extra-­EU manufacturer and EU-­authorized representative) and the primary position that is given to safety considerations in the EU legislation on medical devices. Finally, unlike the ’90s Directives, the 2017 Regulations explicitly provide obligations for importers and distributors. Generally, before placing medical devices on the market, importers must ensure that the manufacturer and the medical devices comply with the requirements of the medical devices Regulations. Moreover, they must display the details of this compliance on the medical device or on its packaging, and must register in  the EUDAMED database (see later discussion). The new EU regulation defines distributors,16 mentions the activities of distributors,17 and includes them among the economic operators in the medical devices supply chain. Distributors consequently also have specific obligations, notably regarding the verification of manufacturers’ and importers’ compliance with legal requirements, the supply of medical devices to end-­ users or third parties, and their involvement pre-­marketing (compliance with essential requirements) and post-­ marketing (vigilance), even though there is no Good Distribution Practice for medical devices. Although distributors are not covered by the obligation of registration on EUDAMED, Member States are permitted to maintain or introduce obligations on the registration of distributors of medical devices made available on their territory.

3.2  The Main Role of Private Regulators: European ‘Notified Bodies’ It is a peculiarity of the European system that it relies on private regulators to decide whether medical devices may enter the European market. Manufacturers decide on entry for low-­risk class I medical devices. ‘Notified bodies’ decide for class IIa, class IIb, and class III devices. ‘Notified bodies’ are private conformity assessment bodies established where the safety of various products is regulated through European technical harmonization, such as the safety of toys, radios, or pressure equipment. The approach has its roots in what is still called the ‘New Approach’ to standardization of product safety, adopted to facilitate 16  ‘ “[D]istributor” means any natural or legal person in the supply chain, other than the manufacturer or the importer, that makes a device available on the market, up until the point of putting into service.’ Article 2.34 of Regulation 2017/745. 17  ‘For the purpose of this Regulation, the activities of distributors should be deemed to include acquisition, holding, and supplying of devices.’ Recital (28) of Regulation 2017/745.

Regulating Medical Devices in the European Union   791 the free movement of products across national borders within the European Union in order to create the European Union’s ‘internal market’. The approach operates as follows. Minimum regulatory standards, agreed by EU legislation, apply across the whole Union. Manufacturers must apply these standards to all products, including medical devices. Notified bodies are responsible for assessing and certifying the conformity of the products (including medical devices) to those standards as a matter of public interest. Each Member State must mutually recognise the conformity assessment of notified bodies in other Member States and must permit the certified products to be marketed as safe within its territory. Notified bodies are designated and monitored by national competent authorities, such as the French National Agency for Medicines and Health Products (ANSM) for medical devices. Notified bodies have been used by the European Union’s medical devices legislation from its inception. Under the ’90s Directives, there were about 80 notified bodies across Europe. The manufacturer can choose the notified body of the country in which it is established or any other notified body recognised anywhere in the European Union. This arrangement led to a kind of regulatory competition between notified bodies, as does state regulation in the United States. It was alleged that the resultant varying quality would potentially lead to an uneven level of patient protection and a possible distortion of competition between manufacturers of similar products. For this reason, Member States’ control over notified bodies has been strengthened in the 2017 Regulations pertaining to medical devices.

3.3  The Oversight Role of Public Regulators: The Balance Between European and National Contrary to the European Union’s regulation of medicines, where the European Medicines Agency (EMA) coordinates regulation across the Union, the oversight role of public regulators of medical devices comes mainly from Member States and their national competent authorities. Indeed, because there is no equivalent to the EMA, there is only a limited coordination role for the European Commission. As mentioned earlier, the national competent authorities have control over the designation and monitoring of notified bodies and must only inform the European Commission. National competent authorities also authorize clinical trials of devices, especially regarding the newly required clinical investigations for implantable devices and class III medical devices for which a positive opinion from an ethics committee is needed. However, national competent authorities do not intervene in the awarding of the CE mark. National competent authorities also collect data forwarded by the manufacturers, especially regarding their vigilance notifications, and are in charge of market surveillance. Member States may also authorize derogation from the conformity assessment procedure, on a duly justified request, where the classical procedures have not been carried out but the use of the medical device is in the interest of public health or patient safety or health, as noted earlier.

792   Aurélie Mahalatchimy Member States are in charge of implementing and controlling the proper implementation of EU rules. Moreover, key aspects of medical devices regulation, such as the prescription of medical devices and the management of national healthcare systems, including prices and reimbursement of medical devices, remain at national discretion. There is much less involvement of EU-­level regulators than in the case of medicines regulation. The European Commission plays a coordination role in the classification of medical devices and the management of the European database for medical devices: EUDAMED. Regarding classification, two groups are in charge of advising the European Commission. First, a medical devices expert group including representatives of the Member States competent authorities, European trade federations and professional organizations, and notified bodies is in charge of advising the European Commission on borderlines and classification of medical devices. Second, a medical devices coordination group, composed of persons designated by the Member States based on their role and expertise in the field of medical devices and IVD medical devices, has been established by the 2017 Regulations to provide advice to the European Commission and to assist it (as well as the Member States) in ensuring a harmonized implementation of the medical devices regulations. This group notably ensures coordination and cooperation between notified bodies. As this group is also competent on borderline and classification of medical devices, its links with the older medical devices expert group are not clear, and the future of the latter may be in question. Furthermore, although medical devices are generally not covered by the EMA, the EMA or national competent authorities must be consulted regarding combination products. First, the European Commission can consult the EMA regarding borderline products involving medicines, as is also the case for the European Chemicals Agency and the European Food Safety Authority regarding biocidal products or food products. Second, the EMA or national competent authorities must be consulted by a notified body on medical devices composed of substances or combinations of substances that are absorbed by or locally dispersed in the human body. Third, they must also be consulted by a notified body on companion diagnostics which are newly defined and explicitly considered by the 2017 Regulation on IVD medical devices. Fourth, they must also be consulted by a notified body on medical devices incorporating an ancillary medicine (e.g., substances derived from human blood or human plasma, or from biotechnology products). Fifth, the EMA or national competent authorities are in charge of delivering marketing authorization dossiers for medicines with integral device component, such as prefilled syringes. In that case, the dossier must contain the EU manufacturer’s declaration of conformity or the CE certificate issued by a notified body.

4  Risk Management Medical devices are health products. As such they raise risks for patients’ safety, but also for other persons such as third parties or health professionals who manipulate these

Regulating Medical Devices in the European Union   793 devices. That is why safety and the management of risks is a priority in the European Union’s medical devices Regulations, with the broad objective of protecting public health. Although all medical devices are associated with some risk, it is necessary to reduce the risks linked to the use of medical devices and to accept the residual risks only after a benefit-­risk assessment, thus the EU legislation on medical devices requires the management of risks all along the therapeutic chain. The ‘pyramidal system’ of responsibility established by the 2017 Regulations deals with risk management by setting wide and strengthened obligations for manufacturers who must ensure traceability all along the lifecycle of a medical device, some freedom to the national competent authorities regarding the management of vigilance on their territories, and a new coordinating role for the European Commission. Indeed, the latter has both an arbitral role where there is disagreement between Member States and a control role over the national competent authorities regarding preventive measures adopted in case of potential risk to health. With a new system that is more complete, especially regarding traceability and vigilance, and clearly strengthened requirements on risk management, the 2017 Regulations should better protect public health. Nevertheless, the new system has proved difficult to set up, and already there have been delays in the implementation of some parts of the new regulations, especially the full operationality of the EUDAMED database.

4.1  Managing Risk from the Time of Market Entry The management of risk is required for medical devices to enter the European market. Much of this has been covered previously in the discussion of market access. Once a device is cleared for access, risk can be managed on a case-­by-­case basis, as with medicines, by requiring a healthcare professional’s prescription. There are no specific EU-­ level rules on the prescription of medical devices, but national rules can exist on this matter and EU law does not affect them.18 Beyond the prescription of medical devices, there will inevitably be manufacturing defects or other problems with some devices, so it is important to be able to trace a patient’s device back to its manufacturer, distributor, or other participants in the distribution chain. It also is important to be able to identify other devices that may have the same problem. Accordingly, the 2017 Regulations on medical devices enhance transparency to provide a better management of risks and strengthen the confidence of medical devices’ users. To this end, two main tools have been developed: the traceability proc­esses and the EUDAMED database on medical devices. Both contribute to the management of risks all along the development chain of medical devices. For a long time, traceability has been required, from the description of the raw material as part of the quality management system to details about the manufacturing proc­ ess. But the new regulation goes further in expanding traceability to the identification of 18  Article 1.15 of Regulation 2017/745.

794   Aurélie Mahalatchimy and cooperation among all different economic operators. Moreover, users are also targeted by traceability requirements. For instance, implantable class III medical devices, as well as devices to be listed by the European Commission, have to be registered and traced by the health establishments supplying them. This is also recommended for other types of devices. Other examples for implantable medical devices are the implant cards delivered with the devices to patients. Moreover, the 2017 Regulations established a unique device identification (UDI) system to contribute to the efficiency of traceability by allowing better identification of medical devices and reducing the risks linked to errors in device choice and selection. This system includes two parts: a UDI device identifier (UDI-­DI) specific to a manufacturer and a device and a UDI production identifier (UDI-­PI) that identifies the unit of device production and, if applicable, the packaged devices and that relies on an internationally recognised nomenclature. The UDI-­DI is given by the manufacturer, before commercialization, and communicated to the European Commission with other information such as the name and address of the manufacturer, the medical device nomenclature code, and risk class of the device. Thus, the UDI system contributes to the traceability and control over medical devices all along the supply chain. As such, it also participates in the fight against falsified devices that have been targeted within the new EU Regulation 2017/745. The Regulation defines ‘“falsified device” as any device with a false presentation of its identity and/or of its source and/or its CE marking certificates or documents relating to CE marking procedures’. This definition is similar to that for falsified medicines. But, unlike medicines, the obligations regarding falsified devices are not primarily on manufacturers. Rather, importers and distributors must inform the competent authority of their Member State when they consider or have reason to believe that a device presents a serious risk or is a falsified device. It is then part of the market surveillance activities of the competent authorities to confiscate, destroy, or otherwise render inoperable falsified devices. Finally, EUDAMED, in which all medical devices have to be registered by their manufacturer before commercialization, also participates in the efficiency of traceability and in the better management of risks. In addition to the name, label, and instructions for use, EUDAMED will provide data about devices placed on the market, potential clinical investigations, certificates issued by notified bodies where ­relevant, and information about the relevant economic operators. More specifically, EUDAMED includes an electronic system related to clinical investigations permitting Member States, the European Commission, and notified bodies to exchange information. Indeed, the sponsor can submit a single application for several Member States and report serious events, device deficiencies, and related updates via EUDAMED. Consequently, this system reduces the delay for information exchange and accelerates the decision on investigation and corrective measures where necessary. EUDAMED also includes an electronic system related to vigilance and surveillance after commercialization.

Regulating Medical Devices in the European Union   795

4.2  Monitoring and Managing the Risks of Marketed Devices As is also the case for medicines, EU law requires post-­market monitoring of the risks associated with medical devices. First, vigilance implies the reporting of serious incidents19 and of corrective actions taken by the manufacturer. Manufacturers must report serious incidents and corrective actions to the national competent authorities of the Member States for all medical devices on the EU market. The competent authorities of the Member States assess these notifications in collaboration with the manufacturer and the notified body where relevant. Vigilance also applies broadly beyond the reporting of serious incidents. It is part of a larger process of post-­market surveillance and market surveillance that aims to prevent the risk of incidents. Vigilance involves every actor: the manufacturers, but also Member States and their national competent authorities, the notified bodies, and the European Commission. Manufacturers and Member States both have surveillance responsibilities. For manufacturers, the 2017 Regulations require a post-­market surveillance plan to be established for class I medical devices as well as a post-­market surveillance report to be updated when necessary. For the other classes of medical devices (class IIa, class IIb, and class III devices) which raise more risks, a Periodic Safety Update Report (PSUR) must be prepared and updated annually in addition to the surveillance plan. The PSUR analyzes data gathered through post-­market surveillance plans and discusses any preventive or corrective actions taken. These post-­market surveillance data are gathered all along the medical device lifecycle and used to update the benefit-­risk balance and the materials on the medical device, such as those relating to technical documentation, risk assessment, and clinical evaluation. Because the data gathered are used to update the relevant part of technical documentation, the process contributes to transparency. The national competent authorities of the Member States are in charge of market surveillance. To this end, they can organise expected or unexpected inspections of manufacturers, distributors, hospitals, and health professionals marketing or using the medical devices. If a device presents ‘an unacceptable risk to the health or safety of patients, users or other persons, or to other aspects of the protection of public health’ (these notions being freely interpreted by the Member States in the absence of a European definition) or does not comply with the legislative requirements, the national competent authorities of the Member States can recall a device, restrict its use, or require other corrective action to bring the device into compliance with the 2017 Regulation. The national competent authorities are also obliged to report the situation to the

19  It should be highlighted that Regulation 2017/745 distinguishes serious incidents and serious events. The former is linked to the medical device without a necessary impact on health (Article 2.64 of Regulation 2017/745), while the latter occurs during a clinical investigation (see earlier discussion) and is not necessarily linked to the studied medical devices (Article 2.57 of Regulation 2017/745). Both can be ‘serious’ in specific situations.

796   Aurélie Mahalatchimy European Commission and the other Member States where a certificate has been issued for the device concerned. Furthermore, Regulation 2017/745 on medical devices distinguishes between ‘recall’ (‘any measure aimed at achieving the return of a device that has already been made available to the end user’) and ‘withdrawal’ (‘any measure aimed at preventing a device in the supply chain from being further made available on the market’). All market participants—manufacturers, importers, and distributors—have the obligation to withdraw or recall a device when they consider or have reason to believe that a device they have made available on the market is not in conformity with Regulation 2017/745. For instance, in December 2018, Zavation Medical Products Pan Medical US recalled of all its medical devices marketed in Europe from August 2016 because of falsified end-­date certificates issued by the society it acquired, Pan Medical US.20 Member States and their competent authorities must also withdraw or recall a device for the same reasons, thus fulfilling their duty to protect health.

4.3  Unique Challenges Presented by Cybersecurity Cybersecurity raises unique challenges, including for networked medical devices. Some of these challenges are related to intrinsic safety problems, such as with pacemakers or implantable cardiac defibrillators or insulin pumps, which, in the event of a data compromise, could result in harm to a patient. But cybersecurity also poses challenges regarding the confidentiality of personal data stored or transmitted by medical devices. There is a need for trust in the use of electronic and connected medical devices. Somewhat unlike in the United States, where the FDA has begun to collaborate with other government agencies and provide recommendations to manufacturers to address cybersecurity risks, EU regulation thus far has not focused specifically on cybersecurity. Nevertheless, the EU view seems to be changing, as illustrated by three complementary domains. First, within EU Regulation 2017/745 on medical devices, cybersecurity of devices is intrinsically considered through the new inclusion of software within the definition of medical devices: for instance, insulin dosage-­planning stand-­alone software (class IIb), software for the presentation of the heart rate or other physiological parameters for intensive care monitoring (class IIb), and software that uses an algorithm to characterize viral resistances to various drugs. More specifically, some safety and performance requirements of this Regulation relate to cybersecurity, notably because devices that incorporate electronic programmable systems or software are specifically targeted. For instance, their design or manufacture must remove or reduce the risks associated with 20  French ANSM, Recall of medical devices from Pan Medical US corporation, 28/12/2018, https:// ansm.sante.fr/S-­informer/Informations-­de-­securite-­Retraits-­de-­lots-­et-­de-­produits/ Retrait-­du-­marche-­de-­tous-­les-­dispositifs-­medicaux-­Pan-­Medical-­US-­Corporation-­distribues-­depuis-­ aout-­2016-­Rappel.

Regulating Medical Devices in the European Union   797 the IT environment or take into account the specific features of the mobile platform (e.g., size and contrast ratio of the screen). These new rules within the EU Regulation on medical devices show that EU law is taking into account in specific and detailed disposals the challenges raised by electronic and networked medical devices regarding the information stored or transmitted by the devices. Second, EU Regulation 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (the General Data Processing Regulation [GDPR]) completes this legal frame regarding the personal data stored or transmitted by medical devices.21 Third, the challenges raised by medical device cybersecurity could be considered in the European Union’s general cybersecurity policy currently under development. As part of the EU cybersecurity strategy adopted in 2013, a second mandate has been given to the European Union Agency for Network and Information Security (ENISA). A Directive on security of network and information systems, one promoting a culture of risk management, was adopted in 2016. The latter introduces security requirements as legal obligations for key economic actors, especially operators providing essential services and suppliers of key digital services. Nevertheless, neither of these two texts mentions medical devices. In contrast, the European regulation on Cybersecurity explicitly mentions medical devices, although in one recital only.22 It provides uniform requirements for cybersecurity via a certification scheme for specific IT processes, products, and services and to transform ENISA into a permanent agency for cybersecurity within the European Union. It remains to be seen how this regulation will complement the medical devices Regulation 2017/745 and GDPR regarding cybersecurity in the field of medical devices. In general, the EU approach appears more segmented than the US approach to this topic.

5  Innovation and Competition The EU Directives on medical devices and the 2017 EU Regulations on medical devices constitute in themselves, as supportive legislations, the main incentives for competition of medical devices in order to create a single European market for medical devices. More targeted incentives are missing. For instance, as in the United States, there is no ‘generic’ device concept as it exists for medicines. And, unlike in the United States, there is no period of exclusive marketing regarding data exclusivity for medical devices. Consequently, in the European Union, incentives for innovation in the medical devices field are limited to patent protection. As for competition, the absence of generics

21  See the chapter by Hoffman and Herveg, ‘Privacy and Integrity of Medical Information’, in this volume. 22  Regulation (EU) 2019/881.

798   Aurélie Mahalatchimy makes it essentially brand-­to-­brand, and advertising is only partially regulated at the European level, leaving EU Member States discretion to regulate at the national level.

5.1  Incentives for Innovation: Patent Protection Patents for medical devices rely on the same legal landscape as the one applicable for medicines.23 Device manufacturers generally enjoy a 20-­year period of patent protection. Also, as with medicines, while the devices themselves can be patented, patents are not available for the methods by which the devices are used. Thus, for example, a company could patent a new scalpel for making surgical incisions but not the method of making a horizontal rather than vertical incision.24 Moreover, there is no Supplementary Protection Certificate for medical devices. The Court of Justice of the European Union even stated that a supplementary protection certificate could not be granted for a substance which was an integral and active part of a patented medical device.25

5.2 Competition As mentioned, all competition is brand-­to-­brand for medical devices because there is no ‘generic’ device concept. Due to the decentralized system based on several competent authorities (notably the notified bodies), the European legal framework has been considered as an incentive by itself. Indeed, as said before, the choice of a notified body by the manufacturer anywhere in the European Union has led to a kind of regulatory competition between notified bodies. The European Union also had the lowest approval times in the world among leading markets for high-­risk devices.26 Nevertheless, the strengthened requirements for notified bodies as well as for the medical devices as such regarding pre-­market and post-­market requirements will probably have an impact on European competitiveness. Beyond the early uncertainty due to the implementation of the new legal frameworks which could lead to more burdens for 23  See the chapter by Mahalatchimy, ‘Regulating Medicines in the European Union’, in this volume. 24  Article 53(c) of the European Patent Convention and Recital 35 of Directive 98/44/EC. Moreover, inventions the commercial exploitation of which would be contrary to ‘ordre public’ or morality are also excluded. Id. Art. 53(a). European patent law provides a nonexhaustive list of these inventions: processes for cloning human beings, processes for modifying the germ line genetic identity of human beings, and uses of human embryos for industrial or commercial purposes, processes for modifying the genetic identity of animals which are likely to cause them suffering without any substantial medical benefit to man or animal, and also animals resulting from such processes. Rule 28 of the implementing regulations to the European Patent Convention and Article 6(2) of Directive 98/44/EC. 25  CJEU, 25 October 2018, Boston Scientific Ltd, C-­527/17, ECLI:EU:C:2018:867. 26  Emergo Group, ‘Compare the Time, Cost, and Complexity of Getting Regulatory Approval’, December 2017, https://www.emergobyul.com/resources/worldwide/global-­regulatory-­comparison­tool?field_market_tid=All&cost=All&field_device_risk_value=3.

Regulating Medical Devices in the European Union   799 companies (lack of notified bodies, additional cost for new regulatory compliance), the new frameworks aim to ensure the smooth functioning of the internal market as well as a high level of protection for public health. Although only time will tell us whether the competitiveness of the medical devices sector in Europe has been maintained and how much this is the result of the new legal frameworks, two important comments can be made regarding a positive impact of the European regulation on competition. First, the strengthened requirements, that must lead to safer medical devices as well as enhanced transparency, should encourage more trust in this sector in Europe. Second, EUDAMED will make exchanges of information easier and quicker. Consequently, the European Union may be more competitive compared to other countries regarding clinical investigations and post-­market vigilance and surveillance.

5.3 Advertising The new EU Regulation 2017/745 covers claims in labeling, instructions for use, and making available, putting into service, and advertising of devices. Only CE-­marked devices can be placed on the EU market and promoted. Moreover, it is prohibited to make explicitly false or misleading statements. It is also forbidden to fail to inform of the risks associated with the use of the device and to make off-­label promotion in which the user or patient would be misled with regard to the device’s intended purpose, safety, and performance. However, the Regulation does permit manufacturers to exchange scientific information (without advertising the device as safe and effective) before it is CE-­marked. In addition to these medical device-­specific EU provisions, other EU provisions on advertising and promotion may be applicable to medical devices, especially Directive 2006/114/EC on misleading and comparative advertising and Directive 2005/29/EC on unfair business-­to-­consumer commercial practices in the internal market. Apart from these EU provisions, the advertisement of medical devices is regulated at the national level. For instance, in Italy, the advertising of nonprescription medical devices requires the prior authorization of the Ministry of Health.27 As another example, in France, the prior authorization of the ANSM is required for medical devices and IVD medical devices that are listed in specific regulations,28 and French law distinguishes between advertising to the general public and to healthcare professionals.29 27  Article 21 of the Legislative Decree no. 46 of 24 February 1997 (the Italian law on medical devices). To go further on this Italian rule, see Italian Supreme Court Decision no. 10892, 7 May 2018. 28  Order of 24 September 2012 establishing the list of medical devices with high risk for human health, and for which advertising requires a previous authorization in accordance with Article L.5213-­4 of the French Public Health Code; Order of 24 September 2012 establishing the list of in vitro medical devices for which advertising requires a previous authorization in accordance with Article L.5223-­3 of the French Public Health Code. 29  Decree 2012–743 of 9 May 2012 regarding advertising for medical devices.

800   Aurélie Mahalatchimy Moreover, two databases, mentioned earlier, aim to spread information to the public: the UDI Database, which includes core elements for the identification of marketing devices, and the EUDAMED database, still under construction. Note, however, that the EUDAMED database already existed under the medical devices Directives’ central repository for exchanges of information on market surveillance between national competent authorities and the European Commission only. The 2017 Regulations provide for a much larger EUDAMED database that includes different modules on actors, UDI and devices, notified bodies and certificates, vigilance, clinical investigations and performance studies, and market surveillance. The implementation of the new EUDAMED database was initially scheduled for March 2020, but the European Commission has recently delayed it to May 2022 in order to achieve full functionality of the different modules and to submit them to an independent audit before EUDAMED is fully operational. Such delay may be seen as an opportunity for the industry to adapt itself to the new requirements. Indeed, the EUDAMED database is challenging for companies in terms of time and resources, especially IT resources.

6  Examples of Challenging Products Covered by the New EU Regulations Emerging technologies, in particular genetic testing, 3D printing, and e-­health (discussed later), may challenge the European Union’s regulatory scheme, especially as the issues they raise are not entirely understood. However, the 2017 EU Regulations on medical devices and IVD medical devices have taken into account all these technologies as far as they are medical devices. It is difficult to know whether the new disposals are fully operational and adapted to these emerging technologies while their implementation is still ongoing. Nevertheless, one common particularity of the EU scheme applicable to these emerging technologies is that it is fragmented. Beyond the sharing of requirements between national legal frameworks and the European legal framework (which is an intrinsic difficulty of EU law), several European legal texts may apply to these technologies. Indeed, most of the time, their legal frame is not limited to the EU legislation on medical devices because it also includes EU rules on unfair commercial practices, protection of personal data, patients’ rights in cross-­border healthcare, and general product safety. Although the application of each of these sets of rules may raise concern, their combination, complementarity, and sometimes overlapping areas will also raise issues which will need to be measured once the necessary delay in the full implementation of the 2017 Regulations on medical devices and IVD medical devices is over.

6.1  Direct-­to-­Consumer Genetic Testing Direct-­to-­consumer (DTC) genetic tests for medical purpose are considered to be in vitro diagnostic medical devices, and, as such, the EU Regulation 2017/746 regulates

Regulating Medical Devices in the European Union   801 them. Interestingly, during the process of adoption of this text, the European Parliament called for a ban on DTC advertising of genetic testing, as the previous IVD medical devices Directive did not regulate this aspect of genetic testing. The final text, however, did not include such a prohibition and instead only prohibits misleading claims for medical devices, including in DTC advertising of genetic testing. Therefore, it does not deviate much from the existing generally applicable Directive 2005/29/EC on unfair commercial practices against misleading advertising. Moreover, EU Regulation 2017/746 provides rules on genetic tests in general (i.e., for physician-­ordered tests) without focusing on DTC genetic tests. It requires that the individual being tested is provided with relevant information on the nature, significance, and implications of the genetic test, and, in particular, an ‘appropriate access to counselling’ is required.30 Nevertheless, Member States can adopt or maintain measures that are more protective of patients, as is the case in France where a genetics counselor must provide the counseling, and, consequently, DTC genetic tests are prohibited.31 Consequently, apart from advertising and general rules on genetic tests, DTC genetic testing remains regulated at the national level only, involving a fragmented regulatory landscape in Europe, with some countries essentially banning it (France, Germany) while others only restrict it on the basis of general laws regarding patients’ rights and healthcare services (Luxembourg, Poland).

6.2  3D Printing Three-­dimensional printing, or ‘additional manufacturing’, is transforming how products are designed, developed, manufactured, and distributed, including in the medical technology sectors. It refers to ‘the various processes used in the manufacture of products, by depositing or fusing materials layer by layer’32 to obtain 3D objects. Within the health field, 3D printing has been used or is being developed for the rapid and cost-­ effective production of implants and prosthetics, such as dental implants and prosthetic limbs and for surgical planning and tools, and also 3D bio-­printing, such as the printing of cells, tissues, and organs, although this latter is in its infancy. Three dimensional printing raises important legal challenges linked to the standardization process, legal definitions, and intellectual property (patents, copyrights, and trademarks). At the moment, the EU legal regime regarding 3D printing is unclear overall in all of those domains. Indeed, there is no sui generis regulatory regime governing the whole printing process. Where 3D printing processes or printed objects correspond to the legal definition of medical devices, the legal regime of medical devices will apply. 30  However, it does not ‘apply in cases where a diagnosis of a medical condition and/or a disease which the individual being tested is already known to have is confirmed by a genetic test or in cases where a companion diagnostic is used’. Article 4.2 and 4.3 of Regulation 2017/746. 31  Article 1132–1 of the French Public Health Code. 32  European Commission, The disruptive nature of 3D printing, January 2017: https://ec.europa.eu/ growth/tools-­databases/dem/monitor/sites/default/files/DTM_The%20disruptive%20nature%20of%20 3D%20printing%20v1.pdf. Accessed 15 February 2019.

802   Aurélie Mahalatchimy However, some will be considered as standard medical devices (such as a mass-­ produced pillar for dental implants), while others may be custom-­made medical devices (such as the implant designed for a particular patient and to be adapted on the pillar). For 3D bioprinting, piecemeal legislation is relevant in relation to tissue engineering and regenerative medicine, all being applicable at different stages of production.33

6.3  E-­health: Novel Diagnostic and Monitoring Technologies Many information and communication technology-­based tools are being developed at an exponential rate to assist with prevention, diagnosis, treatment, health monitoring, and lifestyle management. These innovative tools include electronic health records, telemedicine services, personal wearable devices, and portable communicable systems. The European Commission defines e-­health as ‘the application of information and communications technologies across the whole range of functions that affect the health sector’.34 Using its public health competency, the European Union has established a European strategy in the field of e-­health with the aim of favoring the development of a European e-­health area. But the Commission has also adopted more concrete measures, using its internal market competency, to set up e-­networks to improve the interoperability of health systems in the European Union.35 These are very much in their infancy as each EU Member State retains responsibility for its own health system. Besides e-­health services, many e-­health products should be qualified as medical devices and, as such, are regulated by EU Regulation 2017/745 which includes software. When they are not, they are regulated by the EU Directive on general product safety.36 The particularity of e-­medical devices is that the personal data they include or transmit are health data governed by the GDPR. These health data are considered sensitive, and, as such, their processing is in principle prohibited although exceptions exist. However, the definition of data concerning health as provided in the GDPR does not include well-­being data, which are difficult to define. Consequently, many e-­objects used for well-­being data, such as a connected watch/bracelet counting the number of daily steps, will not be governed by the prohibition on processing health data, although the e-­objects may be covered by other generic provisions of the GDPR.37 The line between health data and well-­being data appears very unclear. Consequently, the data 33  Li P and Faulkner A, ‘3D Bioprinting Regulations: A UK/EU Perspective’ (2017) Eur J Risk Regulation 8(2), 441–447. ISSN 1867-­299X. 34  Communication from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions: e-­Health—making healthcare better for European citizens: An action plan for a European e-­Health Area, 30 April 2004, COM(2004)356. 35  Directive 2011/24/EU. 36  Directive 2001/95/EC. 37  Rial-­Sebbag E, ‘Les produits de santé connectés: quelle sécurité?, in N. de Grove-­Valdeyron, Les nouveaux enjeux de la politique pharmaceutique européenne: pour des produits de santé sûrs, innovants et accessibles’ (2018 mai) Journées Louis Dubouis, 23–24; Toulouse, Les Actes de la Revue de droit de l’Union Européenne (Editions Clément Juglar Janvier 2019) 23–32, 30.

Regulating Medical Devices in the European Union   803 provided by the same watch/bracelet could be considered well-­being data if they are used by the source person only, whereas these same data will be considered health data if they are required by or transmitted to a healthcare professional. The same is true for the watch/bracelet’s legal qualification as a medical device.38 Overall, the EU regulation of e-­medical devices includes specific provisions both in the EU Regulation on medical devices and in the EU’s GDPR. The legal qualification of e-­objects and their associated data depends not only on the legal definitions but also on the objects’ intended use. The EU regulatory landscape thus relies on a case-­by-­case analysis, and companies may use the grey zones to choose the legal regime that is most adapted to their economic strategies. For instance, they could claim to produce well-­ being data and then process and sell them.

7 Conclusion The European framework on medical devices has historical specificities which have led to less burden for the commercialization of medical devices than for medicines, mainly through harmonization of essential requirements only and access to the market without direct intervention of European or national authorities, with control performed by the notified bodies. This has facilitated access of medical devices to the market, although the real access of European patients to medical devices is limited by heterogeneous national regulations on pricing and reimbursement evolving continuously towards a rationalization and decrease of expenses due to the imperative of controlling national health budgets. Beyond this pricing and reimbursement challenge, which is the same for medicines, the 2017 Regulations on medical devices have established strengthened requirements to better achieve the objective of ensuring a high level of protection of public health as well as trust in the medical devices sector. In addition to the reinforced role of and control over notified bodies or the clarification of the economic operators’ obligations, one main innovation of the medical devices’ Regulations is to make necessary the concept of clinical evaluation based on a continuous analysis of clinical data from research to commercialization. One consequence may be that the criterion of efficacy of medical devices, which has never been officially recognized as fundamental, as it is for medicines, except in the national contexts of reimbursement regarding effectiveness (cost-­efficacy), may be more necessary with the implementation of the new Regulations. Moreover, in addition to the provisions in the Regulations discussed earlier, a range of other guidelines and benchmarks that apply to medical devices have been developed by 38  For instance, the European Court of Justice of the European Union has considered that drug prescription assistance software, since it has a medical purpose, is a medical device. Judgment of the Court (Fourth Chamber) of 7 December 2017, Syndicat national de l’industrie des technologies médicales (Snitem) and Philips France v Premier ministre and Ministre des Affaires sociales et de la Santé, C-­329/16, ECLI:EU:C:2017:947.

804   Aurélie Mahalatchimy national health authorities39 but also by professional bodies.40 In this regard, as well as regarding the strengthening of safety requirements of these health products, especially for class III medical devices, the 2017 medical devices Regulations bring the legal regime for medical devices in the EU closer to that for medicines. The marker between the evidence to be generated for safe access to good-­quality medical devices and the promotion of innovation and economics objectives has been moved in favor of public health and patient protection. The appropriate level of regulation and evidence to be generated will clearly be higher than under the previous medical devices’ Directives. Although such an approach can be recognized from the public health objective, one can wonder whether the consequences for the medical devices market, which will become more regulatorily burdensome, have been adequately considered. Indeed, apart from the EUDAMED database and patent protection, the European incentives for innovation and competition which rely more on the legal framework itself under the European Directives, has not been particularly strengthened, as this chapter has shown. While only time will tell whether the new legal framework does not impose more burdens or barriers on innovation and access to the market than necessary, we can expect that specific incentives, such as a period of exclusive marketing regarding data exclusivity for medical devices, as found in the United States, will be needed.

Bibliography Adèle PA, Desmoulin-Canselier S, ‘Droit des dispositifs médicaux: vers une réforme ou un simple réaménagement?’ (2016) 5 Revue de Droit Sanitaire et Social 930. Allensteller C, Medical Devices, EU Policy making and the implementation of health and safety in France (Transactions Publishers 2008). Allensteller C and Permanand G, ‘EU Regulation of Medical Devices and Pharmaceuticals in Comparative Perspective’ (2007) 24(5) Review of Policy Research 385. Bister S, ‘La sécurité des médicaments et des dispositifs médicaux, quelles améliorations depuis le traité de Lisbonne’, in De Grove-Valdeyron N, Les nouveaux enjeux de la politique pharmaceutique européenne: pour des produits de santé sûrs, innovants et accessibles (Editions Clément Juglar 2019) 11. Callens S et al., ‘e-Health’, in Den Exter A, European Health Law (Maklu 2017) 581. Costes L, ‘Applis e-santé: élargissement par la CJUE de la qualification de dispositif médical’ (2018) 145 Revue Lamy droit de l’immatériel 44. De Bruyne J, Vanleenhove C, ‘Liability in the Medical Sector: The “Breast-Taking” Consequences of the Poly Implant Prothese Case’ (2016) 24(5) European Review of Private Law 823.

39  For instance, the French ‘Haute Autorité de Santé’ provides numerous recommendations on patient care. 40  For instance, MedTech Europe has adopted a Code of Ethical Business Practice.

Regulating Medical Devices in the European Union   805 De Grove-Valdeyron N, ‘Politique de santé de l’Union européenne et transformation numérique des soins: quels enjeux pour quelle compétence?’ (2019) 624 Revue de l’Union européenne 39. Dumartin C, Aulois-Griot M, ‘Traçabilité et vigilance: deux outils complémentaires au service de la sécurité du dispositif médical et du patient’ (2018) 1 Revue de Droit Sanitaire et Social 60. Garnier E, Perroy AC, ‘Le Règlement européen n° 2017/745 sur les dispositifs médicaux: une clarification des responsabilités des opérateurs économiques’ (2018) 1 Revue de Droit Sanitaire et Social 19. Hervey TK and McHale JV, European Union Health Law (Cambridge University Press 2015). Hervey TK, Young CA, Bishop LE, Research Handbook in EU Health Law and Policy (Edward Elgar 2017). Hodges C, ‘Do We Need a European Medical Devices Agency?’ (2004) 12(3) Medical Law Review 268. Hutchinson A, Fischer N, Schroler P, ‘Is There a Future for Medical Device SPCs? Past, Present and Future Perspectives’ (2017) 16 (3) BIO-Science Law Review 143. Le Gal Fontes C, Chanet M, ‘Le rôle et les conditions de surveillance des organismes notifiés: une réforme tant attendue’ (2018) 1 Revue de Droit Sanitaire et Social 34. Kalokairinou L, Borry P, Howard HC, ‘Regulating the Advertising of Genetic Tests in Europe: A Balancing Act’ (2017) 54 Journal of Medical Genetics 651. Kalokairinou L, Howard HC, Slokenberg S et al., ‘Legislation of Direct-to-Consumer Genetic Testing in Europe: A Fragmented Regulatory Landscape’ (2018) 9(2) Journal of Community Genetics 117. Li P and Faulkner A, ‘3D Bioprinting Regulations: A UK/EU Perspective’ (2017) 8 (2) European Journal of Risk Regulation 441. Monziols G, ‘Le droit des dispositifs médicaux de diagnostic in vitro: une analyse au prisme de la sécurité sanitaire’ (2018) 1 Revue de Droit Sanitaire et Social 71. Nicolle F, ‘CJEU Closes the Door on SPCs for Medical Devices’ 2019, 41(5) European Intellectual Property Review 324. Peigné J, ‘Le nouveau cadre juridique des dispositifs médicaux’ (2018) 1 Revue de Droit Sanitaire et Social 3. Peigné J, ‘La notion de dispositif médical issue du règlement (UE) 2017/745 du 5 avril 2017’ (2018) 1 Revue de Droit Sanitaire et Social 5. Pelkman S, ‘The New Approach to Technical Harmonization and Standardization’ (1987) 25 Journal of Common Market Studies 294. Quinn P, ‘The EU Commission’s Risky Choice for a Non-Risk Based Strategy on Assessment of Medical Devices’ (2017) 33(3) Computer Law & Security Review 361. Rage-Andrieu V, ‘L’apport du règlement 2017/745 à l’évaluation clinique des dispositifs médicaux’ (2018) 1 Revue de Droit Sanitaire et Social 47. Rial-Sebbag E, ‘Les produits de santé connectés: quelle sécurité?’, in De Grove-Valdeyron N, Les nouveaux enjeux de la politique pharmaceutique européenne: pour des produits de santé sûrs, innovants et accessibles (Editions Clément Juglar 2019) 23. Roscam Abbing HDC, ‘New EU Rules for in Vitro Diagnostics Genetic Tests: A First Step in the Right Direction’ (2017) 24(2) European Journal of Health Law 121. Schulte B., Baustel R., ‘Europe- EU-Medical Device Innovation: Impact of the New European Regulations’ (2017) 67 International Product Liability Review 16.

806   Aurélie Mahalatchimy Van Drongelen A et al., ‘National Institute for Public Health and the Environment, The Impact of the New European IVD Classification Rules on the Notified Body Involvement: A Study on the IVDs Registered in the Netherlands’ RIVM Letter Report 2018–0082 (https://www. rivm.nl/bibliotheek/rapporten/2018-0082.pdf). Vollebregt E, ‘Europe’s New Medical Devices Regulations’ Profound Impact on Digital Health’ (2017) 4(4) Digital Health Legal 4.

CONTROL , USE , A N D A L L O C AT ION O F B O D Y PA R T S : ORGANS, HUMAN TISSUE, BLOOD

chapter 38

Propert y i n H um a n Body Pa rts An Old Legal Question for a New Technological Age Justine Pila

1 Introduction Since the informational properties of DNA were discovered in the 1950s, advances in medical and other scientific fields have continued to reveal new and important uses of human body parts for therapeutic and other ends. The result has been a shift in cultural attitude to the body, reflected in a growing tendency to dissociate human genetic material from the bodies in which it is contained, and a growing openness to treating bodily materials as objects of direct commercial exchange. As demands for access to transplantable organs, blood and urine, embryonic stem cells, reproductive materials, and other human tissue for personal, medical, and scientific purposes have increased, this openness has facilitated calls for the recognition of property rights in respect of detached bodily materials especially.1 A central aim of these calls has been to address existing legal uncertainties2 and to give individuals greater control over the use of their bodies and its associated benefits, including a right to share in any profits it generates. 1  From a vast literature see, e.g., LB Andrews, ‘My Body, My Property’ (1986) 16 Hastings Center Rep 28, and other writings in the bibliography in favour or against this view. For some scholars the protection of bodies and bodily materials by property is as much required by law as it is justified by policy. See, e.g., S Douglas and I Goold, ‘Property in Human Biomaterials: A New Methodology’ (2016) 75 Cambridge LJ 478. 2  On which see, e.g., R Magnusson, ‘Proprietary Rights in Human Tissue’ in N Palmer, E McKendrick eds., Interest in Goods (Lloyds Commercial Library 1998); MJ Allen, ‘Human Tissue Ownership and Use in Research: What Laboratorians and Researchers Should Know’ (2010) 56 Clinical Chemistry 1675; L Skene, ‘The Current Approach of the Courts’ (2014) 40 JME 10; Holdich v Lothian Health Board [2013] CSOH 197 (Lord Stewart) (Holdich), especially [29]–[52].

810   Justine Pila For the moment, however, the formal position in most jurisdictions, including in Europe and the United States, is that neither the bodies nor the bodily materials of living or deceased persons are protected by property ‘as such’.3 At the same time, a growing practice of dealing with them in the manner of other property objects4 has led courts to recognize certain exceptions to this position where doing so has proved expedient on the facts of individual cases, including (paradoxically) in order to deny a property claim. As a result, courts in different jurisdictions have recognized people as having sufficient rights of property in respect of their bodily materials—or at least, sufficient rights of possession5—to make an unencumbered and irrevocable gift of them,6 to punish and prevent their unauthorized taking or destruction by third parties,7 and to enable their use for paternity testing,8 among other things. Some US courts have also attributed quasi-property status to human reproductive materials to enable people to use or prevent their use to conceive,9 albeit for reasons invoked by other courts to deny them property.10 Where the courts have been consistent is in their reticence to make definitive statements accepting or rejecting the property status of bodily materials in the current 3  The phrase ‘as such’ appears in Article 21 of the European Convention on Human Rights and Biomedicine (CHRB) with the apparent intention of limiting the scope of its prohibition against the use of bodies and bodily materials for financial gain (to bodies and bodily materials ‘as such’) and, notoriously among patent lawyers, in Article 52(3) Convention on the Grant of European Patents (European Patent Convention) with the effect (whether or not intended) of denuding the article 52(2) list of unpatentable ‘non-inventions’—including ‘discoveries’—of much of its scope and significance. 4  On the long history of body parts and bodily materials being publicly offered and acquired for sale in countries around the world, including with occasional official support, see R Scott, The Body as Property (Allen Lane 1981) ch 1. 5  On the distinctions between rights of use/possession and property and between property objects and objects for which possessory remedies may be granted, see Holdich (n 2) [47] & [75]. 6  Greenberg v Miami Children’s Hospital Research Institute Inc 264 F Supp 2d 1064 (SD Fla 2013) 1074 (declining ‘to find a property interest for the body tissue and genetic information voluntarily given’); Washington University v Catalona 490 F 3d 667 (8th Cir 2007) (affirming the donee of human tissue to be its exclusive owner). 7  R v Welsh [1974] RTR 478 (urine); R v Rothery [1976] Crim LR 691 (blood); R v Herbert (1960) 25 Criminal LJ 163 (hair). 8  Roche v Douglas as Administrator of the Estate of Edward John Hamilton Rowan (Dec) [2000] WASC 146 (2002) 22 WAR 331. 9  Hecht v The Superior Court of Los Angeles County 16 Cal App 4th 836 (Ct App 1993) (finding that stored sperm is a ‘unique type of “property” ’ whose disposition may be determined by the man from whom the sperm came); York v Jones 717 F Supp 421 (ED Va 1989) (accepting frozen embryos deposited with an in vitro fertilization clinic to be the property of the parents held on bailment by the clinic). Cases in Australia and the UK are in agreement. See, e.g., Yearworth v North Bristol NHS Trust [2010] QB 1 (CA); Bazley v Wesley Monash IVF Pty Ltd [2011] 2 QD R 2017 (Queensland SCt); Re Edwards [2011] NSWSC 478. But see Evans v Amicus Healthcare Ltd [2004] EWCA Civ 727 [88] (rejecting the relevance of property to resolve a dispute over access to fertilized sperm). 10  Case C–34/10 Brüstle v Greenpeace eV [2011] ECR I-09821 (Grand Chamber) (denying the patentability of any invention the devising of which involves the destruction of a human embryo on dignity grounds). See further Moore v Regents of the University of California 793 P2d 479 (Cal 1990) (Arabian J) (rejecting the property status of human tissue with reference to human dignity); App no 46470/11 Parrillo v Italy [2015] ECHR 755 [200] (citing Article 1 of the European Convention on Human Rights to conclude that ‘human embryos cannot be reduced to “possessions” ’).

Property in Human Body Parts   811 era of advanced technologies. The following passage from a case involving an attempted kidney donation is representative of the general judicial attitude that pertains.11 Considering . . . that the “no property right” jurisprudence was developed long before the age of transplants and other medical advances, we need not identify or forecast the circumstances in which someone may conceivably have actionable rights in the body or organ of a deceased person. For purposes of this case it is enough to say, in answer to the first part of the first certified question, that plaintiff, as a specified donee of an incompatible kidney, has no common-law right to the organ. For that reason, his cause of action for conversion must fail, as it is necessarily based on his claimed right to possess the kidney in question.

Underlying all of the cases in this area is a deep ambivalence in the face of growing moral and regulatory complexity compounded by a rapidly changing technological and social environment. In that environment, patient rights, individual choice, and consumerism frequently compete with freedom of science and public health in ways that implicate and destabilize foundational concepts of life, personhood, the public interest, and even health and medicine.12 In the absence of property (as such), the legal position is that people are free to use their bodies and bodily materials as they see fit and to decide how they may be used after death, within certain regulatory constraints. For example, all research involving human tissue is subject to ethics review based on a growing and increasingly detailed body of international bioethical principles developed from the Nuremberg Code.13 Central is ensuring that participants receive appropriate information about projects and consent to participate in them, and that their and others’ privacy and data protection interests are fully protected in accordance with international standards and relevant domestic and regional regulatory provisions.14 Increasingly, research ethics councils also require proof of community consultation, particularly for transnational and controversial projects such as biobanking, research in indigenous communities, and the collection of materials for unspecified future use.15 In addition, domestic and international laws 11  Colavito v New York Organ Donor Network Inc 8 NY 3d 43 (NY Court of Appeals 2006) 53. Cf. Colavito v New York Organ Donor Network Inc 356 F Supp 2d 237 (ED NY 2005) (denying the property status of corpses on public policy grounds and rejecting in consequence a claim of conversion by the intended recipient of a donated kidney). 12  T Murphy, Health and Human Rights (Hart 2013) ch 1. On the destabilizing effects of modern technologies on conceptions of health and medicine, see M Brazier, ‘Human(s) as Medicine(s)’ in S McLean ed., First Do No Harm (Ashgate 2006) 187; AL Sussman, ‘The Case for Redefining Infertility’, The New Yorker 18 June 2019. 13  Some kinds of research are deemed unacceptable. See, e.g., United Nations Declaration on Human Cloning (A/C.6/59/L.27/Add.1) (b) (calling for a prohibition on human cloning). 14  On the right to privacy see, e.g., Parrillo (n 10) (deciding that since embryos contain the genetic material of the woman from whose body they originate and identify her, their use engages the Article 8 ECHR right to a private and family life); AB v Leeds Teaching Hospital NHS Trust [2004] EWHC 644 (QB) (legal claims in respect of corpses supported by the Article 8 right). 15  D Nicol et al, ‘Impressions on the Body, Property, and Research’ in Goold et al. eds. (n 1) ch 2.

812   Justine Pila prohibit people from consenting to any use of their bodies or bodily materials by others that would offend the criminal laws of the state or inalienable human rights such as autonomy and human dignity. The same limits apply to third parties. Hence, at the same time as limiting people’s freedom to deal with their bodies and bodily materials as they wish, criminal and other laws empower people in the exercise of that freedom by protecting it. An important further source of such protection is the modern right of bodily integrity, based on human dignity, which prevents any medical intervention on or use of a person’s body or bodily materials to which she has not consented.16 While these rights give individuals a measure of control over their persons, it is less extensive than the control they would have under a property regime. Increasingly, it is also balanced against other constitutional rights aimed at protecting third-party and communitarian interests, such as freedom of science and public health.17 Very different is the position concerning objects that have been created from human bodies or bodily materials, both of which may be protected by property and fully commercialised. Examples range from preserved skeletons to embryonic stem cell cultures, copies of DNA sequences (cDNA), and so-called biologics: drugs produced by genetic engineering from protein antibodies or other biological materials. The reason for the different position of such things is the separate provision at law for products of nature transformed by a person’s exercise of skill or labour into a new and different species of object. Where the new and different object is tangible, such as a preserved skeleton or culture of embryonic stem cells, it may be a ‘good’ protected by personal property vesting in the person who ‘created’ it.18 Where the new and different object is intangible, such as a method of culturing stem cells or producing a biologic, or the immortal cell line or biologic achieved by applying that method, it may be an ‘invention’ or other type of intellectual product protected by intellectual property (IP) vesting also in the person who created it or was responsible for its creation. Indeed, such products are among the most commercially lucrative of property objects, underlining what is at stake for industry in discussions of body property. For example, in 2018 the anti-inflammatory drug adalimumab (marketed as Humira) generated nearly US$20 billion in sales for AbbVie, and the cancer drug trastuzumab (marketed as Herceptin) generated approximately US$7 billion dollars for Roche. Both were created from human biological material and have been protected by a large number of patents in jurisdictions around the world: at the date of writing, 403 for Humira alone.19 The distinction between unprotected bodies and bodily materials as such and protectable human creations incorporating or derived from them is fraught with concep16  Charter of Fundamental Rights of the European Union [2012] OJ C 326/391 (CFR) art 3. 17  On the role of fundamental rights, including to property, see M Goodwin, ‘Human Rights and Human Tissue: The Case of Sperm as Property’ in R Brownsword et al., eds., The Oxford Handbook of Law, Regulation, and Technology (Oxford University Press 2017); LS Underkuffler, The Idea of Property: Its Meaning and Power (Oxford University Press 2003) 106. 18  See, e.g., Doodeward v Spence (1908) 6 CLR 406 (HCA) (Doodeward). 19  See https://www.drugpatentwatch.com/p/biologics/tradename/HUMIRA (last accessed 12 November 2019).

Property in Human Body Parts   813 tual and philosophical difficulty. That difficulty is simultaneously underlined and compounded by the expansive interpretation in some jurisdictions of what counts as an act and object of human creation sufficient to entitle an individual to personal or intellectual property. In all countries, exercising work or skill to transform an unowned object (res nullius) into something new and different will suffice to entitle a person to property in the result. For example, extracting T cells from a person’s blood and using the CRISPR/Cas9 gene-editing tool and other techniques to modify them for use as a cancer treatment, as has been done by researchers from the University of Pennsylvania recently,20 would in all countries be regarded as involving the transformation of the original T cells into something new and different. If neither the T cells nor the added genetic material is the subject of property or other access restrictions, there will be no impediment to a claim of ownership in respect of the result, whether as a tangible object in a Petri dish or an intangible invention comprising a method of producing the modified cells (and cancer treatment) or the modified cells (and cancer treatment) themselves.21 In Europe and other jurisdictions, however, merely reducing the T cells to a new and useful form by extracting them from a person’s body and placing them in a Petri dish will also suffice for certain intellectual and personal property purposes.22 As a result, and in those jurisdictions at least, both species of property can subsist in detached bodily materials, notwithstanding the preceding ‘no property’ rule, in virtue of the work or skill required to detach them and/or the value of the objects once detached. The availability of intellectual and personal property for objects incorporating or derived from bodies and bodily materials simultaneously diminishes and increases the practical significance of the preceding ‘no property’ rule: diminishing it by filling what initially appears to be a property void with property rights and increasing it by vesting such rights in the persons responsible for the work or skill required to ‘produce’ the relevant object rather than the persons whose body or bodily materials they are. The result in practice is to create opportunities for the profitable use of human bodies from which the latter persons may be excluded, as well as opportunities for third-party ownership of detached bodily materials owing to the work or skill involved in detaching them. Well-known examples include the patents granted for products comprising and derived from cells removed from the bodies of John Moore and Henrietta Lacks, the ownership 20  The modifications included (a) inserting a gene into the cells to cause them to target other cells with an excessive amount of the antigen NY-ESO-1 associated with certain types of cancer and (b) editing the cells to suppress their production of different proteins that might (i) interfere with that targeting and (ii) prevent them from killing the targeted cancer cells. For the details see L Thulin, ‘Four U.S. CRISPR Trials Editing Human DNA to Research New Treatments’ Smithsonian.com 3 September 2019; ‘Early Results from First-In-U.S. Trial of CRISPR-Edited Immune Cells for Cancer Patients Suggest Safety of Approach’, Penn Medicine News 6 November 2019. 21  See, e.g., US Patent application no. US2019247432 (A1) entitled ‘Genetically Modified Immune Cells Targeting NY-ESO-1 and Methods of Use Thereof ’ (15 August 2019) filed by the University of Pennsylvania. 22  See, e.g., Yearworth (n 9); Directive 98/44/EC of the European Parliament and of the Council of 6 July 1998 on the legal protection of biotechnological inventions [1998] OJ L 213/13 (Biotech Directive) arts 3(2), 5(2).

814   Justine Pila and profits from which lay entirely with the scientists responsible for the products themselves (and their employing institutions) to the exclusion of Moore and Lacks.23 In recent years, these opportunities have increased exponentially as developments in modern digital and biotechnologies have enabled the informational and biological properties of human tissue to be unlocked and exploited for a range of therapeutic, scientific, technological, and commercial ends. Among other things, these developments have enabled material previously regarded as surplus or waste, such as residual embryos from in vitro fertilization and the blood from umbilical cords and placentas, to be transformed into valuable scientific and commercial resources owing to the genetic material they contain. Hence the growing tension that exists between the persons whose bodies and bodily materials are needed to realize these opportunities, and those with a commercial, scientific, or other interest in seeing them realized. In legal terms, that tension has been expressed in different ways, including via support for the application of different property models.24 According to one model, bodies and bodily materials are (or should be recognized as being) personal property capable of being gifted, abandoned, sold, and/or held on deposit or trust in the manner of other objects.25 According to another model, they are common or collective property, or at least a scarce and valuable community resource in which the public at large has an interest that is (or should be) enforceable by property.26 Central is the question of who should control access to the diverse and substantial benefits able to be reaped from the use of human bodies and bodily materials in the current technological age, and on what legal and moral bases. In other words, the status of body parts as property or not has its most practical significance for questions regarding people’s interests in the commercialization of their bodily materials.

2  The ‘No Property’ Rule The law reports of all common law jurisdictions contain judicial statements rejecting the property status of human bodies, corpses, and bodily materials (as such), whether

23  According to Rebecca Skloot, a search of the US Patent and Trademark Office database in 2013 turned up ‘more than seventeen thousand patents involving HeLa [Henrietta Lacks’s] cells’: R Skloot, The Immortal Life of Henrietta Lacks (Pan Books 2010) 222. 24  For a cultural critique of the application of these models and the competition between so-called gift and property approaches to donation specifically, see D Joralemon, ‘Organ Wars: The Battle for Body Parts’ (1995) 9 Med Anthropol Q 335. 25  I Goold, ‘Abandonment and Human Tissue’ in Goold et al. eds. (n 1) ch 9 and other sources in the bibliography. 26  C Stewart et al, ‘Public Umbilical Cord Blood Banking and Charitable Trusts’, in Goold et al. eds (n 1) ch 5 (invoking models of charitable gifts and charitable trusts).

Property in Human Body Parts   815 directly or by denying the availability of property-based relief.27 Given the association of body ownership with slavery, it is unsurprising that such statements tend to go unchallenged in so far as the bodies of living persons and their materials are concerned. By contrast, commentators have repeatedly doubted the ‘no property’ rule in respect of corpses and lawfully detached bodily materials. For example, it has been said that the judicial authorities traditionally cited for the absence of property in corpses do not in law support that result,28 and that the denial of personal property with respect to these and lawfully excised human tissue is outdated and ill-conceived.29 In order to determine the validity of such doubts and the property status of bodies and bodily materials in ­general, it is appropriate to begin by considering the nature of property and the tests deployed at common law to establish its existence. Among the issues that arise in this regard are whether the noncommercial property models that some support in this context are models of property at all, and what significance moral and constitutional values have and ought to have in the recognition and protection of rights of property.

3  The Nature of Property and Test for Its Existence In common law and other jurisdictions, rights of property—or in legal language, real rights—are conceived as transferrable rights to exclude all others from one or more use of a certain thing (the property object).30 Examples are the rights typically conferred in respect of land and manufactured goods such as cars and pens, which entitle their owner to exclude all others from occupying or otherwise interfering with the protected object and, in doing so, protect the property owner’s freedom to enjoy exclusively its benefits. While such rights plainly affect and help determine the relationship between the property holder and other persons, they derive in scope and content from the thing to which they attach and are in this sense rights to (or against) that thing. In taxonomical schemes, they are typically contrasted with personal rights, which, unlike property, are defined with reference to a person rather than an object. The most commonly cited examples of personal rights are the rights conferred by contract and tort, each of which is specific to 27  Yearworth (n 9) [30]–[32]; Moore (n 10) 135. Civil law jurisdictions have been even less accommodating of property claims in respect of human bodies. See generally HAMJ Ten Have and JVM Welie eds., Ownership of the Human Body (Springer 1998). 28  R Hardcastle, Law and the Human Body (Hart 2007). 29  Andrews (n 1); A Campbell-Tiech, ‘Annotation: A Corpse in Law’ (2002) 117 Br J Haematol 809, 811; P Matthews, ‘Whose Body? People as Property’ (1983) Curr Legal Problems 193; and sources in bibliography. 30  While legal writers often reference rights of property as rights in rem, I have avoided that term for the reasons given by Neil MacCormick (in N MacCormick, Institutions of Law: An Essay in Legal Theory [Oxford University Press 2007] 136) and used his language of ‘real rights’ to include realty and personalty instead.

816   Justine Pila the claimant and enforceable by her only against the one or other identifiable persons who, by entering a contract or assuming a relevant duty of care or other formal responsibility, have become legally bound to conduct themselves toward her in a certain way. Other examples include rights of reputation and bodily integrity, which share the general exigibility of property (in the sense of being enforceable against the world) but are nonetheless personal rather than property in virtue of their attachment to a person rather than an object.31 The definition of property rights as transferrable rights to exclude all others from certain benefits of a given object raises three initial questions: What types of transferrable right are essential to property? What types of object can property protect? And how do we know whether a given object is protected by property and identify the initial beneficiaries of its protection? In answering these questions, regard must be paid both to the property systems that exist and the (actual and desirable) reasons for having them.32

3.1  Property Rights and Objects Generally speaking, and within limits, the common law is agnostic about the range and content of the rights conferred by property, consistent with the ‘bundle of sticks’ conception of property that pervades much contemporary property thinking.33 Essential for rights to constitute rights of property is that they be exclusionary in nature, attach to a discrete and objectively discernible object, and be capable of transfer and enforcement against all others. While property is closely associated with positive rights of exclusive possession and use, the example of IP demonstrates that a right need not cease to be proprietary merely in virtue of its omission of either or both of these. Patents, for example, do not confer rights to use the protected object, for which separate regulatory approval is required, much as for property in cars and other dangerous goods. In addition, the intangible nature of the objects protected by IP makes them incapable of supporting exclusive occupation. That is why, in an eighteenth-century dispute involving literary property, the King’s Bench in England rejected a conception of property as requiring exclusive occupation in preference for one based on John Locke’s theory of property in human labour and the objects we mix it with.34 Nor need property subsist indefinitely rather than for a limited period or be enjoyed free from restriction on its valid exercise. Indeed, no rights of property are so enjoyed, since all are subject to the background laws 31  C Webb, ‘Three Concepts of Rights, Two of Property’ (2018) 38 OJLS 246, 258. 32  This is consistent with Ronald Dworkin’s interpretive theory of law. In the context of property specifically, see also J Waldron, ‘Property and Ownership’ in EN Zalta ed., The Stanford Encyclopaedia of Philosophy (Winter 2016 edition); https://plato.stanford.edu/archives/win2016/entries/property/ (last accessed 11 October 2019). 33  W Hohfeld, ‘Fundamental Legal Conceptions as Applied in Judicial Reasoning’ (1913) 23 Yale LJ 16; (1917) 26 Yale LJ 710; and sources in the bibliography. 34  Millar v Taylor (1769) 4 Burr 2303, 98 ER 201 (KB) (Millar v Taylor); J Locke, Two Treatises of Government (1690) S 27.

Property in Human Body Parts   817 of the relevant jurisdiction, including its criminal laws and fundamental rights.35 That is why the owner of land in a residentially zoned area may not build a factory on it, why the owner of a car may not drive it recklessly, and why the owner of a pen may not use it to conceal another’s written expression. Less clear is the importance of commercialization to property and whether a transferrable right to exclude others from the use of an object remains a right of property if the transfer is unable to take place on the open market. Prohibiting the commercial exploitation of property rights and objects is certainly unusual, since a central aim of granting property is to support self-regulating market economies, the efficient allocation of resources, and opportunities to generate individual and collective wealth. Hence the tendency of theorists to treat property as linked to freedom of contract.36 Nonetheless, and while property must be conceived to retain its legal distinctiveness from the two species of personal right mentioned previously, it is a sufficiently flexible and discrete concept to be capable of limitation in its permissible means of exploitation, as modern competition and other laws reflect.37 From a public policy perspective, imposing such limits in certain contexts also has the benefit of enabling objects to be protected by transferrable exclusionary rights without this necessarily entailing their complete commodification by exposing them fully to market valuation and exchange. Indeed, this is a theme of much contemporary writing about body property. Particularly influential in this regard has been the argument of Margaret Radin that certain things, including human bodies and bodily materials, are so integrally connected to the world of objects and community as to exhibit internally plural meanings that the law ought to honour by limiting without preventing their treatment as commodities.38 The recognition of limited rights of property—such as exclusive rights to possess but not dispose of one’s bodily materials, or exclusive rights to dispose of such materials but without financial gain—is an obvious way of achieving this imperfect commodification of bodies and bodily materials.39 The law is similarly agnostic regarding the types of object capable of supporting rights of property. For example, while valid property claims will often involve a tangible product capable of exclusive possession, intangible objects such as inventions, authorial works, and other intellectual products have for centuries also been protected by property. Only two things would seem to be required from a positive law perspective: first, that there exists an object sufficiently determinate in its essential aspects to support transferrable exclusionary rights and corresponding legal duties and, second, and of especial importance for bodies and bodily materials, that there are no moral or public

35  Waldron (n 32). 36  Cf J Penner, The Idea of Property in Law (Oxford University Press 1997) (arguing that the right of alienation is not intrinsic to property but rather depends on contract law). 37  This is consistent with regulatory views of property. Webb (n 31); K Gray, ‘Property in Thin Air’ (1991) 50 Cambridge LJ 252, 304; and sources in bibliography. 38  See generally M Radin, Contested Commodities (Harvard University Press 1996) ch 7. 39  SR Munzer, A Theory of Property (Cambridge University Press 1990).

818   Justine Pila policy objections to recognizing such rights and duties or enforcing them by exercise of a state’s coercive powers.40

3.2  Property Origins and Beneficiaries There remains the final and related question of how property originates, or, more pragmatically, how the law determines whether property exists and identifies its initial beneficiaries. These are among the most difficult and obscure questions of property law and theory. Historically, and at common law at least, property rights have been treated as arising from a person’s assertion of a sufficiently close or special connection to an object to justify the grant of exclusionary rights in respect of it. Assertions of property are generally implicit rather than explicit and inferred from such acts as producing, using, or maintaining continual possession of a putative property object.41 For tangible objects, such as cars and pens, possession has especial legal weight as ‘the outward expression of ownership’.42 For intangible objects, such as poems and technologies, production displaces possession in evidentiary significance, consistent with the Lockean theories of property discussed earlier.43 Whatever the case, the presumptive beneficiaries of any property that exists are accepted to be the person(s) responsible for the act from which its existence was inferred. Of course, it is not the act or assertion itself that triggers property, but rather the state’s recognition of its legitimacy by agreeing to enforce the actor’s claims against those of other persons. This raises the normative question of why the state might and should agree to recognize the legitimacy of individual property claims, to which one or more of a range of consequentialist and deontological answers may be given. One is that doing so will encourage individuals to coordinate their investment of time and energy efficiently by focusing on different (natural and manufactured) artefacts and facilitating their allocation to the persons in whose hands they have greatest use. Aside from reducing the risk of wasted effort, coordinated investment and efficient allocation reduces the risk of social conflict over scarce resources and helps to ensure that the use of those resources is maximized, along with human effort and activity. It also rewards and encourages the production and social diffusion of objects of value to individuals and communities, recognizes people’s rights in the work of their hands, and creates opportunities for social and economic interaction and exchange, including for the purpose of 40  For examples of the courts’ express reliance on morality and public policy to deny property in respect of bodies and bodily materials see, e.g., the cases cited in n 10 and n 11. 41  Cf the Roman Law view of property acquisition by occupatio (occupation of unowned objects [res nullius] or land [terra nullius]), accessio (the accession of one object to another), and specificatio (a human act of creation). For a discussion of the relevance of Roman Law principles for the property status of bodies and bodily materials see Holdich (n 2). 42  F Pollock, R Wright, An Essay on Possession in the Common Law (Oxford University Press 1888) 4. 43  See n 33 and associated text.

Property in Human Body Parts   819 generating (autonomy-enhancing and community-building) wealth. And finally, it promotes individual freedom and good citizenship, including by reducing social conflict over scarce resources, providing the security of possession of things that people need to live and flourish, and enabling people to express and develop their personal and altruistic selves by engaging in creative and other productive acts and voluntarily donating things of value to other individuals and the general public. Any conception of property can be expected to reflect one or more of these answers and the normative commitments they entail. From the preceding discussion, three obstacles to the recognition of property in human bodies and bodily materials can be identified. The first concerns their ability to support transferrable exclusionary rights. The second concerns the implications of their recognition and protection for morality and public policy. And the third, related to but different from the first, concerns their ability to satisfy the test of property’s existence. The first and third of these difficulties are practical and conceptual and specific to living bodies and attached bodily materials. The second is political and philosophical and extends also to corpses and detached bodily materials.

4  No Property in Living Human Bodies or Attached Bodily Materials as Such 4.1  Capacity to Support Transferrable Exclusionary Rights and Satisfy the Test of Property’s Existence An argument can be made that human bodies are not capable of supporting the transferrable exclusionary rights of property. The reason is that they lack the discrete ex­ist­ ence required of property objects by virtue of their inseparability from the persons whose body they are. Since we are all the exclusive occupants of our bodies and cannot in practice be vacated of such occupancy, possession is not among the rights that property in a body might confer. In addition, the possession we each have of our bodies gives us a large degree of practical control over their use, which limits further beyond occupancy the range of transferrable exclusionary rights that might feasibly be recognized in respect of them. Such limits are additionally re-enforced and extended by the legal and constitutional guarantees of human dignity and autonomy mentioned previously, which represent further obstacles to the recognition and enforcement of property rights in modern civil societies. The same ‘lack of discrete existence’ objection to property can also be raised for attached bodily materials owing to their inseparability from the bodies of which they are part. In deciding whether such materials are property, the courts have additionally relied on the absence of any act from which a claim of property might legitimately be inferred. Specifically, they have doubted the ability of attached bodily materials to satisfy

820   Justine Pila the test for the existence of property owing to the insufficiency of self-possession as a basis for inferring claims of property. According to the UK House of Lords in R v Bentham,44 for example: One cannot possess something which is not separate and distinct from oneself. An unsevered hand or finger is part of oneself. Therefore, one cannot possess it. . . . A person’s hand or fingers are not a thing.

On this view, the same considerations of necessary occupancy and use that cast doubt on the ability of human bodies to support transferrable exclusionary rights also limit the scope for their occupants to assert the required claim to property in respect of them. This line of reasoning has a certain sense to it. We do not, after all, create our bodies, and we are as much occupied by as occupants of them. Having said that, and issues of selfpossession aside, we certainly invest time and effort in maintaining them and, increasingly, also in modifying them, assisted by a growing range of technologies,45 and it is difficult to see why this investment is not a sufficient basis for inferring the assertion— not to mention the connection—required of property. In addition, we create others’ bodies and are also able increasingly to design them to taste, assisted again by increasingly sophisticated technologies; a recent example being the use by Chinese scientist He Jiankui of the CRISPR gene-editing tool to ‘create’ the world’s first genetically engineered twins.46 In this context especially, the most convincing reason for the law’s rejection of human bodies as property may simply be the unwillingness of judges and other law- and policy-makers on moral grounds to countenance the treatment of rights-bearing persons as having the fungibility and commensurability of other products amenable to market valuation and exchange.47 Rejecting the presumption of commercialization changes the analysis and invites an alternative view of the special connection between persons and their bodies as the basis of noncommercial or even nontransferrable rights of self-ownership.48 The latter view in particular has been associated with John Locke’s theory of property.49 In the submission of counsel in Re A (Conjoined Twins: Surgical Separation),50 for example: Existing case law [regarding the ability of doctors to separate conjoined twins to save the life of one at the expense of the other] is based upon the presumption of 44  R v Bentham [2005] UKHL 18 [8]. 45  M O’Connell, To Be a Machine (Granta 2017); R Godwin, ‘The Impossible Body’, The Guardian 22 September 2018. 46  A Ahuja et al, ‘Mastering Evolution: The World’s First Gene-Edited Babies’, The Financial Times 30 November 2018. Cf Evans (n 9). 47  See, e.g., the sources cited in n 10 and n 11. 48  M Quigley, Self-Ownership, Property Rights, and the Human Body: A Legal and Philosophical Analysis (Cambridge University Press 2018). 49  As discussed, this theory has mainly been important for its support of property in respect of objects derived from human labour. 50  Re A (Conjoined Twins: Surgical Separation) (2000) 57 BMLR 1 (Fam) 82.

Property in Human Body Parts   821 bodily integrity. John Locke’s assertion that ‘every Man has a Property in his own Person. This no Body has any Right to but himself ’ (Two Treatises of Government, 1690) . . . underpins much of the moral dialogue in this area.

In the European Union, the concept of bodily integrity referenced in Re A has the status of a fundamental right under the EU’s Charter of Fundamental Rights ‘to respect for [one’s] physical and mental integrity’ including, ‘[i]n the fields of medicine and biology’, respect for the free and informed consent of the person concerned. . . , the prohibition of eugenic practices. . . , the prohibition on making the human body and its parts as such a source of financial gain, [and] the prohibition of the reproductive cloning of human beings.

Thus defined, and as suggested by its description as a personal right previously, it is difficult to see the right of bodily integrity as one of property.51 So, too, and consistent with its equation to bodily integrity, the right of self-ownership suggested by the statement of Locke can be interpreted as less proprietary than personal depending on how prohibitively one reads its denial of the scope for third-party rights.52 On this view especially, each of the rights referenced brings to mind the so-called moral rights from the field of IP. Designed to supplement the property rights of authors and inventors, these include certain nontransferrable (and often inalienable) rights to be associated with, share in any profits from, and control the public presentation of any authorial works and inventions that one creates. Of especial current relevance is the right of authorial integrity, premised on the Hegelian idea of works as extensions and expressions of personhood, which enables authors to prevent derogatory treatments of their works prejudicial to their honour or reputation as authors specifically.53 While interpreted at law as an individual right, the right of authorial integrity can also be conceived as a polyvalent constitutional right that protects authorial discourse (rather than works of authorship) from the corrupting influences of commerce and politics.54 In this and other respects it offers a valuable counterpart to the right of bodily integrity in the current technological and economic climate and is a possible model for its explication and development, whether or not as a supplement to body (including genetic) property.55 51  N 27 and associated text. 52  See also E Wicks, ‘Property Rights in the Body’ in Human Rights and Health Care (Hart 2007) (discussing self-ownership and arguing that possessory rights are better able to explain and govern the relationship between our minds and bodies). 53  Convention for the Protection of Literary & Artistic Works 1886/1971 (Berne Convention) art 6bis(1). 54  See CB Graber, G Teubner, ‘Art and Money: Constitutional Rights in the Private Sphere’ (1998) 18 OJLS 61. 55  On the relevance of authors’ and inventors’ moral rights as a model for regulating access to the benefits of bodies and bodily materials, see further J Pila, ‘Intellectual Property Rights and Detached Human Body Parts’ (2014) 40 JME 27.

822   Justine Pila The difficulty of formulating an adequate test for property’s existence highlights the instability of its conceptual underpinnings and the elusiveness of its origins. In the context of this uncertainty, legal tradition and practice assume especial importance. Central to each has been the unwillingness of judges and other law- and policy-makers across jurisdictions to recognize human bodies and bodily materials as supporting the transferrable exclusionary rights of property, whether commercial or noncommercial and whether on conceptual or normative grounds.

4.2  The Morality of Enforcing Property Claims The image of a person attempting to deal with another’s body as an object of property highlights the second difficulty, which is one of morality. For supporters of moral rights of bodily integrity (or nontransferrable rights of self-ownership), this difficulty does not arise owing to the rights’ association with human dignity. For others, however, it may offer a more convincing explanation and justification for the denial of property in respect of human bodies and bodily materials than technical considerations regarding their capacity to support transferrable rights. This is especially so given the relevance of moral values to understandings of that capacity; it being, as discussed earlier, partly the insulation of individuals from certain consequences of property protection via criminal and constitutional laws that precludes the recognition of bodies and bodily materials as objects of transferrable exclusionary rights on conceptual and practical grounds. Consistent with this, discussions of the moral implications of recognizing body property tend to lead from autonomy, and a view of the state as deriving its purpose, and hence its legitimacy, from its unique capacity to protect the freedom of individuals to be self-determining subject to a prohibition against causing harm to others.56 Indeed, and as seen earlier, supporting autonomy is a central justification for recognizing property claims. But it is also difficult to reconcile with the state’s enforcement of third-party exclusionary rights in respect of others’ bodies. While an individual might conceivably seek to confer such rights consensually by selling her body for financial gain, even liberal theorists have tended to reject the permissibility of such attempts, quite apart from their plausibility under human conditions of social inequality, for either of two reasons. The first is their subversion of the very freedom that libertarian theory seeks to protect, and the second is their causing of harm to others by demeaning humanity.57 The latter view in particular has been widely supported in body property debates, including with reference to human dignity and the respect it requires for persons as ends rather than as means to an end. By treating bodies and bodily materials as instruments, property is

56  JS Mill, On Liberty (1859). 57  D Archard, ‘Freedom Not to be Free: The Case of the Slavery Contract in JS Mill’s On Liberty’ (1990) 40 Philosophical Q 453.

Property in Human Body Parts   823 said to offend this value and corrupt the ideals associated with humanness, including those of moral agency and responsibility.58 By focusing on the intrinsic qualities of bodies and bodily materials, dignity-based objections to property raise difficult issues regarding the nature and origins of personhood and the legal and moral status of the body at different stages of its development. In this, they support a right of bodily integrity even more controversial and expansive than the Hegelian-premised right of authorial integrity.59 For the same reason, and unlike some objections from autonomy, dignity-based objections pertain even in societies of perfect equality in which all private arrangements are truly consensual.60 Having said that, context is not irrelevant. For example, an argument can be made that as long as the state and market encourage people to treat themselves and others as commodities in the myriad ways that they do currently, human dignity is especially vulnerable and in need of particular protection from the further erosion that treating bodies and bodily materials as property would entail. Indeed, it is presumably the awareness of this vulnerability in the current era of body-focused technologies that accounts for the increased emphasis on bodily integrity as a standalone fundamental right. Viewed in this light, conceptions of bodily integrity as a polyvalent right enforceable against even its individual bearers to protect personhood from the corrupting influences of the market and state have especial relevance.61

4.3  Conclusion: No Property but Freedom of Use There is reason to doubt the capacity of human bodies and attached bodily materials to support the transferrable exclusionary rights of property and even better reason to oppose their protection by property on moral grounds. This is particularly the case if commercialization is conceived as an essential aspect of property, and property rights as entailing the freedom to buy and sell the objects to which they attach. So serious are the threats posed to autonomy and dignity particularly by treating human bodies and bodily materials as objects of market valuation and exchange that even the ‘moral’ use of property to solve problems arising outside the market, such as the shortage of transplantable organs and essential medicines, is difficult to countenance. The question remains: If living bodies and the materials contained in them are not protected by transferrable exclusionary rights, what use might people lawfully make of 58  MJ Sandel, ‘What Money Can’t Buy: The Moral Limits of Markets’, The Tanner Lecture on Human Values (Oxford University Press 1998); Radin (n 38); and other sources in the bibliography. 59  The recognition of authors’ moral rights in article 6bis of the Berne Convention following the Rome Revision of 1928 was a major reason for the failure of the United States to accede to the Convention until 1988. 60  Sandel (n 58). 61  S Orbach, Bodies (Profile 2019); A Barker, ‘Person in the News: Harry, Duke of Sussex,’ Financial Times 5/6 October 2019, 11 (reporting Prince Harry’s criticism of the British tabloid media for having ‘commoditised’ his mother and wife ‘to the point that they are no longer treated or seen as a real person’ and contributed thereby to his mother’s death).

824   Justine Pila each? In the absence of property, the legal position is one of freedom to use our and others’ bodies as we decide, subject to the limits mentioned previously. These include the need to respect human dignity and autonomy and to refrain from subjecting any person, with or without her consent, to slavery, servitude, torture, or other inhuman or degrading treatment.62 They also include the need to protect the bodily integrity of all persons, including by refraining from undertaking any health-related intervention without first obtaining her free and informed consent.63 And whether as a standalone provision or requirement of bodily integrity, they include the prohibition against using any body or bodily materials for financial gain.64 In addition to limiting the freedom that people enjoy to exploit their bodies and bodily materials, each of these principles supports the ‘no property’ rule discussed earlier by doubting the capacity of human bodies and the materials they contain to be the objects of transferrable exclusionary rights. More ambivalent is the position regarding corpses and materials lawfully removed from the body of a person who is living or deceased.

5  No Property in Corpses or Detached Bodily Materials as Such Neither the fact of a person dying nor the fact of a body part having a separate existence can trigger transferrable rights where none existed previously. Each can, however, transform the legal subject of a person into the legal object of a thing, paving the way for its protection by property. Despite this, the courts of common law jurisdictions have consistently refused to recognize the general law of personal property as extending to corpses and detached bodily materials beyond the limited extent described previously. The question arises as to why this is so. If corpses and detached bodily materials are capable of supporting transferrable exclusionary rights and do not therefore face the same practical and conceptual hurdles as the recognition of property in living bodies and attached bodily materials face, why have general property principles not been extended to them?

5.1  The Moral Ambivalence of Property The most obvious explanation is morality. Specifically, and as for living bodies and attached bodily materials, protecting corpses and detached bodily materials as objects 62  United Nations Declaration of Human Rights 1948 arts 1, 3 to 5; CHRB arts 1 and 2; CFR arts 1 to 3. 63  CFR art 3; CHRB art 5. 64  CFR art 3; CHRB art 21; Human Tissue Act 2004 s 32; Council of Europe Convention against Trafficking in Human Organs CETS No 126.

Property in Human Body Parts   825 of property would leave certain individuals and groups vulnerable to exploitation, as well as degrading the ideals of moral agency and responsibility associated with humanity. Having said that, since the risks to autonomy and dignity are less acute for corpses and detached bodily materials, especially in the case of nonessential regenerative materials such as blood and urine, and since not recognizing property creates its own moral difficulties, including by limiting access to essential medical treatments requiring the use of others’ bodily materials, there is greater scope for competing moral argument in this context. That scope has been fully explored by commentators in recent years, with many advocating the recognition of weak forms of property at least, comprising noncommercial transferrable exclusionary rights, as a way of reconciling the various interests in corpses and detached bodily materials that exist, and securing something of a moral compromise between the ‘no’ and ‘full’ property positions.

5.2  The Existence of Quasi-Property and Third-Party Property Rights In the meantime, however, the law has continued to progress in other directions by developing alternative common law and statutory forms of protection that are either inconsistent with or diminish the legal space and appetite for recognizing personal property. As the Court of Appeal of England and Wales remarked in R v Kelly,65 echoing a similar statement of the Supreme Court of California in Moore ten years previously, however questionable the historical origins of the [“no property”] principle, it has now been the common law for 150 years at least. . . If that principle is now to be changed, it must be by Parliament, because it has been express or implied in all the subsequent authorities and writings.

Among these ‘subsequent authorities and writings,’ three deserve special mention for their significance. The first is the Human Tissue Act 2004 (UK) and corresponding measures in other states,66 which establishes a regulatory framework for the use and storage of corpses and the removal, use, and storage of material from any living or deceased body, based on individual rights of possession and consent. The second is the IP legislation of all countries based on the Agreement on Trade-Related Aspects of Intellectual Property Rights 1994 (TRIPS Agreement),67 which ensures the availability of property for a range of intangible human productions, including databases, authorial works, and 65  R v Kelly (1999) QB 621 (CA) 630. 66  In the US see, e.g., Uniform Commercial Code § 2–108(1)(c) (making the operation of federal statutory provisions governing commercial transactions subject to state laws governing such matters as ‘the transfer of human blood, blood products, tissues, or parts’); R Hakimian et al., ‘50-State Survey of Laws Regulating the Collection, Storage, and Use of Human Tissue Specimens and Associated Data for Research’ (US Department of Health and Human Services 2004). 67  Compliance with which is a condition of World Trade Organization membership.

826   Justine Pila inventions. And the third is the series of common law cases, including Doodeward (in Australia) and Yearworth (in the UK), which mirror these IP instruments by extending the protections of personal property to tangible human creations and productions incorporating or derived from the same.

6  Quasi-Property Rights of Possession and Consent as Tools to Regulate the Storage and Use of Corpses and Detached Bodily Materials Early common law authorities had established a duty on the part of deceased persons’ executors and relatives to dispose of a corpse and correlative rights for the same to store a corpse for that purpose.68 Beyond such limited rights of possession, however, corpses were either unowned or collective property. The absence of personal property or equivalent rights in respect of them was unproblematic until the introduction of anatomy to medical curricula in the fourteenth century, when there arose a demand for access to corpses for teaching purposes. Almost immediately, a practice of exhuming bodies and offering them for sale to medical schools developed. By the nineteenth century, there existed in England and certain American colonies a highly organized industry of obtaining corpses for sale in this way and even, on occasion, by murdering people. When the details of this industry became known, a public outcry in England and elsewhere triggered legislative action to regulate the use of corpses for teaching purposes, including (in England) via the Anatomy Act 1832. By provision of this Act, any professor, student, or other person wishing to practice anatomy with a corpse could only do so with a licence from the state and permission from an executor or other person in lawful possession of a corpse, which permission could be overridden by contrary expression of the deceased person while alive or the deceased person’s closest surviving relative after death. A person using a corpse in accordance with the Act was also required to bury it within six weeks of its receipt. During the nineteenth and twentieth centuries, the basic principle of the Anatomy Act—that a corpse might validly be stored and used for an authorized scientific purpose subject: (a) to the user being appropriately licensed by the state and complying with public health and other regulatory conditions, and (b) to any contrary wish of the deceased person or her surviving relatives—was retained and adopted in other jurisdictions and contexts, including the conduct of certain postmortem examinations and the use of human tissue in therapy, medical education, or research. In the late 1990s, how68  In the US, see PF Nemeth, ‘Legal Rights and Obligations to a Corpse’ (1943) 19 Notre Dame LR 69. In the United Kingdom see n 65.

Property in Human Body Parts   827 ever, another public scandal involving corpses revealed problems with its negative (‘no objection’) consent requirement.69 This time the scandal involved the retention and disposal by certain hospitals in England of organs and other human tissues removed from children during or following postmortem examinations without the knowledge or consent of their parents. When the practice came to light, public enquiries and consultations were held, and legal proceedings commenced. Among these proceedings was AB v Leeds, involving allegations by certain parents of wrongful interference with their children’s bodies, and negligence by their children’s doctors in the conduct of their fiduciary (doctor–patient) relationships. The wrongful interference case was based on principles that had been developed and accepted in the United States and other jurisdictions previously, and cast the unauthorized retention and disposal of the children’s organs and other bodily materials as an infringement of the parents’ rights to possess their children’s bodies for the purpose of burying them. The negligence case was based on the defendants’ duty to inform the parents, when seeking their consent to examine their children’s bodies, that materials would be removed and would or might be retained and to comply with any wishes of relevance that the parents expressed. While the legal basis for the wrongful interference case was accepted, it was held to afford no relief on the facts. That left the negligence case, on which the parents succeeded. In a judgment clearly sympathetic to their position and sensitive to their distress, the court went some legal distance to uphold their claim. Central was its finding that obtaining their consent to the postmortem examinations was not a purely administrative matter, but rather part of the doctor–patient relationship that had commenced with the doctors’ treatment of the children and continued after the children’s death. By not telling the parents that their children’s materials would be removed and potentially retained, the defendants had breached the duty of care on which this relationship was based, making them liable to the parents in negligence. Like all judicial decisions, the judgment in AB v Leeds reflects the special nature of its facts, including its involvement of children’s bodies and hospital postmortem examinations that required parental consent and served the limited purpose of establishing each child’s cause of death. Bearing these facts in mind, two remarks of the court stand out. The first is that human bodies, whether living or deceased, are appropriately treated ‘with dignity and respect’, and doctors are regarded as ‘the custodians’ of any materials removed from them.70 And the second is that it would be inappropriate—‘peculiarly insensitive’, the judgment states—to suggest that any competition between the possessory rights of parents and doctors in the body of a child not be resolved in favour of the parents.71 In combination, these reflect a prioritizing of the moral value of bodily integrity in its nonindividuated conception over other considerations such as public health and the freedom of doctors to pursue their work unimpeded by parents’ grief.

69  Cf VM Sheach Leith, ‘Consent and Nothing But Consent? The Organ Retention Scandal’ (2007) 29 Sociology Health Illness 1023. 70  AB v Leeds (n 14) [132]. 71  Id. [133].

828   Justine Pila The same prioritizing can be seen to underpin the statutory response to the organ retention scandal in the form of the Human Tissue Act 2004. The effect of this Act at the time of its introduction was to repeal existing legislation governing the use of corpses and bodily materials for teaching and other purposes, including the Human Tissue Act 1961, the Anatomy Act 1984, and the Human Organ Transplants Act 1989, and to establish in its place a detailed regulatory framework governing the storage and use of corpses, and the removal, storage, and use of all material from living and deceased bodies other than gametes, embryos outside the body, and hair and nail from the body of living persons for postmortem, anatomical examination, public display, research, medical treatment (including transplantation), education and training, and other specified purposes. Retaining the two principles of the Anatomy Act in strengthened form, the Human Tissue Act prohibits any person from undertaking any conduct within its scope: (a) without first obtaining ‘appropriate consent’ from the person whose body or bodily materials they are or, if that person has died, from his or her nominated representatives or relatives72; and (b) on a site not licensed specifically in connection with the removal, storage, or use of the relevant materials by a regulatory authority empowered to that end. It also vests that authority with a range of additional powers and responsibilities, including the establishment of codes of practice for such matters as obtaining consent, and the assessment of applications for organ, bone marrow, and peripheral blood stem cell donations from living persons. In the limited space not covered by its provisions, other laws governing the storage and use of corpses and the removal, storage, and use of material from human bodies continue to operate. These include a mix of common law and statutory authorities governing the disposal of bodies following a person’s death73; a statutory regime governing the in vitro creation, storage, and use of human embryos, gametes, and living sperm for treatment and research purposes and the licensing of authorized activities in connection with them74; and private contractual arrangements concerning the removal, use, and possession of hair and nail. In this regulatory context particularly, there has been limited space or motivation to recognize additional rights of personal property beyond the exceptions previously referenced. The framework established by the Human Tissue Act is based on principles of consent and human dignity,75 and has been criticized for prioritizing the interests of individuals and their families over those of third-party patients and the medical and scientific communities and, through its licensing provisions particularly, for stifling research and technological development with excessive bureaucracy.76 By comparison, US law has been regarded as striking a more appropriate balance and as creating a more 72  Cf Human Tissue Act, ss 2 to 7 (explicating the requirement and its exceptions). 73  Dobson v North Tyneside Health Authority [1996] EWCA Civ 1301 (noting the duties of burial conferred at common law and its attendant rights of possession); Public Health (Control of Disease) Act 1984 s 46. 74  Human Fertility and Embryology Act 1990. 75  The Human Tissue Authority commits expressly to undertake its work in compliance with these and two other principles said to underpin the Human Tissue Act: quality, and honesty and openness. 76  AN Warrens et al., ‘Addressing Problems in Transplantation Research in the United Kingdom’ (2012) 93 Transplantation 860; Wicks (n 52).

Property in Human Body Parts   829 research-friendly environment in which the market (via commercial biobanks and private funding arrangements) has played a greater regulatory role than the state.77 An important early authority, which also reflects and has helped set the tone for US regulatory priorities, is the decision of the Californian Supreme Court in Moore v Regents of the University of California. The dispute in Moore arose from the use by a doctor and his colleagues of bodily materials removed from the claimant during treatment for hairy-cell leukemia. The materials, which included extensive amounts of blood, bone marrow aspirate, sperm, and the claimant’s spleen, were used without the claimant’s knowledge to support the doctor’s and others’ research, including their development of commercially lucrative technologies. Upon discovering this, the claimant brought two legal claims similar to the claims in AB v Leeds: breach of the doctor–patient relationship based on the doctor’s failure to have disclosed his research and commercial interests in the claimant’s bodily materials when seeking consent to remove them, and conversion of the claimant’s personal property into the relevant technologies and related products. Moore’s first claim succeeded: the doctor clearly owed him a fiduciary duty and had breached that duty by failing to disclose his personal interest in Moore’s bodily materials and then removing more of those materials than required for his medical treatment. By contrast, the second claim of wrongful interference failed for three (related) reasons. The first was the detrimental impact that allowing it would have had on medical advances by requiring scientists ‘to investigate the consensual pedigree of each human cell sample used in [their] research’.78 The second was the sufficiency of the claimant’s other rights to protect his legitimate autonomy and patient interests. And the third was the inconsistency of his claim to property with the substance and purpose of the detailed federal and state regulations concerning the use and disposal of human bodily materials that existed in California and other parts of the United States. The last of these reasons recalls the statement of the English and Welsh Court of Appeal in R v Kelly earlier, that whatever the arguments of principle in favour of personal property in detached bodily materials, its judicial recognition would be inappropriate given the other laws that exist, including some inconsistent with or premised on the absence of personal property.

7  Property in Human Productions Derived from Corpses and Detached Bodily Materials Of particular importance in Moore itself, and a central aspect of the regulatory context of current relevance, are the laws of IP. Broadly stated, these confer a range of transferrable exclusionary rights in respect of human creations and productions of an expressive or 77  Private funding has been especially important for research involving the use of human embryos owing to US governmental restrictions on the use of federal funding to that end. 78  Moore (n 10), 135. See also Greenberg (n 6), 1076.

830   Justine Pila industrial nature, including those incorporating or derived from bodies and bodily materials, supplemented by the (nonproprietary) moral rights referenced previously. Strictly speaking, IP rights are not exceptions to the ‘no property’ rule discussed earlier, given that the objects to which they attach are, as the court in Moore remarked, factually and legally distinct from the objects to which the ‘no property’ rule applies. In addition, and however legitimate or otherwise it might be, the rule establishing the availability of IP for human creations and productions is of wider application and longer standing existence than its designation as an ‘exception’ might suggest. In the context of patents, for example, it has existed in statutory form since the fifteenth century.79 Abstracted from its specific applications, ‘intellectual property’ denotes the right of persons who exercise skill or labour to produce an expressive or industrial object to property rights in respect of it. Hence, and regardless of their legitimacy,80 there is nothing surprising from an IP perspective about either the decision of Doodeward, that a person who preserved a two-headed stillborn foetus in liquid nitrogen owned personal property in respect of the result (as a new and different product created by transformative human skill or work), or Yearworth, that a person who used his body to generate and ejaculate sperm similarly owned personal property in the sperm (as an existing product reduced by human skill or work to a new and useful state). To use the language of property theory discussed earlier, one might say that by transforming the foetus and reducing the gametes to a new and useful state, the claimants in each case asserted a claim to property that the law elected to recognize; or, if reasoning from the Lockean theory of property, that they applied the (self-owned) labour of their bodies to existing (unowned) objects in a manner sufficient to earn them property.81 The jurisprudence of patent law, especially, offers clear support for the Doodeward and Yearworth principles of personal property in objects incorporating or derived from corpses and detached bodily materials. The reason is the availability of patents for any ‘inventions’, defined in Europe to mean any ‘technical teaching’ in the sense of a causal, perceivable, and repeatable result obtained by utilizing the forces of nature by human action on the physical world.82 The classic statement of this principle comes from the 1969 decision of the German Federal Supreme Court in Rote Taube involving a method of breeding a more beautiful dove.83 In the court’s decision, the method was not an invention, and therefore not patentable, since it could not be repeated indefinitely to 79  The Venetian Republic introduced a patent statute in 1474. 80  Yearworth (n 9) is an especially difficult case owing in part to its support of different legal treatment for male and female reproductive materials respectively; see Holdich (n 2). See further Evans (n 9). 81  See also R v Kelly (1999) QB 621 (applying Doodeward to hold that three human heads, several limbs, and certain other parts of corpses that had been prepared and preserved by surgeons were the property of the Royal College of Surgeons); Bazley (n 9); Re Edwards (n 9). A statutory version of the Doodeward principle is contained in the UK’s Human Tissue Act s 32(a)(c). 82 G2/07 Broccoli and G1/08 Tomatoes [2011] EPOR 27 (EPO Enlarged Board) [6.4.2.1]. 83  Rote Taube Case No X ZB 15/67 (1969 BGH); (1970) 1 IIC 136 (Rote Taube).

Property in Human Body Parts   831 produce the same result. Given this, its contribution to knowledge was merely the once-achieved dove produced by applying the method on one occasion. Were the method repeatable, however, it would have been a ‘technical teaching’ and its result, whether conceived as the steps comprising the human action or the dove, suitable for protection by property. This was despite its involvement of living phenomena since, and as the court remarked, these are subject to the same scientific laws as other phenomena and to the same rules of patentability. Three points of current importance emerge from Rote Taube. The first is that productions in different fields of technology are treated equally. This is a central principle of patent law in all countries,84 and explains the presumptive availability of patents for inventions involving living phenomena, including human biological materials, such as the biologics mentioned previously. The second is that, while a patent may protect a dove or other product as the embodiment of a technical teaching, that teaching will by definition be capable of repeated instantiation in tangible objects, such as living doves, which objects might logically then also be protected by property, consistent with the Yearworth case. And the third is that an act of human creation is not itself required to devise an invention and earn a person patent protection: merely acting on the world to produce a causal, perceivable, and repeatable result will suffice.85 Hence the principle, consistent again with Lockean theory and Yearworth, that applying work or skill to remove something from the state of nature in which it originates and make it available in a new and useful form is enough to merit a person rights of property. This principle has been tested and affirmed expressly by patent authorities in the context of inventions incorporating or derived from human bodily materials since the 1990s. In Howard Florey/Relaxin,86 for example, the European Patent Office (EPO) supported the patentability of an isolated gene encoding the human protein H2-relaxin on the grounds that isolating the gene required a technical process and that the result of isolating it was a product that could be properly characterized by its chemical structure (giving it a discrete and objectively determinable existence). When the European Union introduced its Biotech Directive in 1998, it entrenched this principle by providing expressly that elements isolated from the human body or otherwise produced by a technical process are inventions for which a patent might validly be granted.87 Only recently has the patent community’s commitment to this principle of property in detached bodily materials as technical teachings been revisited and scaled back in some countries. This occurred first with the US Supreme Court’s 2013 decision in Association for Molecular Pathology v Myriad88 that isolated forms of the breast cancer genes BRCA1 and BRCA2 were not inventions owing to their insufficient difference 84  Reproduced also in TRIPS Agreement 1994 art 27(1). 85 Cf Diamond v Chakrabarty 447 US 303, 308 (supporting a view of patentability as including ‘anything under the sun that is made by man’ [emphasis added]). 86  Howard Florey/Relaxin [1995] EPOR 541 (EPO Opposition Division) (Howard Florey/Relaxin). 87  Biotech Directive art 5(2); also art 3(2). 88  Association for Molecular Pathology v Myriad Genetics Inc 569 US 576 (2013) (Myriad).

832   Justine Pila from the genes in their natural forms. For the sequences to be inventions, the court held, their putative inventors would have needed to alter them structurally by removing their introns, for example, to make them human ‘creations’.89 The effect of this decision was to invalidate decades of patents granted by the US Patent Office, supported also by the decision of the Federal Circuit in Myriad, that isolated genes merit protection as the product ‘of human transformation, requiring skill, knowledge, and effort’ to ‘reduc[e] a portion of nature to concrete form’.90 The distinction between unprotected human bodily materials and other products of nature, on the one hand, and protectable objects incorporating or derived from them, on the other, is a fine one. It rests, conceptually and normatively, on the existence of a human act sufficient to transform the former into the latter. That transformation justifies both the recognition of property where none existed previously, and the vesting of that property in the person responsible for the transformation to the exclusion of all others, including the persons whose bodily materials were transformed and any persons needing access to the resulting object on medical or other grounds. As long as the recognition of property depends on a person’s ability to demonstrate a certain connection to an object, as by possessing, using, or producing it, or on a person having applied her labour to a preexisting object, as by reducing the object to a new and useful state, there can be few limits to the type of objects capable of supporting rights of property. In the IP context, this position has been defended with reference to the social value of the inventions and other objects whose creation and dissemination IP rights facilitate, especially in the era of advanced bio- and medical technologies. To the extent that recognizing personal property in the raw materials on which those technologies depend might impede their development, it seems unlikely to garner much support in the current technological and commercial age, notwithstanding the brake on patenting that decisions such as Myriad reflect.91 As discussed earlier, there is a distinction between protecting bodies and bodily materials as such and inventions and other (tangible and intangible) products incorporating or derived from them. But the distinction can be very fine indeed, and, in some cases, the availability of property for objects declared to fall in the latter category is difficult to reconcile with the provision in Article 21 of the Convention on Human Rights and Biomedicine (CHRB), reproduced as an aspect of the EU Charter of Fundamental Rights Article 3 right of bodily integrity, that ‘[t]he human body and its parts shall not, as such, give rise to financial gain’—the ‘as such’ proviso notwithstanding. It also raises the same moral questions as recognizing property in respect of bodies and bodily materials raises.

89  Id. 594–595. 90  Association for Molecular Pathology v United States Patent and Trademark Office 689 F 3d 1303 (CA Fed 2012) 1331–1332. 91  On the scientific impact of Myriad, see J Liddicoat et al., ‘Continental Drift? Do European Clinical Genetic Testing Laboratories Have a Patent Problem?’ (2019) 27 Eur J Human Genetics 997.

Property in Human Body Parts   833

8  The Moral Limits of (Intellectual) Property The argument for IP, including patent law particularly, depends largely on its role in supporting the development and disclosure of new and useful technologies with the potential to address global health, environmental, and other pressing social issues. Given this, it is unsurprising that when the first patent applications for isolated genes and other biotech products were filed, patent offices and commentators adopted a ‘public interest’ frame to assess their legitimacy. In Howard Florey/Relaxin, for example, the EPO was clear not only that the DNA sequence was an invention and therefore patentable, but that granting the patentees broad protection was ‘perfectly justified . . . in view of the fact that H2-relaxin has been made available to the public for the first time’.92 Nonetheless, a striking feature of early biotech patenting cases such as Howard Florey/ Relaxin is the discussion they contain regarding the morality of patents for objects derived from living phenomena, including human bodily materials particularly. That discussion reflects an early effect of biotech patenting in having challenged the economic orientation of traditional public interest conceptions of patent property (as existing to cure the market failure created by the ability to reproduce resource-intensive inventions at zero cost), including by introducing new values such as human dignity to be weighed in the balance when considering their social utility. In Howard Florey/Relaxin, the morality of the patent was raised in opposition to its grant in reliance on a long-standing but rarely invoked exception to patentability for inventions the commercialization of which is contrary to ordre public or morality.93 According to the arguments advanced, using human tissue from a pregnant woman to devise an invention for protection by patent grant would offend the moral values of dignity and autonomy for two reasons. First, it would treat life as property and exploit the female condition of pregnancy for a technical process oriented toward profit. Second, it would support the dismemberment of women and their piecemeal sale to commercial firms around the world. In dismissing these arguments, the EPO emphasized the limited scope of the ordre public/morality exception to patentability, and the inappropriateness of patent authorities considering ‘fundamental ethical questions’ when deciding whether or not to grant a patent.94 On its view of the substance of the objections themselves: since the women had consented to their tissue being used ‘within the framework of necessary gynaecological operations’ and ‘human tissue or other material, such as blood, bone etc., [had] been widely used for many years as a source for useful products’, including many life-saving substances, there could be no basis for regarding the isolation of human genetic material as immoral. Indeed, the use of bodily materials for other 92  Howard Florey/Relaxin (n 86) [5.3]. 93  EPC art 53(a); Biotech Directive art 6(1); see also TRIPS Agreement art 27(2). 94  Howard Florey/Relaxin (n 86) [6.5].

834   Justine Pila purposes was explicitly approved in Article 13 of the then Draft Bioethics Convention of the Council of Europe, subject only to satisfying appropriate information and consent procedures. In addition, since the DNA fragment could be chemically synthesized, there would be no need to obtain further human tissue to reproduce the invention. Nor would granting the patent support slavery or the dismemberment of women: to suggest otherwise ‘betray[ed] a fundamental misunderstanding of the effects of a patent’, which do not confer rights to human beings at all, but rather rights to ‘exclude for a limited period of time third parties from commercially using the patented invention’. And, finally, regarding the argument that granting the patent would allow the patenting of ‘life’: ‘DNA is not “life,” ’ the EPO said, ‘but a chemical substance which carries genetic information and can be used as an intermediate in the production of proteins which may be medically useful’.95 At the time of its decision, Howard Florey/Relaxin offered a rare window into the previously closed world of patent-granting practice. In so doing it revealed the nature of patent offices and other specialist users of the system as an epistemic community able to exploit their interpretive custody of patent law to exclude ‘outsider’ perspectives and limit what qualifies as an issue of patentability to reflect their particular (pro-technology and pro-patenting) commitments. At the same time, the decision revealed the effect of modern biotechnology in motivating environmental and other public advocacy groups to exploit such doctrinal mechanisms as existed to gain entry to the system and have their moral concerns regarding those commitments heard, even if not always fully or properly addressed. While the opposition to the patent in Howard Florey/Relaxin failed, political action around biotech patenting continued, nowhere more so than in the European Union. There it culminated in two legal developments of especial importance. The first was the Biotech Directive of 1998, which supplemented the expansive conception of inventions as technical teachings described earlier with a restatement and explication of the traditional ordre public/morality exclusion for the biotech context with explicit reference to human dignity.96 And the second was the decision of the Court of Justice of the European Union in Brüstle, applying the Biotech Directive, that any invention the derivation of which involves the destruction of a fertilized ovum capable of developing into a human being is immoral on human dignity grounds and unable in consequence to 95  Id. [6.3.1–6.3.4]. 96  See Biotech Directive art 6 (reproducing EPC art 53(a) and declaring the following subject matter to be unpatentable on its basis: ‘(a) processes for cloning human beings; (b) processes for modifying the germ line genetic identity of human beings; (c) uses of human embryos for industrial or commercial purposes; and (d) processes for modifying the genetic identity of animals which are likely to cause them suffering without any substantial medical benefit to man or animal, and also animals resulting from such processes’); Recital 38 (‘Whereas the operative part of this Directive should also include an illustrative list of inventions excluded from patentability so as to provide national courts and patent offices with a general guide to interpreting the reference to ordre public and morality; whereas this list obviously cannot presume to be exhaustive; whereas processes, the use of which offend against human dignity, such as processes to produce chimeras from germ cells or totipotent cells of humans and animals, are obviously also excluded from patentability’).

Property in Human Body Parts   835 support a patent. In a clear reversal of the earlier Howard Florey/Relaxin approach, the Advocate General in this case spoke to the dangers of granting property for new technologies incorporating or derived from human bodily materials even where, as on the Brüstle facts: the bodily materials would otherwise have gone to waste, fully informed consent from the materials’ donor had been obtained, and the claimants had no intention to commercialize the property. In so doing he invoked the ‘slippery slope’ argument against the protection of bodies and bodily materials by even the most limited species of property, suggesting, in effect, that once the gate is opened, there will no containing the forces of the free market. Hence the spectre raised in his opinion of war prisoners being dismembered in the pursuit of profit through patenting.97 However alarmist the language of Brüstle might seem, in the context of wider debates about body property particularly, it serves as a sobering reminder of the trajectory of modern capitalism. As argued by Karl Polanyi more than 70 years ago,98 since fully selfregulated market economies require the commodification of nature and human beings, their pursuit can only push human societies to the edge of the precipice. Modern biotechnology, and in particular biotech patenting, offers support for this argument. Not only does patent law encourage and reward the use of technology to transform people into objects of use, it completes the transformation by recognizing those objects as replaceable and measurable in the manner of other commodities, and plays a central role while doing so in the construction of modern life and culture through the interpretation of such central patent law concepts as ‘invention’, ‘discovery’, and even ‘human embryo’. As studies by others have shown, patent law also supports bio-piracy and the practice among Western researchers and pharmaceutical companies of exploiting the vulnerability of indigenous and developing communities to access their genetic and other biological resources for use in the development of commercially lucrative technologies. If IP in general and patent law in particular offers a cautionary tale to those who would extend the reach of property even further than it exists currently, it also demonstrates the malleability of the ‘property’ concept, and offers some ideas for how it might be used to manage the risks of doing so.99 In IP, more than in other areas of property, balancing the benefits of protection against their risks and costs is a central aim and a central reason for the existence of tailored statutory (rather than common law) property regimes. In the name of achieving such a ‘balance’, for example, patents require application to and registration by the state, and are conditional upon an applicant’s ability to demonstrate that the invention for which the patent is sought is new, inventive, susceptible 97  Compare the allegations of organ harvesting from Falun Gong and other prisoners of conscience in China and the many resolutions and declarations by parliaments and governments around the world in response to them in recent years, including House Resolution 343 by the US House of Representatives in June 2016 and Written Declaration 0048 by the European Parliament in July 2016. See https://endtransplantabuse.org (last visited 10 October 2019). 98  K Polanyi, The Great Transformation: The Political and Economic Origins of Our Time (1944, republished by Beacon Press 2001) especially 136–137. 99  Phillips v Mulcaire [2012] UKSC 28 (‘[Definitions of] “intellectual property” . . . are not particularly helpful because . . . “there is no universal definition of the term, which is no doubt why Parliament has adopted a variety of definitions for different situations” ’).

836   Justine Pila of industrial application, and clearly and sufficiently described in the patent application. It is also in the name of ‘balance’ that certain inventions are ineligible for patent protection, including methods of medical treatment, plant and animal varieties, and inventions the commercialization of which would be contrary to morality or public policy; that the rights conferred by a patent do not extend to use of the invention and expire after 20 years; and that the persons who devise inventions, whether or not they are also the patentees, have moral rights of the type described previously, including rights to be recognized as having devised their inventions and share in any profits that patents granted for their inventions generate. And finally, it is in the name of ‘balance’ that the freedom of patentees to exploit their property is limited by competition laws and other patent-specific mechanisms, such as compulsory licensing, and that states are required by international and regional measures to support ‘the fair and equitable sharing’ of any (patent-derived or other) benefits arising from the use of other states’ genetic resources when devising technologies.100 In these and other ways, the property conferred by patents is limited and supplemented by certain rights analogous to those of reputation and bodily integrity aimed at recognizing and protecting the interests of persons beyond the patentees, including the donors of any bodily materials used to devise a patented invention, the persons responsible for devising it, and any third parties needing access to it for essential medical or other reasons.

9 Conclusion In recent decades, advances in digital and biotechnologies have increased exponentially the practical value of bodily materials by unlocking their genetic (informational and therapeutic) properties. Inevitably, new industries have developed and existing industries have been adapted to exploit these developments for a variety of scientific and commercial ends. These include developing databases of genetic materials and information to support research about the identity and health of individuals and their relations, and devising diagnostic tests, medicines, and other technologies. The social value of these is immense, as is the scope for profiting commercially from them. As long as that scope exists, people can be expected to devote substantial resources to exploiting it by obtaining access to genetic material and deriving commercially valuable information and technologies from it, and by charging others for access to that information and those technologies. In each of these endeavours, property is engaged and has a central role to play.

100  See Convention on Biological Diversity 1992, Articles 15, 8(j) and 16(5), and the 2010 Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from Their Utilization (Nagoya Protocol), implemented in the EU by Regulation (EU) No 511/2014 [2014] OJ L 150/59.

Property in Human Body Parts   837 Whether the law should recognize personal property in respect of detached bodily materials, particularly for the benefit of those whose materials they are, or even for others, is a question that has attracted much attention in recent years. While the immediate effect of such recognition may indeed be increased legal certainty and a firmer basis for individuals and communities to share in any profits that the use of their genetic and other bodily materials generates, it would also have other less visible and immediate effects that any consideration of this issue, in whatever jurisdiction, must properly accommodate. Aside from the risks to human dignity mentioned previously, these include contributing to the current culture of consumerism that pervades modern capitalist societies by further entrenching and extending the role of individual choice in the current regulatory framework governing the use of bodies and bodily materials.101 In this context particularly, utilitarian thinking and the thirst for new technological and commercial opportunities must be tempered by careful moral reasoning.

Acknowledgments I am grateful to the editors for their comments on an earlier draft.

Bibliography GJ Annas, MA Grodin eds., The Nazi Doctors and the Nuremberg Code (Oxford University Press 1992). L Bennett Moses, N Gollan, ‘ “Thin” Property and Controversial Subject Matter: Yanner v Eaton and Property Rights in Human Tissue and Embryos’ (2013) 21 J L Med 1. L Bennett Moses, ‘The Problem with Alternatives: The Importance of Property Law in Regulating Excised Human Tissue and In Vitro Human Embryos’ in I Goold et al. eds., Persons, Parts and Property: How Should we Regulate Human Tissue in the 21st Century? (Bloomsbury 2014) ch 12. J Black, ‘Regulation as Facilitation: Negotiating the Genetic Revolution’ (1998) 61 MLR 621. W Boulier, ‘Sperm, Spleens, and Other Valuables: The Need to Recognize Property Rights in Human Body Parts’ (1995) 23 Hofstra LR 693. D Beyleveld, R Brownsword, ‘Human Dignity, Human Rights and Human Genetics’ (1998) 61 MLR 661. P Calandrillo, ‘Cash for Kidneys? Utilizing Incentives to End America’s Organ Shortage’ (2004) 13 George Mason LR 69. D Dickenson, Property in the Body: Feminist Perspectives (Cambridge University Press 2009). P Drahos, ‘Biotechnology Patents, Markets and Morality’ (1999) 21 EIPR 441. P Drahos, ‘Regulating Patent Offices: Countering Pharmaceutical Hegemony’ (2008) 5 SCRIPTed 501. JB Frank, ‘Body Snatching’ A Grave Medical Problem’ (1976) 49 Yale J Biol Med 399. 101  Z Bauman, ‘Viewpoint: Sociology and Postmodernity’ 36 Sociological Rev 800; T Murphy, ‘The Texture of Reproductive Choice: Law, Ethnography, and Reproductive Technologies’ in T Murphy ed., New Technologies and Human Rights (Oxford University Press 2009) ch 6.

838   Justine Pila M Frunza et al, ‘Could Body Parts be Considered Property? Bioethical, Anthropological and Metaphysical Issues’ (2013) 58 Studia UBB Philosophia 125. DM Gitter, ‘Ownership of Human Tissue: A Proposal for Federal Recognition of Human Research Participants’ Property Rights in their Biological Material’ (2004) 61 Washington and Lee LR 257. I Goold, Flesh and Blood: Owning our Bodies and their Parts (Bloomsbury 2016). I Goold, ‘Why Does It Matter How We Regulate the Use of Human Body Parts?’ (2014) 40 JME 3. M Goodwin, ‘Human Rights and Human Tissue: The Case of Sperm as Property’ in R Brownsword et al. eds., The Oxford Handbook of Law, Regulation, and Technology (Oxford University Press 2017) ch 8. K Greasley, ‘Property Rights in the Human Body: Commodification and Objectification’ in Goold et al. eds., Persons, Parts and Property: How Should we Regulate Human Tissue in the 21st Century? (Bloomsbury 2014) ch 7. C Hawes, ‘Property Interests in Body Parts; Yearworth v North Bristol NHS Trust’ (2010) 73 MLR 130 J Herring, P-L Chau, ‘Interconnected, Inhabited and Insecure: Why Bodies Should Not be Property’ (2014) 40 JME 39. AM Honoré, ‘Ownership’ in A Guest ed., Oxford Essays in Jurisprudence (Oxford University Press 1961) 107. J Kent, Regenerating Bodies: Tissue and Cell Therapies in the 21st Century (Routledge 2012). I McLean, J Poulton, ‘Good Blood, Bad Blood, and the Market: The Gift Relationship Revisited’ (1986) 6 J Public Policy 431. TW Merrill, HE Smith, ‘What Happened to Property in Law and Economics?’ (2001) 111 Yale LJ 357. A Mossoff, ‘Locke’s Labor Lost’ (2002) 9 U Chicago L School Roundtable 155. S Moyn, ‘The Secret History of Constitutional Dignity’ in C McCrudden ed., Understanding Human Dignity (Oxford University Press/British Academy 2014) ch 4. SR Munzer, A Theory of Property (Cambridge University Press 1990) 41–56. SR Munzer, ‘An Uneasy Case against Property Rights in Body Parts’ (1994) 11 Social Philosophy & Policy 259. D Nicol et al., ‘Impressions on the Body, Property, and Research’ in Goold et al. eds., Persons, Parts and Property: How Should we Regulate Human Tissue in the 21st Century? (Bloomsbury 2014) ch 2. J Penner, ‘The Bundle of Rights Picture of Property’ (1996) 43 UCLA LR 711. A Pottage, ‘The Inscription of Life in Law: Genes, Patents, and Bio-Politics’ (1998) 61 MLR 740. N Priaulx, ‘Managing Novel Reproductive Injuries in the Law of Tort: The Curious Case of Destroyed Sperm’ (2010) 17 Eur J Health L 81. N Ram, ‘DNA by the Entirety’ (2015) 115 Columbia LR 873. C Rose, Property and Persuasion: Essays on the History, Theory and Rhetoric of Ownership (Routledge 1994). M Rosen, ‘Dignity: The Case Against’ in C McCrudden ed., Understanding Human Dignity (Oxford University Press 2013) ch 7. D Resnik, ‘The Morality of Human Gene Patents’ (1997) 7 Kennedy Institute Ethics J 43. PE Schaafsma, ‘An Economic Overview of Patents’ (1997) 79 J Patent & Trademark Office Soc 241.

Property in Human Body Parts   839 SV Shiffrin, ‘Lockean Arguments for Private Intellectual Property’ in SR Munzer ed., New Essays in the Legal and Political Theory of Property (Cambridge University Press 2001) ch 5. JC Stedman, ‘Invention and Public Policy’ (1947) 12 Law Contemporary Problems 649. C Stewart et al., ‘The Law of Gifts, Conditional Donation and Biobanking’ (2013) 21 J L Med 351. C Stewart et al., ‘The Problems of Biobanking the Law of Gifts’ in Goold et al. eds., Persons, Parts and Property: How Should we Regulate Human Tissue in the 21st Century? (Bloomsbury 2014) ch 3. RM Titmuss, The Gift Relationship: From Human Blood to Social Policy (Allen and Unwin 1970).

chapter 39

Orga n Tr a nspl a n tation Joaquin Cayon-­De Las Cuevas and David Orentlicher

1  Introduction: Shortage of Organs for Transplantation It is commonplace to say that the success of transplantation therapies has led inexorably to their generalisation. Far from the experimental nature of the first interventions carried out in the middle of the twentieth century, new technologies have produced an increase in survival and a substantial improvement in the quality of life. This new scenario has made transplants a therapeutic alternative that is increasingly demanded and used. Indeed organ transplantations may mean the difference between life and death for hundreds of thousands of people worldwide. Transplants are often the most cost-­ effective treatment, and sometimes the only way to treat liver, lung, or heart failure. However, it should not be forgotten that the transplantation process1 is different from other therapies in a very important way—there is a persistent and large gap between the need for transplants and the availability of organs. The shortage of organs affects all countries. Official figures show that there is no country where the availability of organs is sufficient to meet existing demand. In the United States, more than 100,000 people are on a waiting list for an organ, but only about 40,000 transplants were performed in 2020. In the case of the European Union, approximately 34,000 transplants were carried out in 2019, while 10 patients died every day waiting for a transplant and nearly 60,000 patients 1  In this chapter we consider transplantation as a whole process. According to the CoE law, the term ‘transplantation’ covers ‘the complete process of removal of an organ or tissue from one person and implantation of that organ or tissue into another person, including all procedures for preparation, preservation and storage’ [Article 2(4) of the Additional Protocol concerning Transplantation of Organs and Tissues of Human Origin (2002)].

842   Joaquin Cayon-­De Las Cuevas and David Orentlicher remained on the waiting list at the end of the year.2 The existence of the organ shortage is inevitable, absent breakthroughs in synthetic organs or xenotransplantation. The shortage of organs is compounded by enormous variability in the donation and transplant activity of different countries. The access to transplant therapies for patients varies depending on where they are in the world. These disparities in access reflect factors that are highly complex and sensitive, including legal (type of legislation and consent systems in place in the country), organisational (performance of national transplant programmes and teams), and cultural (such as the awareness in the general population, the health literacy, or religious beliefs). Thus, for example, organ transplants are mostly carried out in countries that have reached a certain degree of development and have promoted the implementation of the systems that make transplantation possible. Still, as indicated, the shortage of organs is a global problem that affects all countries, both those that have implemented effective donation and transplant systems and those that still lack them. Accordingly, in the regulation of organ transplantation, considerable attention is paid to strategies for increasing the organ supply and policies for allocating the organs that are available.

2  Legal Governance of Organ Transplantation The legal regulation of organ transplantation in the United States and Europe has important similarities and differences. In terms of similarity, the two jurisdictions share a dual kind of system. In Europe, some law is set at the supranational level and other law at the national level. In the United States, some law is established at the national level and other law at the state level. In terms of difference, even at the supranational level, European oversight is divided. Both the European Union (EU) and the Council of Europe (CoE) exercise authority. In contrast, American oversight resides in the single national ­government. In addition, while national regulation in the United States is binding and enforceable in all states, supranational regulation in Europe is largely advisory to the Member States such that the Member States have to legislate and enforce their own laws for EU and CoE guidelines to take effect. Europe and the United States also differ in how they divide regulatory authority between public and private actors. European countries rely more on the law to set standards while the United States relies more on professional setting of standards.

2  Data from OPTN ­-Report 6 March 2020 (39.718)- and ONT, which is in charge of the World Transplant Registry. Available at: http://www.ont.es/Documents/07.09.2020%20NPONT%20 RegistroMundial_.pdf. See also Newsletter Transplant Vol. 25, 2020. Available at: http://www.ont.es/publicaciones/ Documents/NEWSLETTER%202020_baja.pdf

Organ Transplantation   843

2.1  European Governance There are two key aspects to European transplantation law. First, while there are important efforts to establish common policies at the supranational level, the ‘margin of appreciation’ for national authorities leaves much of transplantation law to the discretion of national authorities. Thus, for example, whether to employ an ‘opt-­in’, actual consent system or an ‘opt-­out’, presumed consent system for posthumous organ donation is decided country by country. Second, within the category of policies set at the supranational level, authority is further divided between the EU’s legal system, which is much more focused on quality and safety transplantation standards, and the CoE’s system, which is fundamentally based on a human rights approach and addresses questions such as equitable access to transplants and the problem of organ trafficking. The margin of appreciation for national authority plays a critical role in organ transplantation law. The EU’s deference to national authorities follows from the ­provisions of the Treaty on the Functioning of the European Union (TFEU), which only allows EU actions on some public health matters, keeping significant national competence in the healthcare area. In this regard, under Article 168(7), Union action shall respect the responsibilities of the Member States for the definition of their health policy and for the organisation and delivery of health services and medical care. The responsibilities of the Member States shall include the management of health services and medical care and the allocation of the resources assigned to them.

This key provision has been considered a subsidiarity rule with regard to EU intervention in the field of health care.3 In fact, it explains why EU transplantation law must allow a certain margin of discretion to Member States. For this reason, EU regulation on organs and tissues does not provide a comprehensive framework as such. Indeed, the Organ Transplantation Directive and the Tissue Directive—which share similar structure, content, and principles—do not provide the whole picture of the EU situation since a significant part of the regulation is left to national states. Even in areas where EU regulation is in place, there is a significant role for the Member States. In the particular field of organs and human materials, Article 168(4)(a) TFEU allows the European Parliament and Council to adopt ‘measures setting high standards of quality and safety of organs and substances of human origin, blood and

3  As Guy and Sauter underline ‘it would therefore appear that “subsidiarity” in the context of Article 168(7) TFEU is concerned with the “systemic” focus of healthcare in light of the references to “organization and delivery of health services and medical care”. This “systemic” focus can be distinguished from an “individual” focus, which broadly encompasses the implications of the internal market rules for patients, healthcare professionals and healthcare products, as well as human rights aspects in respect of access to healthcare’ [Mary Guy and Wolf Sauter, ‘The History and Scope of EU Health Law and Policy’ (2016). CCP Working Paper 16-­02] http://dx.doi.org/10.2139/ssrn.2718045.

844   Joaquin Cayon-­De Las Cuevas and David Orentlicher blood derivatives’, but ‘these measures shall not prevent any Member State from maintaining or introducing more stringent protective measures’. As mentioned, supranational regulation also flows from the CoE, and the key legal document is the Council’s 1997 Biomedicine (Oviedo) Convention.4 While the Convention is directly binding on the State Parties, it is not possible to invoke its infringement before the European Court of Human Rights (ECtHR). Article 29 only allows the possibility of requesting an advisory opinion from the ECtHR on legal questions concerning the interpretation of the Convention, with the States explicitly required to ‘provide appropriate judicial protection to prevent or to put a stop to an unlawful infringement of the rights and principles’.5 This reliance on Member States for enforcement parallels the EU legal framework. For example, the EU issues its fundamental rules in the form of Directives: the Directive on standards of quality and safety of human organs intended for transplantation—hereinafter, Organ Transplantation Directive6—and the Directive on standards of quality and safety for tissues and cells—Tissues Directive.7 As with Directives generally, these Directives establish a type of norm that obliges Member States as regards the intended result but not as regards the means for its implementation.8 In the field of transplantation, the Biomedicine Convention has been developed through an Additional Protocol concerning Transplantation of Organs and Tissues of Human Origin (2002),9 which has made it possible to establish a certain European code on organ donation and transplantation. The aim of this Protocol is to protect human dignity and integrity as well as rights and fundamental freedoms in the face of scientific 4  Officially ‘Convention for the protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine’ (ETS No.164). 5  The Convention on Human Rights and Biomedicine has been described as the best current example of how to promote the protection of human rights in the biomedical field at a transnational level. Roberto Andorno, ‘The Oviedo Convention: A European Legal Framework at the Intersection of Human Rights and Health Law’ (2005) 2(5) J Intern Biotechn L 133. 6  Directive 2010/45/EU. The Organ Transplantation Directive has been implemented by the Commission by Directive 2012/25 laying down information procedures for the exchange, between Member States, of human organs intended for transplantation and other related acts. See Commission Staff Working Document on the mid-­term review of the ‘Action Plan on Organ Donation and Transplantation (2009–2015): Strengthened Cooperation between Member States’ (SWD(2014) 147 final of 25.4.2014) and Report from the Commission to the European Parliament and the Council on the exercise of the power to adopt delegated acts conferred on the Commission pursuant to Directive 2010/53/EU of the European Parliament and of the Council of 7 July 2010 on the standards of quality and safety of human organs intended for transplantation (COM(2015) 123 final of 10.3.2015). 7  Directive 2004/23/EC 8  By contrast with regulations which are binding in their entirety and directly applicable in all Member States, directives are only binding, ‘as to the result to be achieved, upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods’ (Article 288 of the Treaty of Functioning of the European Union). 9  The Additional Protocol has been signed by 22 European countries and ratified by 15 of them (Bulgaria, Croatia, Estonia, Finland, Georgia, Hungary, Iceland, Montenegro, North Macedonia, Portugal, Republic of Moldava, Romania, Slovenia, Spain, and Switzerland). It entered into force on 1 May 2006.

Organ Transplantation   845 and medical advances. The Additional Protocol contains general principles and specific provisions regarding the transplantation of organs and tissues of human origin for therapeutic purposes. In particular, the general principles set out in the Additional Protocol include equitable access to transplantation services for patients, transparent rules for organ allocation, health and safety standards, the prohibition of financial gain by donors, and the need for donors, recipients, health professionals, and the public to be properly informed. Finally, specific provisions cover the removal of organs from living and deceased persons, the use made of the organs and tissues removed, the prohibition of financial gain, confidentiality, and sanctions and compensation. Also at the CoE level, an important treaty must be taken into account: the Convention Against Trafficking in Human Organs of 2015, which calls on governments to establish as a criminal offence the illegal removal of human organs from living or deceased donors in certain cases. Furthermore, the Convention also provides protection measures and compensation for victims as well as prevention measures to ensure transparency and equitable access to transplantation services. Regarding cross-­border transplantation, EU Member States may conclude, or allow a competent authority to conclude, agreements with European organ exchange organisations provided that such organisations ensure compliance with the requirements laid down in the Organ Transplantation Directive. This entails delegating to those organisations, inter alia: (a) the performance of activities provided for under the framework for quality and safety and (b) specific tasks in relation to the exchanges of organs to and from Member States and third countries.10 At the transnational level, Eurotransplant is the largest nonprofit organisation, and it is responsible for a patient-­oriented allocation of donor organs in Austria, Belgium, Croatia, Germany, Hungary, Luxembourg, the Netherlands, and Slovenia, serving a total population of around 137 million people.11 Another transnational organisation is Scandiatransplant, which includes Denmark, Finland, Iceland, Norway, Sweden, and Estonia, covering a population of about 28.8 million inhabitants. Most of the organ exchange organisations have been created at the national level, and there is no organisation that brings together all European countries.

2.2  US Governance As in Europe, regulation of organ transplantation in the United States is dual in nature, with some law set at the national level and other law at the state level. Sometimes the national and state rules address different matters; at other times, they address the same matters, allowing for enforcement by either federal or state regulators and sometimes both.12 For example, if someone pays for an organ, the purchaser and seller are subject 10  Article 16 of the Organ Transplantation Directive. 11  Officially called ‘Eurotransplant International Foundation’, the organisation was created by Jon J van Rood in 1967 and legally founded on 12 May 1969. 12  Federal law generally allows for charges at both levels, while states vary in whether they allow charges that duplicate federal charges.

846   Joaquin Cayon-­De Las Cuevas and David Orentlicher to federal and state charges. Similarly, if a transplant program violates standards of care, federal or state health agencies can bring enforcement actions. In some cases, individuals may be able to bring tort claims. Overall, though, legal sanctions are uncommon. As with medical practice generally in the United States, organ donation was originally governed primarily by state law, partly by applying general legal rules and partly by enacting laws specific to organ transplantation. As mentioned, states ban the sale of organs, and they also have established procedures for people or members of their family to consent to posthumous organ donation. For examples of applying general principles, just as patients have to consent before physicians can perform an appendectomy or other surgery, so must they consent before physicians remove bone marrow or a kidney for transplantation. As a corollary, people cannot be forced to donate organs, even when necessary to save another person’s life.13 Thus, just as a pregnant patient must be able to choose an abortion when continuing the pregnancy would pose a health risk, so must anyone be able to decline organ donation because of the health risks to donors. Another important feature of medical practice being governed by state law is the reliance on the medical profession for the setting of standards, and that also is largely the case for organ transplantation. For example, principles of professional liability prevent physicians from subjecting patients to undue risk, and courts generally will defer to professional standards of practice in deciding when the level of risk is acceptable. Thus, experts in cardiology and cardiac surgery, rather than legislators, decided when advances in medicine made bypass surgery or stent insertion a reasonable option to treat coronary artery disease. Similarly, organ transplant surgeons employing their professional expertise decided when it was sufficiently safe to perform live kidney or partial-­liver donations. Starting in 1984 with the National Organ Transplant Act (NOTA), the federal government assumed a substantial role in organ transplantation, holding transplant programs to national standards in important ways. Even then, the government has deferred to physicians working with other transplant professionals and members of the public to establish the standards. Most notably, NOTA created the Organ Procurement and Transplantation Network (OPTN) as a private, nonprofit entity to determine standards for organ transplantation, including rules for the allocation of organs among those in need of a transplant.14 As a condition of participation in the Medicare and Medicaid programs, transplant programs must be members of the OPTN and adhere to its standards.15 Some states also require OPTN membership as a condition for licensing organ transplant programs. Since its creation, the OPTN has contracted with the nonprofit United Network for Organ Sharing (UNOS) to carry out its responsibilities. UNOS is a membership organisation that includes the 58 organ procurement organisations throughout the country, as well as more than 300 transplant surgery centers, medical laboratories that perform tests for organ matching, volunteer and advocacy groups, and members of the general public. The organ procurement organisations are 13  McFall v Shimp, 10 Pa. D. & C. 3d 90 (Allegheny County 1978). 14  42 USC. §274(b). 15  42 USC. 1320b–8(a)(1)(b).

Organ Transplantation   847 responsible for recovering organs in their geographic area and allocating the organs to patients in need. Some of the procurement organisations cross state boundaries (e.g., the New England Organ Bank), some follow a state’s boundaries (e.g., Alabama Organ Center), and some serve part of a single state (e.g., OneLegacy in the Los Angeles area). Perhaps the most important element of NOTA was to establish a national structure for organ transplantation. The OPTN issues guidelines on a number of features of organ transplantation, including steps to be taken in screening live donors and obtaining their informed consent to donation. In addition, the OPTN determines the rules for allocating organs among those waiting for a transplant. As mentioned, NOTA also imposed a ban on the sale of organs used in transplantation, with a broad definition of ‘human organ’ that includes not only hearts, livers, and kidneys, but also bone marrow, corneas, bone, and skin. However, the ban on sales does not apply to blood, sperm, or ova, nor does it apply to retrieval of adult stem cells from circulating blood. While people cannot be paid for their organs, they can receive reimbursement to protect them from economic loss. Thus, it is permissible to cover their medical, travel, or housing expenses, as well as lost wages.16 While the enactment of NOTA created a major federal role in organ transplantation, it still left much of the field to the states. In particular, state law governs the retrieval of organs from deceased persons for donation. Following the provisions of the Uniform Anatomical Gift Act (UAGA), which was first issued in 1968 and has been updated multiple times, states protect the right of individuals to agree to the retrieval of organs after their death and the right of family members to authorise retrieval in the absence of a decision by the deceased person.

3  Organ and Tissue Living Donation 3.1 Introduction Because living organ donors assume a significant risk to their own health for the benefit of another person, it is essential that donors make a truly voluntary decision, not, for example, because of family pressures or the prospect of financial gain. Accordingly, in both Europe and the United States, payments for organs are prohibited. In addition, transplant programs take special steps to protect against undue pressure. For example, after prospective donors undergo screening to see whether they are medically qualified to donate, their physicians will offer them the opportunity to change their mind and let the intended recipient and family be told that the prospective donor was medically disqualified.

16  42 USC. §274e(c)(2).

848   Joaquin Cayon-­De Las Cuevas and David Orentlicher

3.2  Overview: Europe Living donation coexists with removal from deceased persons in most Member States of the EU. However, it should be noted that both types of removal are not in a situation of parity since living donation is subsidiary to the removal from deceased persons. Removing organs or tissues from living donors for implantation purposes may carry some risk for that donor. For this reason, the principle of subsidiarity is embedded in Article 9 of the Biomedicine Convention’s Additional Protocol when requiring there should not be any suitable organ or tissue available from a deceased person. This provision mentions two additional conditions for living donation. First, removal of organs or tissue from a living person can only be carried out for the therapeutic benefit of the recipient. Second, there should not be any alternative therapeutic method of comparable effectiveness. Accordingly, if the results of a living donor transplantation are expected to be significantly better than those expected utilising a graft removed from a deceased person, live donation may be the preferred therapeutic option for a particular recipient.17 In addition, dialysis treatment is not considered to provide results in terms of the patient’s quality of life comparable with those obtained by a kidney transplant. Traditionally, living donation used to be carried out for the benefit of persons with a biological link to the recipient. Nevertheless, it has evolved over the years in such a way that good results can be obtained even where there is no genetic relationship between donor and recipient. Concerning this issue, Article 10 of the Additional Protocol covers both options, although it does not mention biological ties but ‘close personal relationships’ as defined by law. Such a personal relationship between donor and recipient is a broad notion which depends on cultural values and might include members of the recipient’s immediate family, parents, brothers, sisters, spouses or long-­standing partners, godparents, close personal friends, etc. The Protocol also allows the donation in favour of recipients without such a close personal relationship, but only under the conditions defined by law and with the approval of an appropriate independent body. It is important to be aware that the independent body required by Article 10 is not the same as the official body identified in Article 13 before which the living donor can give his or her consent. However, as the Explanatory Report clarifies, national law allows for the independent body provided by Article 10 to be the same as the competent body identified in Article 13.18 Finally, it should be noted that Article 10, as a provision for potential organ donors, does not apply for living donation of tissues since the therapeutic interests of a recipient who may not be known at the time of removal have to be taken into account.19

17  Explanatory Report, paragraphs 60–61. 18  Explanatory Report, paragraph 65. 19  Explanatory Report, paragraph 66.

Organ Transplantation   849

3.3  Overview: United States As in Europe, the risks of live donation are cause for concern. Similarly, there are concerns about donor motivation. When the donation occurs within the family, there are concerns about coercion; when the donation is extrafamilial, there are concerns about hidden payments or psychological stability of the donor. Nevertheless, transplant centers generally welcome donation of kidneys within families. Extrafamilial donations also are becoming more common, and studies suggest that transplant centers can screen potential donors to ensure that their desire to donate does not reflect psychopathology. Similar issues are raised with regard to donations of a partial liver, which can be done since the liver can grow new tissue after it has been divided. But the concerns are accentuated since the risks to the donor from the surgery are greater than with kidney donation. Consequently, there is greater reluctance among transplant centers to accept donors who do not have a close relationship with the recipient. Data on long-­term effects of partial liver donation are starting to emerge and are reassuring so far. But it is still too early to draw firm conclusions. Partial lung transplants are not commonly performed in the United States but are more important in countries that do not rely as much on deceased donors. Ethical concerns are heightened because two donors are needed for each recipient. While legal regulation of live donor transplants does not speak to the relationship between donor and recipient, it does include measures to minimise donor risk and enhance the informed consent process. For example, OPTN regulations require organ transplant programs to meet quality standards, perform medical and psychosocial evaluations of the potential donor, and take special steps to ensure that the donor is fully informed and making a truly voluntary choice. The psychosocial evaluation must be performed by a ‘psychiatrist, psychologist, masters prepared social worker, or licensed clinical social worker’ and must include evaluation for • mental health issues, • ‘behaviors that may increase the risk for disease transmission’, • smoking, alcohol, or drug use, • potential donor understanding of the medical and psychosocial risks of donation, and • by exploring the reasons for donation and the relationship of the potential donor to the recipient, the possibility of inducement, coercion, or other undue pressures.20

20  OPTN Policy 14.1.A.

850   Joaquin Cayon-­De Las Cuevas and David Orentlicher

3.4  Donor Capacity to Consent 3.4.1 Europe Following the wording of Article 20 of the Biomedicine Convention, the Additional Protocol prohibits organ or tissue removal when carried out on a person who does not have the capacity to consent. Nevertheless, as an important exception, regenerative tissue such as bone marrow can be removed from a person who does not have the capacity to consent under certain conditions: (1) there is no compatible donor available who has the capacity to consent, (2) the recipient is a brother or sister of the donor, (3) the donation has the potential to be life-­saving for the recipient, (4) the authorisation of his or her representative or an authority or a person or body provided for by law has been given specifically and in writing and with the approval of the competent body, and (5) the potential donor concerned does not object.21 For cell removal, Member States need not apply conditions (1) and (3) as it is established that their removal only implies minimal risk and minimal burden for the donor.22 Capacity to consent typically is recognised at adulthood, but, at the European level, there is not a general provision requiring a certain age since this question is left to national systems. In most of the European countries, only 18-­year-­old persons with full mental capacity may act as a competent living donor. Therefore, neither minors nor persons with a mental disorder can be considered as potential donors. Nevertheless, in the United Kingdom, children can be considered as living organ donors in extremely rare circumstances. The 2004 Human Tissue Act23 defines a child as being under 18 years old. If a clinician intends to consider a child as a living organ donor, they are advised to discuss the case with the Human Tissue Authority at the earliest opportunity. In ac­cord­ ance with common law and the Children Act 1989—which established the legal basis for who has parental responsibility—court approval should be obtained before the removal of a solid organ or part organ from a child for donation.24

3.4.2  United States In the United States, living donation is generally limited to adults (age 18 older) who possess decision-­making capacity. But there is greater willingness than in Europe to allow ‘donation’ by individuals who lack decision-­making capacity, either because of age or intellectual disability. Traditionally, eligibility for donors without decision-­making capacity was determined by state law, and while there has been disagreement on the matter, most courts 21  Article 14 of the Additional Protocol. 22  Article 15 of the Additional Protocol. 23  Human Tissue Act 15 November 2004 has been partially amended by Organ Donation (Deemed Consent) Act 15 March 2019, concerning consent to activities done for the purpose of transplantation and for connected purposes. 24  Code of Practice F (Donation of solid organs and tissue for transplantation. Part. 1-­Living organ donation) published on 20 May 2020. Available at: https://www.hta.gov.uk/sites/default/files/HTA%20 Code%20of%20Practice%20F%20Part%20One%20-%20Living%20Organ%20Donation_0.pdf

Organ Transplantation   851 that have decided the issue have allowed retrieval of a kidney or bone marrow on the grounds that the risk to the donor is low, the benefit to the recipient great, and there was a close emotional relationship between the donor and the recipient (these cases typically involve transplants between siblings). In the absence of a close emotional relationship, courts have been more likely to reject a donation. OPTN policy has narrowed the possibility of donation from individuals who lack decision-­making capacity. For donations other than bone marrow, donation is not permitted for individuals who are ‘both less than 18 years old and mentally incapable of making an informed decision’.25

3.5  Informed Consent 3.5.1 Europe As regards the informed consent to be given by the donor, general rules for informed consent apply here. This means that in order to be valid, consent must be preceded by sufficient information. Therefore, an organ or tissue may be removed from a living donor only after the person concerned has given free, informed, and specific consent to it either in written form or before an official body.26 It should be highlighted that this consent from the living donor is stricter than the general one from a regular patient since it must be specific and provided in a formal way because of its importance. Of course, the donor may freely withdraw consent at any time. In order to avoid undue pressure on the donor, the Explanatory Report to the Additional Protocol explains that donors should be assured that they can refuse to donate or withdraw their consent at any time in complete confidence. To that end, donors should be interviewed in private and helped to cope with the consequences of their decisions.27 In contrast with the granting of consent, there is no requirement for withdrawal of consent to be in writing or to follow any particular form. The donor need simply say no to the removal at any time, even if a procedure performed under local anaesthetic has commenced. Article 14 allows the same protection to donors of regenerative tissue lacking capacity to consent to their removal. However, professional standards and obligations may require that the team continue with the procedure if not to do so would seriously endanger the health of the donor.28 One may wonder what happens if the donor seeks to withdraw consent to the agreed implantation after removal. In this regard, CoE rules only apply for removal so that there is no European common solution for this issue, which has to be solved at the domestic level. Regarding the required information, the donor and, where appropriate, the person or body providing authorisation, should beforehand be given appropriate information as 25  OPTN Policy 14.4.D. 26  Article 13 of the Additional Protocol. 27  Explanatory Report, paragraph 73. 28  Explanatory Report, paragraph 77.

852   Joaquin Cayon-­De Las Cuevas and David Orentlicher to the purpose and nature of the removal as well as on its consequences and risks.29 In this sense, these four classic informative points that already appear in the Biomedicine Convention for patients in general are repeated.30 However, the Protocol adds that donors should also be informed of the rights and safeguards prescribed by law for the protection of the donor. In particular, they shall be informed of the right to have access to independent advice about such risks by a health professional having appropriate experience and who is not involved in the organ or tissue removal or subsequent transplantation procedures.31 This source of information, who may be a doctor or other suitably qualified healthcare worker, must be independent of the team or teams involved in the transplant. However, that person must have appropriate experience with the risks associated with donation and transplantation to be able to give proper advice. This advice can be requested by donors if they wish.32

3.5.2  United States As in Europe, the general requirement of informed consent to a medical procedure applies under state law, with OPTN guidelines and some state laws requiring a more specific and detailed informed consent process.33 Under the OPTN guidelines, potential donors must give written consent confirming that they are willing to donate, that they are giving consent free from inducement and coercion, and they have been informed of their right to decline to donate at any time. To ensure free and willing consent, the OPTN requires the transplant program to employ a confidential evaluation process and provide the potential donor with an Independent Living Donor Advocate who functions independently from the intended recipient’s transplant team and who is an advocate for the rights of the potential donor. The program also must disclose (a) alternative treatment options for the intended recipient, including deceased donor transplantation; (b) the possibility that the donor’s medical evaluation will identify conditions that have to be reported to public health authorities; and (c) the surgical, medical, psychosocial, and financial risks associated with living donation, many of which are specified in the OPTN policy. In addition, potential donors must be informed that it is a federal law crime to accept ‘anything of value’ for an organ donation, including ‘cash, property, and vacations’.34

3.6  Protection of Living Donors 3.6.1  Europe The Organ Transplantation Directive aims to ensure that living donors should be adequately evaluated to determine their suitability for donation in order to minimise the 29  Article 12 of the Additional Protocol (par. 1). 30  Article 5 states that ‘This person [the patient] shall beforehand be given appropriate information as to the purpose and nature of the intervention as well as on its consequences and risks’. 31  Article 12 of the Additional Protocol (par. 2). 32  Explanatory Report, paragraph 72. 33  See, e.g., N.Y. Comp. Codes R. & Regs. tit. 10, § 405.30 (b)(7)-(8). 34  OPTN Policy 14.3.

Organ Transplantation   853 risk of transmission of diseases to the recipients. In addition, living donors face risks linked both to testing to ascertain their suitability as a donor and to the procedure to obtain the organ. Complications may be medical, surgical, social, financial, or psychological. The level of risk depends, in particular, on the type of organ to be donated. Therefore, living donations need to be performed in a manner that minimises the physical, psychological, and social risk to the individual donor and the recipient and does not jeopardise the public’s trust in the healthcare community.35 For this reasons, four different duties of Member States are laid down in order to protect living donors36: 1. To take all necessary measures to ensure the highest possible protection of living donors in order to fully guarantee the quality and safety of organs for transplantation. 2. To ensure that living donors are selected on the basis of their health and medical history by suitably qualified or trained and competent professionals. Such assessments may provide for the exclusion of persons whose donation could present unacceptable health risks. 3. To ensure that a register or record of the living donors is kept, in accordance with Union and national provisions on the protection of the personal data and statistical confidentiality. 4. To endeavour to carry out the follow-­up of living donors. Member States should have a system in place, in accordance with national provisions, to identify, report, and manage any event potentially relating to the quality and safety of the donated organ, and hence of the safety of the recipient, as well as any serious adverse reaction in the living donor that may result from the donation. These legal duties are complemented by an additional one foreseen in the Protocol when it is stated that, before organ or tissue removal, appropriate medical investigations and interventions must be carried out to evaluate and reduce physical and psychological risks to the health of the donor. In this sense, the removal may not be carried out if there is a serious risk to the life or health of the donor.37 One may distinguish four possible players who may assess a serious risk: the donor, the recipient, the medical team, or the independent authority. According to the Explanatory Report, the decision about the risk is a matter for the transplant medical team looking after the donor or the body authorising the donation. In judging the risks involved, the donor’s interests must take precedence, although in some circumstances the balance of risk to the donor compared to potential benefit to the recipient may be taken into consideration. The donation being acceptable or not depends not just on the physical risk associated with the procedure but must include psychological factors.38 35  Recital 23 of the Organ Transplantation Directive. 36  Article 15 of the Organ Transplantation Directive. 37  Article 11 of the Additional Protocol. 38  Explanatory report, paragraph 69, mentions an example of psychological harm. It is the case of the donor who develops an undue sense of ownership towards the recipient or the recipient who feels unduly obligated to the donor. If, following full assessment, the medical team looking after the donor

854   Joaquin Cayon-­De Las Cuevas and David Orentlicher Last but not least, appropriate medical follow-­up should be offered to living donors after transplantation.39

3.6.2  United States As in Europe, a number of steps are taken to protect donors from the risks of transplantation. As mentioned earlier, OPTN policy prescribes a comprehensive psychosocial evaluation and requires transplant programs to assign an independent advocate for the donor. Per OPTN policy, transplant programs also must perform a comprehensive medical evaluation of potential donors. This evaluation includes assessment of the donor’s own medical history, family medical history, employment, housing, health insurance status, and social support system, as well as performance of a physical exam, diagnostic testing, and screening for transmissible diseases and cancer. Kidney and liver donors also must undergo specific testing and screening to assess the status of their kidney or liver health.40 After the medical evaluation, transplant programs may decline donation if the potential donor has a ‘condition that, in the hospital’s medical judgment, causes the donor to be unsuitable for organ donation’. In addition, programs must decline donation for patients with cancer, a high suspicion of coercion or an illegal payment, or uncontrolled psychiatric conditions. Potential kidney donors must be excluded if they have diabetes, uncontrolled hypertension, or organ damage from a history of hypertension. Liver donors must be excluded on the basis of several factors relating to the health status of their liver, including active hepatitis B or C infection.41

4  Organ and Tissue Removal from Deceased Persons 4.1 Europe Organ and tissue removal from deceased persons should not be properly called ‘donation’ since such legal category requires a living donor. This is why the Additional Protocol unifies under the name of ‘removal’ both the living donation and deceased ‘donation’. In this way, according to Article 2 the term ‘removal’ refers to ‘removal for the purposes of implantation’.42

judge there to be a significant risk of death or long-­term severe disability to the donor, the donation procedure should not go ahead. 39  Article 7 of the Additional Protocol. 40  UNOS Policy 14.4A-­C. 41  UNOS Policy 14.4.D. 42  As an exception Article 20 also regulates implantation of an organ or tissue removed for a purpose other than donation for implantation.

Organ Transplantation   855 Removal from deceased persons faces two fundamental key questions. First, when should a person be considered deceased? Second, what kind of consent or authorisation is required in order to transplant the organ or tissue? In relation to the first issue, the Additional Protocol lays down that organs or tissues should not be removed from the body of a deceased person unless that person has been certified dead in accordance with the law. For this purpose, the Protocol establishes a basic guarantee of independence under which doctors certifying the death of a person shall not be the same doctors who participate directly in removal of organs or tissues from the deceased person, or subsequent transplantation procedures, or having responsibilities for the care of potential organ or tissue recipients.43 Concerning consent or authorisation, it must be underlined that this is a key point for the success of any transplantation system. In some countries, it is assumed there is a presumed consent from the deceased person to allow removal so that they should express their opposition via the country’s register of objections or an advanced directive (opt-­ out system). This is the case of Spain, as the world’s leading country in donation rate. According to Article 5(3) of Spanish Law on organ removal and transplantation,44 ‘presumed healthy persons who die in an accident or as a result of an accident shall also be considered as donors, if there is no express objection from the deceased’. Very recently, the United Kingdom has also moved to an opt-­out system through the 2019 Organ Donation Act concerning consent to activities done for the purpose of transplantation and for connected purposes,45 which amends the 2004 Human Tissue Act.46 Other European countries with a presumed consent system are Austria, Belgium, Switzerland, Portugal, Czech Republic, Finland, France, Greece, Hungary, Poland, Slovak Republic, Slovenia, and Sweden. By contrast, in other European countries, the law does not prejudge the wishes of those concerned so that an explicit consent from deceased persons or their relatives and friends is required (opt-­in system). This is the case for Germany, Ireland, Denmark, Estonia, Iceland, Italy, Latvia, Luxembourg, and the Netherlands.47 Opt-­out systems have been criticised for being ineffective and not significantly increasing transplantation rates. Some scholars consider there is scant evidence that presumed consent will be effective. Therefore, rather than legislating for consent of donors, the system should be addressing the misgivings and misunderstandings of families to improve their consent. However opt-­in systems might also be criticised since they do not take into account the public interest in having more organs available. One 43  Article 16 of the Additional Protocol. 44  Law 30/1979, 27 October 1979, on organ removal and transplantation. 45  Popularly known as ‘Max and Keira's Law’, in honor of Kiera Ball, a nine-­year-­old girl who died in a car accident and whose heart was donated to Max Johnson, another nine-­year-­old. 46  Wales took the first step to becoming the first country in the UK to move away from an opt-­in system; see Andreas Albertsen ‘Deemed Consent: Assessing the New Opt-­Out Approach to Organ Procurement in Wales’ (2018) 44(5) J Med Ethics 314. 47  Adam Arshad, Benjamin Anderson, and Adnan Shari ‘Comparison of Organ Donation and Transplantation Rates Between Opt-­Out and Opt-­In Systems’ (2019) 95(6) Kidney Int 1453.

856   Joaquin Cayon-­De Las Cuevas and David Orentlicher may wonder if the principle of autonomy applies for deceased people when dealing with transplantation. In this regard, it can be argued that in such cases the principle of solidarity should displace the principle of decision-­ making autonomy of family members. In any case regarding the impact of the consent system, recent findings suggest no significant gain for established opt-­in countries considering a switch to opt-­out.48 In this sense, it is very likely that Spain’s success can be explained not only by its presumed consent system but also by its investment in infrastructure and equipment, its integrated transplant organisation, and the cultural factors in favour of organ removal from deceased persons.49 From the legal point of view, the Organ Transplantation Directive is without prejudice to the broad diversity of the systems of consent already in place in the EU Member States.50 Article 14 simply sets out that the procurement of organs should be carried out only after all requirements relating to consent, authorisation, or absence of any objection in force in the Member State concerned have been met. In an analogous way, the Biomedicine Convention’s Additional Protocol leaves a wide margin to States Parties to decide the type of consent or authorisation required. In fact, Article 17 merely states that ‘organs or tissues shall not be removed from the body of a deceased person unless consent or authorisation required by law has been obtained’. In this regard, according to Article 8, States should provide information on consent or authorisation, in particular with regard to removal from deceased persons. The provision is therefore silent on whether consent can be expressed or implied, leaving it to national law.51 In any case, the Additional Protocol seems to be aware that whether explicit or implicit, State Parties should take all appropriate measures to promote the donation of organs and tissues.52 The promotion of donation appears as a conventional duty for European countries. In this regard, as the Explanatory Report reminds us, ‘appropriate’ measures are not defined but should include the provisions on 48  Although historically some countries have observed impressive increases after introduction of presumed consent, such as Belgium, others have fared badly with either no difference or an actual drop in organ donation rates. See Arshad, Anderson, and Shari (n 52). 49  The same opinion can be found in Adnan Sharif, ‘Viva España—Lessons from the Spanish Organ Donation System’ (2017)17 Am J Transplant 1964. Rafael Matesanzet et al., ‘About the Opt-­Out System, Live Transplantation, and Information to the Public on Organ Donation in Spain. . . Y olé!’ 17 Am J Transplant 1965. 50  Recital 21. 51  As Den Exter observes, ‘under certain provisions of the Charter of Fundamental Rights Charter’s right to healthcare (Art. 35), and other fundamental rights, such as the right to life (Art. 2), the prohibition of inhuman or degrading treatment (Art. 4), private life (Art. 7), and non-­discrimination (Art. 21), one may argue that these rights require Member States to take all necessary steps to increase the rate of available donors, including adjustment of the consent model’ (André Den Exter, ‘Regulating Organ Donation and Transplantation Under European Union Law: A Success Story?’ (2017) 1(2) J Trafficking Hum Exploitation 129, at 137). Unfortunately, we are afraid this ‘expansive’ interpretation concerning the application of the Charter on the transplantation consent model is far from being adopted by the EU Court of Justice. 52  Article 19 of the Protocol.

Organ Transplantation   857 information to be provided to health professionals and to the public, the need to set up a transplant system, and to have recognised means of giving consent or authorisation.53 Whatever the domestic system, if the wishes of the deceased are not sufficiently established, the team in charge of the removal of organs must beforehand endeavour to obtain testimony from relatives of the deceased. Unless national law otherwise provides, such authorisation should not depend on the preferences of the close relatives themselves for or against organ and tissue donation. Close relatives should be asked only about the deceased persons expressed or presumed wishes. It is the expressed views of the potential donor which are paramount in deciding whether organs or tissue may be retrieved. Parties should make clear whether organ or tissue retrieval can take place if a deceased person’s wishes are not known and cannot be ascertained from relatives or friends.54 In this regard, it is important to point out that the ECtHR has increasingly strengthened the position of closest relatives by explicitly recognising certain rights to set out their views before an appropriate decision is adopted. In Petrova v. Latvia (2014),55 the lack of information and explanation on how and when these rights might be exercised is considered a violation of the relatives’ right to respect for private and family life, as laid down in Article 8 of the European Convention of Human Rights, even though an opt-­ out model is applicable at the national level.56 Six months later in Elberte v. Latvia (2015), the Strasbourg Court found in particular that, although Latvian law set out the legal framework allowing the closest relatives to express consent or refusal to tissue removal, it had not clearly defined the scope of the corresponding obligation or discretion left to experts or other authorities in this respect. The Court also noted that the relevant European and international documents on this matter gave particular importance to establishing the relatives’ views through reasonable enquiries. Finally, it stressed that adequate mechanisms should have been in place to balance the relatives’ right to express their wishes against the experts’ broad discretion to decide on such matters.57 Of course, the Court did not recognise a general right to object to removal of a deceased relative’s organs or tissues, but these Strasbourg decisions might be employed as a precedent in order to de facto introduce familial consent as a special requirement for such a removal.58 53  Explanatory Report, Paragraph 106. 54  Explanatory Report, Paragraph 102. 55  ECtHR judgment 24 June 2014 (Application no. 4605/05). 56  ECtHR judgment 13 January 2015 (Application no. 61243/08). 57  For further analysis of these two rulings, see Sara Fovargue and Lindsey Claire Hogg ‘Mine, Yours, Ours? Autonomy and the Removal and Use of Organs and Tissues before the European Court in Petrova v Latvia and Elberte v Latvia’ (2017) 1(2) J Trafficking Hum Exploitation 143. 58  In the words of Dove et al., ‘what concerns us most is that this decision may serve as precedent to future claims—even in opt in jurisdictions with no mention in domestic law of relatives and consent— of an individualist autonomous human right of a close relative to express authoritative judgement over the possible removal of a deceased relative’s bodily tissue or organs. If opt in jurisdictions already allow in practice in some instances relatives to override even the express wishes of the deceased, how far removed are we from a court finding that, even if not articulated expressly in law, through a country’s permissive actions, they are estopped from denying the relative’s stand-­alone so-­called right to consent?’(Edward S Dove et al., ‘Elberte v Latvia: Whose Tissue Is It Anyway: Relational Autonomy or the Autonomy of Relations?’ (2015) 15(2–3) Med L Intl 77, at 95.

858   Joaquin Cayon-­De Las Cuevas and David Orentlicher Finally, Article 17 of the Protocol does establish a precautionary rule prohibiting the removal if the deceased person had objected to it. This leads to the possibility for a person to decide the disposal of his or her organs and tissues through advance directives.59 In this regard, according to the Explanatory Report, if a person has made known their wishes for giving or denying consent during their lifetime, these wishes should be respected after the person’s death. If there is an official facility for recording these wishes and a person has registered consent to donation, such consent should prevail: removal should go ahead if it is possible. By the same token, it may not proceed if the person is known to have objected. Nonetheless, consultation of an official register of last wishes is valid only in respect of the persons entered in it. Nor may it be considered the only way of ascertaining the deceased person’s wishes unless their registration is compulsory.60 There are sometimes potential conflicts between advance directives and organ donation. Most patients who write anti-­treatment advance directives or otherwise express anti-­treatment desires are also sympathetic to organ donation. The problem with this situation is that, although patients do not realise it, the chance of the patient’s organs being useful can often be increased if certain medical interventions occur when the patient is in the dying process. Sometimes these interventions may increase the time it takes to die.61

4.2  United States Legal rules for posthumous donation have focused on a few issues—obtaining consent for the donation, ensuring that organs can be safety transplanted, and increasing the number of people who agree to donate. Some of these matters are addressed by federal law through OPTN guidelines, and state law also plays a substantial role. As to the latter, there is some variation from state to state, but states typically adopt the provisions of the UAGA, which was initially issued in 1968, revised in 1987 and 2006, and last amended in 2009. By 2020, 46 states had adopted the 2006 version, though it is common for states to modify some of the provisions during the legislative process. To protect the deceased donor (decedent) from the influence of conflicts of interest, the OPTN guidelines prohibit organ recovery or transplantation by either the decedent’s treating physician at the time of death or the physician who declared the time of the decedent’s death.62 To protect potential recipients of the organs from transmission of 59  Under the title ‘previously expressed wishes’, Article 9 of Biomedicine Convention sets out that ‘the previously expressed wishes relating to a medical intervention by a patient who is not, at the time of the intervention, in a state to express his or her wishes shall be taken into account’. 60  Explanatory Report, Paragraph 99. 61  Robert M Veatch, ‘The Conflict Between Advance Directives and Organ Donation: A New Problem in End-­of-­Life Planning’ (2015) 6(2) JAHR—Eur J Bioethics 267, at 269. The author suggests two potential solutions: an easy one (an amendment to an advance directive), and another more complex (the use of surrogates to override advance directives) at 270–272. 62  OPTN Policy 2.14.A.

Organ Transplantation   859 cancer or infectious disease or other health risks, organ procurement organisations must review the decedent’s medical record and as much of the medical and behavioral history that they can obtain and perform testing of decedents and their organs to assess health status. With regard to obtaining consent to donation, every state follows the opt-­in, actual consent model. Under all versions of the UAGA, competent adults may make a gift of their organs, with the gift to take effect upon their deaths. The gift may be for education, research, transplantation, or other therapy. Family members are not permitted to override the decedent’s gift, though organ procurement organisations often are reluctant to remove organs when the family objects. For individuals who have not made a gift and who have not expressed opposition to the use of their organs, family members may authorise the taking of their organs.63 The 1987 UAGA added a limited role for presumed consent. Under the 1987 version, medical examiners may authorise the taking of organs following an autopsy if they do not find any objection after reasonable efforts to ascertain the person’s or family’s wishes. However, the 2006 Revised UAGA eliminated the ability of medical examiners to authorise organ retrieval in the absence of consent.64 While courts initially viewed the medical examiner provisions favorably, more recent court decisions questioned the constitutionality of that authority.65 The later courts did not reject presumed consent entirely, but they did indicate a need for efforts to ensure that objections to donation were identified. The drafters of the 2006 UAGA decided to simply remove the medical examiner provisions, and the states generally followed suit, though some retain their medical examiner provisions for cornea transplants. The 2006 UAGA also expanded the list of persons who may consent to organ donation after a person’s death to include adult grandchildren, adults who ‘exhibited special care and concern for the decedent’, and health care agents.66 Because of the persistent shortage of organs and the failure of many individuals or families to authorise posthumous donation, there have been a number of proposals to increase the donation rate. Some of these proposals were incorporated in the 1987 UAGA, including ‘routine inquiry’ (hospitals asking adult patients whether they want to be organ donors after death) and ‘routine request’ (hospitals also discussing with family members their ability to authorise the taking of organs from deceased patients). With the adoption of routine request in federal Medicare and Medicaid rules, the 2006 UAGA and most states eliminated their routine inquiry and routine request provisions. Under the federal rules, as a condition of participation in the Medicare and Medicaid programs, hospitals must notify their designated organ procurement organisation (OPO) 63  The UAGA does allow for parental override of decisions by unemancipated minors. Rev. Unif. Anatomical Gift Act § 8(g)-(h) (2006). 64  Rev. Unif. Anatomical Gift Act § 23 (2006). 65 Compare Georgia Lions Eye Bank, Inc. v Lavant, 335 S.E.2d 127 (Ga. 1985), and State v Powell, 497 So. 2d 1188 (Fla. 1986), with Brotherton v Cleveland, 923 F.2d 477 (6th Cir. 1991), and Newman v Sathyavaglswaran, 287 F.3d 786 (9th Cir. 2002). 66  Rev. Unif. Anatomical Gift Act at § 9(a).

860   Joaquin Cayon-­De Las Cuevas and David Orentlicher about patients whose death is imminent or who have died and also must collaborate with the OPO to ensure that ‘the family of each potential donor is informed of its options to donate organs, tissues, or eyes or to decline to donate’. Under the federal rule, the OPO ‘determines medical suitability for organ donation’.67 Another policy commonly adopted by states is ‘mandated choice’, under which individuals are asked to express their wishes about posthumous donation. Typically, this occurs when people obtain or renew a driver’s license. Supporters of mandated choice observe that it places the decision in the hands of the individual rather than in the hands of family members. Critics worry that mandated choice usually is implemented without the prospective donor receiving much information, thus preventing genuine informed consent and causing many people to refuse consent. While there has been an increase in the percentage of people who have registered as organ donors, mandated choice does not appear to have increased the number of organ transplants.68

5 Allocation 5.1 Europe Organ allocation is a sensitive issue as a result of the imbalance between supply and demand for organs and tissues.69 The principle of equity of access is the fundamental rule in this matter. This principle can be deduced from the Biomedicine Convention,70 but it also has an explicit mention in the Additional Protocol.71 Thus, the European parties must guarantee that a system exists to provide equitable access to transplantation services for patients. The Protocol’s concern is to ensure that there is no unjustified discrimination against any person within the jurisdiction of a State Party who might benefit from a transplant. Given that there is a severe shortage of most organs and some of the tissues to be transplanted, they should be allocated to maximise the equality of opportunity for patients, taking into account objective medical criteria. Therefore, the allocation system should be as far as possible patient-­oriented. 67  42 CFR § 482.45. 68  Paula Chatterjee et al., ‘The Effect of State Policies on Organ Donation and Transplantation in the United States’ (2015) 175 JAMA Intern Med 1323. 69  Some criticism has been made. For Garwood-­Gowers and Pereira, the normative tendency of transplant-­related discourse to talk about shortage in those cases where demand for organs and tissues outstrip supply should be questioned, given doubts whether transplantation is an appropriate therapy and concerns that, even if it is, we are overemphasizing supply of materials for it and paying insufficient attention to reducing demand. See Austen Garwood-­Gowers and André Pereira, ‘Organ and Tissue Human-­Human Transplantation’, in A den Exter (ed.) European Health Law (Maklu 2017) 391, at 394. 70  Article 3 proclaims the right to an equitable access to healthcare so that ‘Parties, taking into account health needs and available resources, shall take appropriate measures with a view to providing, within their jurisdiction, equitable access to health care of appropriate quality’. 71  Article 3 of the Additional Protocol.

Organ Transplantation   861 It is important to emphasise that organs and tissues should be allocated according to medical criteria. As the Explanatory Report points out, this notion should be understood in its broadest sense, in the light of the relevant professional standards and obligations, extending to any circumstance capable of influencing the state of the patient’s health, the quality of the transplanted material, or the outcome of the transplant. Examples would be the compatibility of the organ or tissue with the recipient, medical urgency, the transportation time for the organ, the time spent on the waiting list, particular difficulty in finding an appropriate organ for certain patients, and the expected transplantation result.72 For ensuring equitable access, waiting lists appear as a necessary tool. In this regard organs and, where appropriate, tissues must be allocated only among patients on an official waiting list, in conformity with transparent, objective, and duly justified rules according to medical criteria. Nevertheless, this requirement does not properly apply when the transplantation of organs removed from a living donor takes place generally between persons having a close personal relationship. By contrast, organs removed from deceased persons should only be allocated to patients registered on an official waiting list. Patients may be registered only on one official transplant list that may have a regional, national, or international scope so as not to prejudice the chances of others. However, this principle does not preclude a system where a patient is registered on a local waiting which is part of a national waiting list.73 There is no European-­wide waiting list because health systems are defined at the national level. Certain guarantees must be provided to the recipients of transplants. First, they must be provided with prior information. In this respect, in terms similar to those that we have already seen for the living donor, the recipient and, where appropriate, the person or body providing authorisation for the implantation should beforehand be given appropriate information as to the purpose and nature of the implantation and its consequences and risks, as well as on the alternatives to the intervention.74 Second, the recipient’s health and safety must be protected. This is why all professionals involved in organ or tissue transplantation should take all reasonable measures to minimise the risks of transmission of any disease to the recipient and avoid any action which might affect the suitability of an organ or tissue for implantation.75 Finally, appropriate medical follow­up should be offered to recipients after transplantation.76 Organ exchange between countries is one of the ways to alleviate the organ shortage. In case of international organ exchange arrangements, the Additional Protocol requires that procedures must also ensure justified, effective distribution across the

72  Explanatory Report, paragraph 37. 73  Recommendation Rec (2001) 5 of the Committee of Ministers to Member States on the management of organ transplant waiting lists and waiting times. Adopted by the Committee of Ministers on 7 March at the 744th meeting of the Ministries’ deputies. 74  Article 5 of the Additional Protocol. 75  Article 6 of the Additional Protocol. 76  Article 7 of the Additional Protocol.

862   Joaquin Cayon-­De Las Cuevas and David Orentlicher participating countries in a manner that takes into account the principle of solidarity within each country.77 Although this invocation of the principle of solidarity is very generic, at the EU level, special rules have been adopted in the field of organ exchange with third countries.78 In this regard, Member States should ensure that organ exchange with third countries is supervised by the competent authority. For this purpose, the competent authority and European organ exchange organisations may conclude agreements with counterparts in third countries. Nevertheless, organ exchange will only be allowed where the organs can be traced from the donor to the recipient, and vice versa, and meet quality and safety requirements equivalent to those laid down in the Organ Transplantation Directive. The Implementing Directive 2012/25/EU develops this issue by laying down information procedures for the exchange, between Member States, of human organs intended for transplantation, in three fields: procedures for the transmission of information on organ and donor characterisation, procedures for the transmission of the necessary information to ensure the traceability of organs, and procedures for ensuring the reporting of serious adverse events and reactions.

5.2  United States Three key questions impact whether individuals receive a transplant: whether they get on a waitlist, how they are prioritised on a waitlist, and whether they can afford the transplant and post-­transplant care (including the availability of public and private insurance). The relevant law for this issue is the 1984 NOTA. As mentioned earlier, NOTA created the OPTN as a private, nonprofit entity to oversee the recovery and ­allocation of organs for transplantation, and the OPTN has contracted with UNOS to operate the OPTN. OPTN issues allocation guidelines for the 58 organ procurement organisations to follow, and it revises the allocation criteria regularly to improve the fairness of the guidelines. While the guidelines differ for hearts, livers, kidneys, and other organs, they generally rely on criteria such as the likelihood of a successful transplant, time spent on the waiting list, and medical urgency (i.e., is the person likely to die or suffer irreparable injury if a transplant is not performed soon). Under the OPTN policy for liver transplantation, for example, the greatest weights are given to medical urgency and the length of time that the recipient will benefit from the transplant. Medical urgency comes into play through the criteria for the highest status on the waitlist (Status 1A for adults79) and the Model for End-­Stage Liver Disease 77  Article 3 of the Additional Protocol. Nevertheless, Tretyakov has critisised this principle, arguing that it is oriented more toward collective, rather than individual welfare. See Konstantin Tretyakov, ‘A Critique of National Solidarity in Transnational Organ Sharing in Europe’ (2017) 5(1) J L Biosci 1. 78  Article 20 of the Organ Transplantation Directive. 79  There are separate waiting lists for adult and pediatric patients. Children have a Status 1A similar to that for adults, plus a Status 1B.

Organ Transplantation   863 (MELD) score ranking.80 Patients do not qualify for Status 1A unless their life expectancy without a transplant is less than 7 days, and sicker patients have higher ranking MELD scores. The length of time that the recipient will benefit from the transplant is reflected in two ways. First, and more importantly, there is consideration of the patient’s life expectancy with a transplant. Adult patients qualify for Status 1A only if they suffer from ‘fulminant’ liver failure. These patients, whose liver failure has usually come on suddenly, typically live longer with a liver transplant than patients with chronic, or long-­ standing, liver failure. If the liver failure has existed for only a short time, then almost all of its harm to the body can be reversed with a liver transplant. When the liver failure has been long-­standing, then much of its damage to the body is permanent and cannot be reversed. As a result, the patient might die even though the new liver is functioning well.81 Length of benefit from the transplant is also reflected in consideration of the likelihood that the liver will be rejected by the recipient’s immune system. Under the OPTN Policy, points are given for blood type similarity since rejection of the transplant is less likely when blood types are compatible. For allocation of kidneys, UNOS assigns different relative weights. The most important difference is the diminished role of medical urgency. Urgency plays a much smaller role in kidney allocation decisions because of the availability of dialysis. Length of benefit and equity both are important aspects of kidney allocation. For example, a key factor is matching kidneys and recipients based on their expected longevity. Donor kidneys that are expected to function the longest are given to patients who are expected to have the longest survival after transplant. The goal is to avoid the problem of less healthy recipients dying with transplants that still function well and healthier recipients needing a second transplant. This matching applies to 20% of kidneys. For the other 80%, waiting time is the most important consideration, and equity is further promoted in the way waiting time is calculated—for example, a patient’s time on dialysis before being added to the wait list is included in the patient’s waiting time to protect patients who are slower to be added to the list. A person’s place of residence may play a critical role in access to organ transplantation. While OPTN is reducing the influence of geography, the guidelines still include preferences for potential recipients closer to the donor hospital. Often the different criteria come into conflict. For example, patients can gain high priority because of an urgent medical need for a liver transplant but then have the priority offset by a shorter life expectancy even with a transplant (as when patients with chronic liver failure are not eligible for the highest priority status and thereby receive lower priority than patients with acute liver failure). While OPTN guidelines govern the allocation of organs for persons who are on a waiting list for a transplant, transplant hospitals are responsible for listing decisions, in 80  The MELD score is based on four measures of liver function, with higher MELD scores reflecting more serious liver disease. 81  Some patients qualify for Status 1A if they have experienced unsuccessful liver transplant surgery. These patients might have developed their initial need for a transplant after chronic liver failure.

864   Joaquin Cayon-­De Las Cuevas and David Orentlicher conjunction with OPTN criteria. Accordingly, while transplant hospitals exhibit considerable agreement in their policies, they also vary in their approaches. In deciding whom to place on the waitlist, transplant providers tend to consider several factors: the likelihood that the transplant surgery will go well, the length of time that the recipient will benefit from the transplant before either the transplant is rejected by the patient or the patient dies, and the quality of life that the recipient will experience with the transplant. For indirect measures of these factors, transplant centers look at the severity and cause of organ failure, the presence in the recipient of other illnesses that affect either life expectancy or the ability to tolerate major surgery, the patient’s age, and psychosocial criteria (e.g., alcohol or other drug abuse, psychiatric illness, and lack of compliance with treatment regimens in the past). With liver transplantation, for example, a number of unofficial policies have developed. Patients with liver cancer generally do not do well with a transplant because of recurrence of the cancer, and transplant centers typically limit transplants to liver cancer patients whose cancers are found early and have not yet spread. Other patients who are denied liver transplants are those who currently engage in drug abuse, have severe infections, or have a non-­liver, non-­skin cancer. Transplantation for patients with liver failure from alcohol abuse is common, but there is variation among centers in deciding which persons with alcoholic liver disease should be considered for transplantation. Factors that are used to select candidates include documented abstinence from alcohol (typically at least 6 months), evidence of social stability (e.g., employment, permanent residence, and marriage), presence of a good family or social support system, absence of illicit substance use, absence of psychiatric illness, and compliance with recommendations by the patient’s treating physicians. Some transplant centers are now performing transplants for some patients with HIV infection. In the past, the OPTN has suggested it would develop standard criteria for deciding when a person should be placed on the waiting list for either a kidney or liver. Such criteria would help address the problem of some transplant centers being too quick to place a patient on a waiting list (sometimes doing so on the theory that, by the time the patient reaches the top of the list, the patient will be ready for a transplant). Note that, to some extent, the allocation guidelines also give guidance for adding people to the waiting list. For example, if a patient’s liver disease is severe enough for the patient to qualify for Status 1A once listed or the patient has a high MELD score, the disease is also severe enough that the patient should be listed. Ability to pay is an obstacle for many potential organ transplant recipients. Medicare pays for kidney transplants, but not for all of the associated costs (e.g., coverage is temporary for drugs to prevent rejection of the transplanted organ by the recipient’s immune system that have to be taken as long as the organ is still functioning). Medicare also pays for heart, lung, and liver transplants but restricts the pool of potential recipients by requiring them to meet certain medical criteria and to qualify for Social Security disability benefits. In addition, with these organs, Medicare also does not pay all of the associated medical costs. Some recipients can rely on private insurance or Medicaid, but those patients too may receive only partial reimbursement for the costs of the transplant

Organ Transplantation   865 and its follow-­up care. Accordingly, for some people, the costs of a transplant may preclude its availability. Even for people who receive a transplant, financial considerations may compromise the long-­term success of the procedure. Data suggest that an important reason for failure of organs a few years after transplantation is the unaffordability of the drugs to suppress the recipient’s immune system. In response to concerns about racial disparities in kidney transplantation, the OPTN amended its allocation policy in November 2002 to increase minority access to kidney transplants. Recent studies had indicated that OPTN could reduce its emphasis on tissue matching while still maintaining a high rate of long-­term success. Accordingly, under the revised policy, tissue matching is relevant only in a few areas (i.e., when the donor and recipient match perfectly or when there is matching at the ‘DR locus’). The change in policy has reduced the magnitude of the disparities. Racial disparities also existed in liver transplantation, but adoption of the MELD scores has resulted in elimination of an association between race and the receipt of a liver transplant once patients are added to the waitlist. However, disparities on the basis of race still exist with respect to being referred to a transplant center for evaluation.82

6  Confidentiality and Data Protection 6.1  Europe It is obvious that both the data of the living or deceased donor and that of the recipient are personal data subject to a duty of confidentiality. This is stated in Article 23 of the Biomedicine Convention’s Additional Protocol, which only allows data collection, processing, and communication according to the rules relating to professional confidentiality and personal data protection. The same reference is made by Article 16 of the Organ Transposition Directive when requiring Member States to ensure that the fundamental right to protection of personal data is fully and effectively observed in all organ donation and transplantation activities, in conformity with EU provisions on the protection of personal data. In particular, Member States should take all necessary measures to ensure that (1) the data processed are kept confidential and secure so that any unauthorised accessing of data or systems that makes identification of donor or recipients possible should be penalised, (2) donors and recipients whose data are processed are not identifiable, and (3)  the principles relating to data quality are met. Given the date on which both the Protocol and the Directive were adopted, the reference to the applicable law on data protection should be understood as referring to the 82  Amit K Mathur et al., ‘Race and Ethnicity in Access to and Outcomes of Liver Transplantation’ (2009) 9 Am J Transplant 2662.

866   Joaquin Cayon-­De Las Cuevas and David Orentlicher current 2016 General Data Protection Regulation (GDPR).83 According to the GDPR, as a general rule, the processing of genetic data and data concerning health is prohibited.84 However, this prohibition does not apply when processing is necessary to protect the vital interests of the data subject or of another natural person where the data subject is physically or legally incapable of giving consent.85 The prohibition also does not apply when processing is necessary for the purposes of preventive or occupational medicine, for the assessment of the working capacity of the employee, medical diagnosis, the provision of health or social care or treatment or the management of health or social care systems and services on the basis of EU or Member State law, or pursuant to contract with a health professional subject to the obligation of professional secrecy.86 Taking into account these exceptions, processing is allowed in relation to persons from whom organs or tissues have been removed or the recipients of organs and tissues in so far as this is required for medical purposes, including traceability.87 Indeed, it must be remarked that traceability is one of the key points of transplantation as a whole. In this regard, the transplantation system should ensure the collection and recording of information required to ensure traceability of organs and tissues.88 This issue is one of the main concerns of the Organ Transplantation Directive. According to its Article 10, Member States must ensure that all organs procured, allocated, and transplanted on their territory can be traced from the donor to the recipient and vice versa in order to safeguard the health of donors and recipients. Furthermore, Member States should also ensure the implementation of a donor and recipient identification system that can identify each donation and each of the organs and recipients associated with it. With regard to such a system, Member States should ensure that confidentiality and data security measures are in place in compliance with Union and national provisions. It is also important to take into account that the competent authority or other bodies involved in the chain from donation to transplantation or disposal should keep the data needed to ensure traceability at all stages of the chain from donation to transplantation or disposal and the information on organ and donor characterisation. Last but not least, data required for full traceability must be kept for a minimum of 30 years after donation.

6.2  United States The federal Health Insurance Portability and Accountability Act of 1996 (HIPAA) and state laws protect the confidentiality of medical information generally and so also protect the confidentiality of organ donors and recipients. Hoffman and Herveg (this volume) discuss confidentiality rules in the United States and Europe. In addition, as 83  Officially ‘Regulation (EU) 2016/679 of the European Parliament and of the Council’ 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC, [2016] OJ L 119/1. 84  Article 9(1) GDPR. 85  Article 9(2)(c) GDPR. 86  9(2)(h) GDPR. 87  Article 23 of the Additional Protocol. 88  Article 3 of the Additional Protocol.

Organ Transplantation   867 mentioned, OPTN guidelines require organ transplant programs to ensure the confidentiality of the process. As in Europe, exceptions to confidentiality exist to ensure traceability of organs in case medical problems are identified with the donor, the organ, or the recipient. For example, living donors must commit to follow-­up testing, and, if an infectious disease or malignancy is detected in the first 2 years after donation, it will be disclosed to the organ recipient’s hospital and reported to the OPTN, and it may be subject to the usual reporting requirements of public health authorities for certain diseases.89

7  Financial Incentives for Organ Donation and Trafficking in Organs While generally prohibited, financial incentives for organ donation have been promoted for some time. Advocates view financial incentives as a key strategy for addressing the persistent shortage of organs and thereby saving many lives. Considerations of justice also can support payments to donors. Noting that transplant surgeons, organ recipients, and others involved in transplantation benefit, commentators argue that donors also ought to benefit. A third argument in favor of some kind of payment is that we may have fewer problems with donors or their families wanting to be involved with recipients or feeling that the recipient owes them something. Thomas Murray has observed that gift-­ giving, unlike business dealings, can result in open-­ended human relationships. In short, gifts create moral relationships that are much more open-­ended, less specifiable, and less contained than contracts. Contracts are well suited to the Marketplace where a strictly limited relationship for a narrow purpose—trading goods or ­services—is desired. Gifts are better for initiating and sustaining more rounded human relationships, where future expectations are unknown, and where the exchange of goods is secondary in importance to the relationship itself.90

And there can be problems with open-­ended relationships after organ transplantation. After studying the relations between organ recipients and the donor or the donor’s family, Renée Fox and Judith Swazey concluded that [w]hat recipients believe they owe to donors and the sense of obligation they feel about repaying “their” donor for what has been given, weigh heavily on them. This psychological and moral burden is especially onerous because the gift the recipient has received from the donor is so extraordinary that it is inherently unreciprocal. It 89  OPTN Policy 14.3. 90  Thomas H Murray, ‘Gifts of the Body and the Needs of Strangers’ (1987) 17(2) Hastings Cent Rep 30, at 31 (but preferring a system of donation for organs because of the important social values that it promotes).

868   Joaquin Cayon-­De Las Cuevas and David Orentlicher has no physical or symbolic equivalent. As a consequence, the giver, the receiver, and their families may find themselves locked in a creditor-­debtor vise that binds them one to another in a mutually fettering way. We have called these aspects of the gift-­exchange dimensions of transplantation, “the tyranny of the gift.”91

Opponents of financial incentives worry that payments for organs will ‘commodify’ the body (i.e., turn the body into a commodity) and undermine the important ethic that people are not like property. There also are concerns about desperate persons taking unacceptable risks for pay and being exploited by wealthier individuals. A third objection rests on concerns that payments for organs will reduce voluntary organ donation and lead to an overall decline in organs for transplantation, as when allowing payments for blood led to a drop in the blood supply. In response to these concerns, advocates observe first that society already permits a great deal of commodification when it permits people to work for a wage. As to the concern about unacceptable risks, payment would only be allowed for organs that a person is allowed to donate. Or, to further protect against risks to donor health, there have been proposals to pay people for their organs, but only if the organs are taken after death, with payment to the person’s family or estate. With regard to exploitation by the wealthy, organs obtained for payment would still be allocated under the usual rules. In other words, organs could be sold by individuals, but not purchased by other individuals. Finally, the experience with payments for organs in Iran indicates that they can increase the total supply of organs.

7.1  Europe Prohibition of financial gain is laid down in the Biomedicine Convention in Articles 21 and 22. As a general principle, the human body and its parts must not, as such, give rise to financial gain. On the other hand, in case of disposal of a removed part of the human body in the course of an intervention, it may be stored and used for a purpose other than that for which it was removed only if this is done in conformity with appropriate information and consent procedures. Concerning the transplantation field, as early as 1978, the CoE issued a Recommendation92 indicating that no substance of human origin might be offered for profit. However, this Recommendation already foresaw the need to compensate for loss of earnings and any expenses caused by the removal or preceding examination. The donor, or potential donor, must be compensated, independently of any possible medical

91  Renée C Fox and Judith P Swazey, Spare Parts. Organ Replacement in American Society (Oxford University Press 1992), at 40–41. 92  Recommendation on the harmonisation of legislation of Member States related to the removal, grafting, and transplantation of human substances [R (78)29].

Organ Transplantation   869 responsibility, for any damage sustained as a result of a removal procedure or preceding examination, under a social security or other insurance scheme.93 One may find the same prohibition in the EU legal framework. The Charter of Fundamental Rights states the prohibition on making the human body and its parts a source of financial gain.94 In addition, the 2010 Organ Transplantation Directive is also based on the principles of altruism and non-­payment, so that Member States should ensure that donations of organs from deceased and living donors are voluntary and unpaid and that the procurement of organs is carried out on a nonprofit basis.95 Furthermore, Member States must prohibit advertising the need for or availability of organs where such advertising is with a view to offering or seeking financial gain or comparable advantage.96 In this respect, financial gain for donors must be banned since it generates inequalities by discriminating against low-­income people as opposed to those who can afford to pay for an organ or tissue. In addition to this reason, the Organ Transplantation Directive introduces a novel argument by stating that altruism is essential because the violation of these principles might be associated with unacceptable risks. Where donation is not voluntary and/or is undertaken with a view to financial gain, the quality of the process of donation could be jeopardised because improving the quality of life or saving the life of a person is not the main and/or the unique objective. Even if the process is developed in accordance with appropriate quality standards, a clinical history obtained from either a potential living donor or the relatives of a potential deceased donor who are seeking financial gain or are subjected to any kind of coercion might not be sufficiently accurate in terms of conditions and/or diseases potentially transmissible from donor to recipient. This could give rise to a safety problem for potential recipients since the medical team would have a limited capability for performing an appropriate risk assessment.97 However, the prohibition of financial gain does not apply to payments which do not constitute a financial gain or a comparable advantage. In this regard, the Directive sets out that the principle of non-­payment should not prevent living donors from receiving compensation, provided it is strictly limited to making good the expenses and loss of income related to the donation. In this respect, Member States are allowed to define the

93  Article 9. 94  Article 3(2)(c). The Charter is legally binding since it is included in Article 6(1) of the Treaty of the European Union. The interaction between the Charter (EU) and the European Convention of Human Rights (CoE) is explicitly laid down in Article 6(3), stating that ‘fundamental rights, as guaranteed by the European Convention for the Protection of Human Rights and Fundamental Freedoms and as they result from the constitutional traditions common to the Member States, shall constitute general principles of the Union’s law’. 95  For further details on prohibition of financial gain, see the ‘Guide for the implementation of the principle of prohibition of financial gain with respect to the human body and its parts from living or deceased donors’, adopted by the European Committee on Organ Transplantation on 11 January 2018. 96  Article 13 of the Organ Transplantation Directive. 97  Recital 19 of the Organ Transplantation Directive.

870   Joaquin Cayon-­De Las Cuevas and David Orentlicher conditions under which such compensation may be granted while avoiding there being any financial incentives or benefit for a potential donor. In particular, the Additional Protocol mentions three types of compensation: (1) compensation of living donors for loss of earnings and any other justifiable expenses caused by the removal or by the related medical examinations, (2) payment of a justifiable fee for legitimate medical or related technical services rendered in connection with transplantation, and (3) compensation in case of undue damage resulting from the removal of organs or tissues from living persons.98 By undue damage is meant any harm whose occurrence is not a normal consequence of a transplant procedure. In this regard, Article 25 of the Additional Protocol makes clear that the person who has suffered undue damage resulting from transplantation procedures is entitled to fair compensation according to the conditions and procedures prescribed by law. Some expenses are controversial. For instance, it has been discussed whether or not burial expenses or cost of repatriation of the bodies of foreign donors constitute an incentive. Some European countries (e.g., Spain) do not consider reimbursement of these expenses a financial gain but this practice has been criticised by arguing it represents an undue incentive.99 There is an ethical debate in Europe about whether certain organ commercialisation should be allowed since sale would not necessarily involve exploitation and could benefit low-­income people. In this sense, the creation of systems of regulated commercialisation of organs has been defended, arguing that in this way black markets and exploitation would disappear and that the number of available organs would increase, thus generating an opportunity to obtain resources for those who lack them. Regardless of the considerations that may be made regarding the effect that this type of solution may have on human dignity, it does not seem very likely, however, that the establishment of this type of market in the human body would be the most appropriate means of putting an end to the exploitation suffered by those who are forced to resort to selling their own organs because they lack any other resource. On the contrary, the establishment of legislation favourable to this type of practice would only deepen this exploitation. It is clear that the hope of receiving financial compensation is a difficult incentive to reject for those who suffer from situations of misery precisely because they have no other alternatives for obtaining such income. It is this circumstance that places them in a situation of special vulnerability that paves the way for obtaining flawed consent.

98  Article 21(1) of the Additional Protocol. Article 24 of the Biomedicine Convention also lays down that the person who has suffered undue damage resulting from an intervention is entitled to fair compensation according to the conditions and procedures prescribed by law. 99  For an argument that the payments should not be considered an incentive, see Mónica Navarro-­ Michel ‘Institutional Organisation and Transplanting the Spanish Model’ in AM Farrell, D Price and ‎M Quigley (ed), Organ Shortage: Ethics, Law and Pragmatism (Cambridge University Press 2011) 153. For the opposite view, see Garwood-­Gowers and Pereira, n 75, at 398.

Organ Transplantation   871 A final reference to the prohibition of trafficking should be made. In principle, Article 22 of the Biomedicine Convention’s Additional Protocol merely states that ‘organ and tissue trafficking shall be prohibited’. This prohibition relies on the principle of non-­ commodification and has been addressed through a particular treaty in the context of the CoE level: the Convention Against Trafficking in Human Organs of 2015. It is important to note that the Convention is not only a European instrument since it is open for signature by the CoE Member States, the EU, and certain non-­CoE Member States.100 This is a framework instrument for the harmonisation of national criminal legislation that pursues three main goals: first, preventing and combating the trafficking in human organs by providing for the criminalisation of certain acts; second, protecting the rights of victims of the offences established; and, last but not least, facilitating cooperation at national and international levels on action against the trafficking in human organs. In particular, the Convention requires that State parties must implement the necessary legislative and other measures to establish as a criminal offence under its domestic law, when committed intentionally, the removal of human organs from living or deceased donors practiced under any of these circumstances: (1) where the removal is performed without the free, informed, and specific consent of the living or deceased donor, or, in the case of the deceased donor, without the removal being authorised under its domestic law101; (2) where, in exchange for the removal of organs, the living donor, or a third party, has been offered or has received a financial gain or comparable advantage; and (3) where, in exchange for the removal of organs from a deceased donor, a third party has been offered or has received a financial gain or comparable advantage. Finally it is important to point out that each party shall consider taking the necessary legislative or other measures to establish as a criminal offence under its domestic law the removal of human organs from living or deceased donors where the removal is performed outside of the framework of its domestic transplantation system or where the removal is performed in breach of essential principles of national transplantation laws or rules.102 This rule is an explicit acknowledgement of the importance of extraterritorial enforcement in tackling illicit organ sales.103

100  The non-­CoE Member States invited to sign the Convention are Canada, Costa Rica, the Holy See, Japan, Mexico, and the United States. At the time of writing this chapter, the Convention has been signed by 25 European countries but only Albania, Croatia, Czech Republic, Latvia, Malta, Montenegro, Norway, Portugal, Republic of Moldava, and Switzerland have ratified it. Costa Rica is the only non-­European country to have signed the Convention but has not ratified it yet. The treaty entered into force on 1 March 2018, once five ratifications were adopted including at least three Member States of the CoE. 101  Nevertheless, any State or the EU may declare that it reserves the right not to apply this provision to the removal of human organs from living donors, in exceptional cases and in accordance with appropriate safeguards or consent provisions under its domestic law. Any reservation made under this paragraph shall contain a brief statement of the relevant domestic law. 102  Article 4(2) of the Convention Against Trafficking in Human Organs. 103  Sheelagh McGuinness and Jean V McHale ‘Transnational Crimes Related to Health: How Should the Law Respond to the Illicit Organ Tourism?’ (2014) 34(4) Legal Stud 682, at 698.

872   Joaquin Cayon-­De Las Cuevas and David Orentlicher

7.2  United States Financial incentives for donation have been much discussed in the United States, and the American Medical Association has endorsed pilot programs to study the appropriateness of incentives,104 but there generally is resistance to the idea for the same reasons as in Europe. Accordingly, as mentioned, current law prohibits payments for organs, and people who traffic in organs are subject to fines of up to $50,000 and incarceration for up to 5 years.105 However, many states have enacted an incentive for living donors that is permitted under current law. In these states, when filing state income tax returns, living donors are entitled to a deduction from their taxable income and sometimes a credit on their taxes due for travel expenses, lodging expenses, or lost wages that are not reimbursed.106 This  incentive can be worth as much as $10,000, though it usually measures in the ­hundreds of dollars. It does not violate federal law since the ban on payments excludes lost wages and medical, travel, or housing expenses. Other state initiatives have been blocked by federal law. For example, in 2000, Pennsylvania enacted a statute authorizing payments of $3,000 to families for medical or funeral expenses of the decedent.107

8 Conclusion As the preceding discussion indicates, organ transplantation principles in Europe and the United States have much in common. Similar safeguards have been adopted to reduce the risk to donors, ensure informed and voluntary consent, and prevent commodification of organs. Perhaps the most important difference is the one that runs throughout healthcare: the greater role that a patient’s financial status plays in access to care. Just as poor Americans suffer from diminished access to healthcare generally, so do they suffer from diminished access to organ transplants. In contrast, the universal access to healthcare principle in Europe makes a patient’s economic status much less important when it comes to access to organ transplantation. 104  American Medical Association Code of Medical Ethics Opinion 6.1.3, initially discussed in Council on Ethical and Judicial Affairs, ‘Financial Incentives for Organ Procurement: Ethical Aspects of Future Contracts for Cadaveric Donors’ (1995) 155 Arch Intern Med 581. 105  42 USC. §274e(a)-(b). Though the Trafficking Victims Protection Act does not explicitly refer to organ trafficking, it does describe trafficking as including violations of laws against ‘false imprisonment, assault, battery. . . ‘. 22 USC. 7101(b)(10). The Act also identifies concerns about serious health risks to victims of trafficking, though the focus is on sex and labor trafficking. 22 USC. 7101(b) (11). 106  Available at: https://www.americantransplantfoundation.org/wp-­content/uploads/2018/08/living_donor_laws_ state_by_state_and_federal.pdf 107  20 Pa. Cons. Stat. Ann. §8622.

Organ Transplantation   873

Acknowledgments We are grateful for very helpful comments by Macey Henderson.

Bibliography Roberto Andorno, ‘Buying and Selling Organs: Issues of Commodification, Exploitation and Human Dignity’ (2017) 1(2) J Trafficking Hum Exploitation 119. John Fabre, ‘Presumed Consent Is Unnecessary’ (2010) 341 Br Med J 922. John Fabre, ‘Presumed Consent for Organ Donation: A Clinically Unnecessary and Corrupting Influence in Medicine and Politics’ (2014) 14(6) Clin Med 567. Austen Garwood-Gowers, ‘Time to Address the Problem of Post-Mortem Transplantation Occurring Without Proper Pre-Mortem Consent’ (2013) 20(4) Eur J Health L 383. Austen Garwood-Gowers and Solvita Olsena, ‘Respect for the Individual: A Human Right in Post-Mortem Use of the Human Body for Transplantation’ in W Weimar, MA Bos, and JJ Busschbach (eds), Organ Transplantation: Ethical, Legal and Psychosocial Aspects. Expanding the European Platform (Pabst 2011) 146. Michael B Gill and Robert M Sade, ‘Payment for Kidneys: The Case for Repealing Prohibition’ (2002) 12(1) Kennedy Instit Ethics J 30. I Glenn Cohen, ‘Transplant Tourism: The Ethics and Regulation of International Markets for Organs’ (2013) 41(1) J L Med Ethics 269. Mark Hall, David Orentlicher, Mary Anne Bobinski, Nicholas Bagley, and I Glenn Cohen, Health Care Law and Ethics, 9th ed. (Wolters Kluwer 2018), chapter 6. Arthur J Matas, ‘In Defense of a Regulated System of Compensation for Kidney Donation’, in W Weimar, MA Bos, and JJ Busschbach (eds), Organ Transplantation: Ethical, Legal and Psychosocial Aspects. Towards a Common European Policy (Pabst 2008) 55. Arthur J Matas, ‘Working Group on Incentives for Living Donation, Incentives for Organ Donation: Proposed Standards for an internationally Acceptable System’ (2012) 12(2) Am J Transplant 306. Jean McHale and Aurélie Mahalatchimy, ‘EU Law and Policy on Human Material’ in TK Hervey, CA Young and LE Bishop (ed), Research Handbook on EU Health Law and Policy (Edward Elgar Publishing 2017) 222. Janet Radcliffe Richards, The Ethics of Transplants: Why Careless Thought Costs Lives (Oxford University Press 2012). Robyn Shapiro, ‘Legal Issues in Payment of Living Donors for Solid Organs’ (2002) 7 Curr Opin Organ Transplant 375. Yosuke Shimazono, ‘The State of the International Organ Trade: A Provisional Picture Based on Integration of Available Information’ (2007) 85(12) Bull WHO 955. Stephen Wilkinson, Bodies for Sale: Ethics and Exploitation in the Human Body Trade (Routledge 2003).

chapter 40

I n troduction to I ncompl ete Com modification a n d Its Cr eepi ng Cou n ter pa rt Natalie Ram and Stéphanie Hennette Vauchez

How valuable is a human cornea? A pint of human blood? A vial of human sperm? And should that value be translated into cold, hard dollars or Euros for anyone (or everyone) involved in moving these body products from one body to another? In both the United States and Europe, the answers to these questions are ethically fraught and often dictated as much by culture as by law. On both sides of the Atlantic, markets in body products have been developing. In both cases, however, these markets are often limited or constrained. This can be related, at least in part, to strikingly converging normative trends, particularly when it comes to expressing unease with markets in body parts. These trends exert considerable force, regardless of how different the starting points of the dominant regulatory paradigms may be. The lived experiences of body product providers in both the United States and Europe are becoming increasingly similar. Within both systems, compensation for human transplantable organs—and at least some tissues and cells—is proscribed by law.1 Even where the law does not formally proscribe remuneration for body product providers, moreover, many continue to donate without payment. Throughout both the United States and Europe, whole blood is given without an expectation of payment or credit.

1  See 42 U.S.C. § 274e (US prohibition). In Europe, in addition to domestic rules that widely express such prohibition, see the insistence of a number of supranational rules (EU and Council of Europe) elevating “voluntary unpaid donation” as a normative standard.

876   Natalie Ram and Stéphanie Hennette Vauchez Primary organizations involved in collecting and distributing this valuable resource are largely nonprofit in nature, although for-­profit entities appear on the US scene.2 But monetary exchange also appears in both systems to some degree. In Europe, even though voluntary unpaid donation remains the norm, most systems allow (and indeed, increasingly develop) forms of “compensation”—a legal category that is reputedly distinct from remuneration. Similar allowances for “reasonable payments” appear in US law.3 Moreover, the US prohibition on profit-­seeking in body products is only partial. Interestingly, although both these normative frameworks differ in their substance, the ways in which they are effectively applied show considerable affinity between them. Sperm is bought and sold in the United States using the traditional language and tools of the market, and compensation serves a vital role in supporting the supply of human sperm for third-­party reproductive use. In Europe, although there is substantial variation among nations, the buying and selling of sperm is gaining currency. Indeed, Denmark is “the sperm capital of the world” and home to the world’s largest sperm bank, Cryos, which pays its male suppliers on a per-­sample basis.4 Cryos exports its product to more than 70 countries, and cross-­border travel to access reproductive resources more heavily regulated in one’s home country is widespread. Moreover, payments for egg providers continue to grow—and to inspire similar concerns and, in some instances, financial limits. In the United States, despite reported payments to egg providers of more than $100,000, egg selling typically eschews economic language, with egg agencies emphasizing altruism. Egg agencies may even reject potential egg providers if they seem too interested in the financial aspects of their role. More significantly, lawmakers and professional organizations have singled out egg provider payments as problematic. The leading professional organization for assisted human reproductive clinics has emphasized that compensation for egg providers is ethically fraught. Prior to 2016, that organization instructed that compensation topping $10,000 was simply not “appropriate.”5 Similar concerns arise across Europe. The marketization of sperm and eggs is typically constrained. By and large, European donors receive substantially lower payments than their US counterparts. But increasing pressure is appearing on these very conditions of the circulation and exchange of body products, especially for reproductive material. The Council of Europe has emphasized that market-­driven payments “are particularly worrisome in the case of gamete donors (especially oocyte donors), where they 2  See, e.g., Erin Carlyle, The Guys Who Trade Your Blood for Profit, Forbes, July 12, 2012, https:// www.forbes.com/sites/erincarlyle/2012/06/27/blood-­money-­the-­guys-­who-­trade-­your-­blood-­for-­ profit/#111fd296282e; Emily Largent, NOTA: Not a Good Act for Tissues to Follow, 19 Quinnipiac Health L.J. 179 (2016); see also Michelle Oberman, When the Truth Is Not Enough: Tissue Donation, Altruism, and the Market, 55 DePaul L. Rev. 903, 912 (2006) (“The distinction between nonprofit and for-­profit is little more than semantic when it comes to the existence of a market, complete with profit incentives, in human tissue”). 3  42 U.S.C. § 274e(c)(2). 4  See Sarfraz Manzoor, Come Inside: The World’s Biggest Sperm Bank, The Guardian, Nov. 2, 2012, https://www.theguardian.com/society/2012/nov/02/worlds-­biggest-­sperm-­bank-­denmark. 5  Ethics Cmte. of the Am. Soc’y for Reproductive Med., Financial Incentives in Recruitment of Oocyte Donors, 74 Fertility & Sterility 216, 219 (2000).

Introduction to Incomplete Commodification   877 may change the donor’s perception of the relative risks and benefits of a donation that is not free of potential health hazards and psychological consequences.”6 In particular, demand for access to human eggs for reproductive use may be driving significant policy change. For instance, while France has long adhered to a non-­remunerative framework for human gametes, as for other human body products, that framework is beginning to crumble, with a 2011 report recommending a more generous understanding of the costs incurred by egg providers.7 These similarities in policy and practices might appear commonsensical, yet they also reflect a remarkable convergence in international norms despite substantially different background assumptions and legal frameworks. In the United States, although federal law proscribes the sale of many human organs and tissues, that law does not reach all body products.8 Human blood, for instance, must be labeled as either “paid” or “volunteer,” but either is legally acceptable.9 Payments for human sperm and eggs are similarly outside the scope of the federal prohibition on sales, and little regulation or state law exists requiring disclosure of compensation or limiting such payments. In combination with a general American affection for market systems, one might have expected the lack of a legal prohibition of markets to give rise to precisely those markets. Indeed, the flourishing of for-­profit businesses trading in human tissue products— despite the legal prohibition on “valuable consideration” for such tissues10—underscores the American appetite for markets. But true markets largely have not developed for many body products outside that legal prohibition—at least where the initial body product supplier is concerned. To be sure, body product exchange in the United States, particularly where human gametes are concerned, supports a multibillion dollar a year industry.11 As described briefly earlier and in more detail in the chapter that follows, however, societal norms and professional guidelines have yielded incomplete commodification for the providers of many of these body products. The chapter that follows explores this incomplete commodification in the American system, linking it in part to traditional discomfort surrounding the role of women in markets. In Europe, meanwhile, supranational governing documents have enshrined unpaid donation across body product types. The Council of Europe’s Oviedo Convention of 1997, for instance, declares that “the human body and its parts shall not, as such, give rise to financial gain.”12 Unlike in the US setting, Europe did not draw lines between organs and tissues. Rather, explanatory notes accompanying the Oviedo Convention clarified that “under this provision organs and tissues proper, including blood, should not be 6  Council of Europe EDQM, Guide to the quality and safety of tissues and cells for human application, op. cit., p. 39. 7 IGAS, Etat des lieux et perspectives du don d’ovocytes en France, op. cit. 8  See 42 U.S.C. § 274e; Flynn v. Holder, 684 F.3d 852, 862 (9th Cir. 2012). 9  See 21 C.F.R. § 606.121(c)(8)(v). 10  See 42 U.S.C. § 274e. 11  See, e.g., Carlyle, supra note 2 (revenues in human blood); Donna Rosato, How High-­Tech Baby Making Fuels the Infertility Market Boom, Money, July 9, 2014, http://time.com/money/2955345/ high-­tech-­baby-­making-­is-­fueling-­a-­market-­boom/(revenues in human reproductive services). 12  Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine (ETS No 164), art. 21.

878   Natalie Ram and Stéphanie Hennette Vauchez bought or sold or give rise to financial gain for the person from whom they have been removed or for a third party, whether an individual or a corporate entity such as, for example, a hospital.”13 Supranational European entities have continued to press this noncommodification principle for blood and tissues as well as for transplantable organs.14 It is against this backdrop of noncommodification commitments that Europe has begun to engage with markets in certain human body products. As the social practices of biomedicine and the pressure of individuals’ reproductive projects and desires have increased patients’ and services’ cross-­ border mobility, EU law has increasingly appeared relevant and, in fact, supportive of circumventing domestic restrictions and constraints. Therefore, the growing role of the European Union as an organizing engine for Europe is of particular salience. The European Union takes as its focus and purpose a free-­flowing supranational market and economy for Europe writ large. Out of inevitable tensions between the Council of Europe’s human rights approach and the European Union’s market focus, commodification has been creeping in. The chapter by Hennette Vauchez explores this creeping commodification in greater detail, drawing out the tensions between Europe’s human rights and market-­driven models and their impact on the rise of market mechanisms for some types of body products. Many differences between the dominant regulatory frameworks that exist in Europe and the United States in the field of human tissues are beyond the scope of the two following chapters. In particular, as these chapters focus on the acquisition of and access to human tissues, they do not examine how and to what extent those conditions of acquisition and access shape the meaning of healthcare (or reproduction) as a human right.15 Thus, the course of commodification of human tissues—including blood, sperm, and eggs—is both the same and opposite for the United States and Europe. The law on the ground in each system may be quite different, but the lived experiences of individuals providing these body products are becoming increasingly similar. The chapters that follow examine the patterns of incomplete commodification in the United States and creeping commodification in Europe. In so doing, they identify, assess, and critique the limits and influence of market forces in shaping body product exchange for traditional human tissues as well as for blood and gametes and explore how these forces fare differently for these body products than for transplantable organs. They also seek to reflect on the gender dynamics that drive these incompletely commodified markets: Do markets emancipate from gender norms, or do they perpetuate them? And do markets reinforce the pressure exerted by demand on body products (gametes in particular), or are they just one way to channel it? 13  Official Explanatory Report to the Convention, §132: https://rm.coe.int/ CoERMPublicCommonSearchServices/DisplayDCTMContent?documentId=09000016800ccde5. 14  The Organ Safety Directive is the strongest example of the principle of non-­commodification of the human body in European law: Directive 2010/45/EU of the European Parliament and of the Council of 7 July 2010 on standards of quality and safety of human organs intended for transplantation. 15  See in particular other chapters in this volume, such as the chapter by I. Glenn Cohen & Emily Jackson.

chapter 41

I ncom pl ete Com modification i n A m er ica n L aw Natalie Ram

In the main, American law bars payments for human body parts and tissue to be used in clinical care.1 But this prohibition on payments does not reach all human body products. Rather, federal law sets out an explicit list of covered organs, and courts have interpreted that list strictly.2 The types of cells and tissues that are not included are not insignificant. Blood, sperm, and eggs fall outside the prohibition on payments. Despite a lack of legal prohibition on payments, however, there is no robust free market for the acquisition of most of the unlisted body products. Americans largely donate whole blood without any financial compensation. And while women who provide eggs for others’ reproductive use may be handsomely compensated,3 even these individuals are subject to price constraints not typical in a free market.4 This is not to say that large sums of money do not change hands in exchange for body products like these. To the contrary, human blood supports a $4.5 billion-­a-­year industry, including “disruptive” start-­up businesses attracting more than half a million dollars of investment capital.5 But most of these dollars do not flow to the individuals who are the source of the body product in question.6 1  See 42 USC § 274e. 2  See Flynn v. Holder, 684 F.3d 852 (9th Cir. 2012). 3  See, e.g., Katherine M. Johnson, The Price of an Egg: Oocyte Donor Compensation in the US Fertility Industry, 36 New Genetics & Soc’y. 354, 363 (2017) (“Average donor compensation was $5890”). 4  See Ethics Cmte. of the Am. Soc’y for Reproductive Med., Financial Incentives in Recruitment of Oocyte Donors, 74 Fertility & Sterility 216, 219 (2000) [hereinafter ASRM, 2000 Financial Incentives for Oocyte Donors]. 5  See Erin Carlyle, The Guys Who Trade Your Blood for Profit, Forbes, July 12, 2012, https://www. forbes.com/sites/erincarlyle/2012/06/27/ blood-­money-­the-­guys-­who-­trade-­your-­blood-­for-­profit/#111fd296282e. 6  A similar disjunction in payments to sources and payments to others can be seen in the context of tissues and cells used in scientific and medical research. See Natalie Ram, Body Banking from the Bench to the Bedside, 129 Harv. L. Rev. 491 (2015). Courts typically hold that individuals have no property interest in their cells once those cells are removed from the body and so no right to compensation for

880   Natalie Ram Moreover, even for tissues that do fall within the legal prohibition on payments, similar patterns of disparate compensation are evident. Indeed, for posthumously retrieved tissues for transplantation, “although the family members of donors are prohibited from receiving payment in exchange for permitting access to their loved ones’ tissues, . . . the tissue industry is definitively—indeed fabulously—profitable, generating an estimated one billion dollars per year.”7 From skin, to heart valves, to corneas, partial commodification abounds in the face of a legal system that purports to frown on commodifying the human body.8 Thus, despite strong demand for body products and the traditional American preference for markets in general, myriad body products are procured in partial, though incomplete, markets. On one hand, when market mechanisms are putatively prohibited, markets have nonetheless crept in at some—but not all—stages of the process of transferring tissue from the donor to the recipient. On the other hand, when markets are not prohibited, they nonetheless do not develop fully. This chapter explores this juxtaposition of law and norms in the American setting. Section 1 briefly sets out the existing legal standards governing payments for body products and identifies their limits. Section 2 considers how, in the face of a legal prohibition on profit-­seeking, markets have nonetheless developed around the clinical use of many human tissues—but only for certain kinds of market participants. Significantly, though there is big money to be made, those dollars largely do not flow to the individuals who are the source of the body products in question. Section 3 flips the analysis, exploring the absence of robust markets even where they are not prohibited. Examining sperm, eggs, and blood as a diverse set of relevant body products not subject to widespread legal prohibitions on payments, this section explains that markets are nonetheless often, though not entirely, underdeveloped or absent, at least for the sources of those body products. Section 3 also suggests one reason for the incomplete commodification of many of these body products: a preoccupation and discomfort with the role of women in markets rather than at home. That is, when it comes to body products, women are deemed to participate primarily for reasons of love, not money. Although gendered assumptions their use in research or stake in any technologies resulting from that research. See Moore v. Regents of the Univ. of Cal., 793 P.2d 479, 480 (Cal. 1990); Greenberg v. Miami Children’s Hosp. Research Inst., Inc., 264 F. Supp. 2d 1064, 1074 (S.D. Fla. 2003) (dismissing conversion claim and finding that individuals providing tissue samples for research on Canavan disease made “donations to research without any contemporaneous expectations of return of the body tissue and genetic samples, and thus conversion does not lie as a cause of action”). The research use of tissue is discussed in the chapter by Laurie and Coleman in this volume. 7  Michelle Oberman, When the Truth Is Not Enough: Tissue Donation, Altruism, and the Market, 55 DePaul L. Rev. 903, 910–911 (2006). Other estimates peg tissue transplantation revenues in the multiple billions of dollars. See Emily Largent, NOTA: Not a Good Act for Tissues to Follow, 19 Quinnipiac Health L. J. 179, 199 (2016) (“tissue transplantation is a multibillion-­dollar industry”). 8  See, e.g., Oberman, supra note 7; Robert A. Katz, The Re-­Gift of Life: Can Charity Law Prevent For- Profit Firms from Exploiting Donated Tissue and Nonprofit Tissue Banks?, 55 DePaul L. Rev. 943 (2006).

Incomplete Commodification in American Law   881 do not fully explain the incomplete commodification of body products, whether within or outside the scope of the legal prohibition on payments, its partial explanatory force suggests further reason for skepticism about the patterns of disparate compensation that have emerged throughout the American system.

1  The Legal Status Quo As in much of the rest of the world, organ and tissue sales in the United States are proscribed by law. As part of the National Organ Transplant Act (NOTA) of 1984, the US Congress declared it “unlawful for any person to knowingly acquire, receive, or otherwise transfer any human organ for valuable consideration for use in human transplantation.”9 This prohibition applies without regard to whether a donor is living or deceased.10 Congress defined “human organ” more broadly than might be expected. It includes “human (including fetal) kidney, liver, heart, lung, pancreas, bone marrow, cornea, eye, bone, and skin or any subpart thereof,” as well as “any other human organ (or any subpart thereof, including that derived from a fetus) specified by the Secretary of Health and Human Services by regulation.”11 The Secretary has expanded on the list of congressionally identified human organs for which “valuable consideration” is forbidden on several occasions. Currently, the prohibition extends by regulation to the gastrointestinal tract and to “any vascularized composite allograft.”12 Congress defined “valuable consideration” in the negative, setting out what it does not include, rather than what it does. Accordingly, “reasonable payments associated with the removal, transportation, implantation, processing, preservation, quality control, and storage of a human organ or the expenses of travel, housing, and lost wages incurred by the donor of a human organ” are permitted.13 So hospitals can bill for the cost of transplant surgery, and people can be reimbursed for any financial costs incurred by virtue of donating, but donors cannot gain financially from their donations. 9  42 USC § 274e(a). 10  Id.; Richards v. Holder, No. 13–13195–LTS, 2014 WL 2805280, at *5 (D. Mass, June 19, 2014) (dismissing claim that federal prohibition on sale of posthumous organs violates takings clause of the US Constitution). 11  Id. § 274e(c)(1). 12  42 C.F.R. § 121.13. Vascularized composite allografts refer to body parts with multiple tissues, as in the case of hands or faces, that are being used for transplants with increasing frequency. 13  42 USC § 274e(c)(2). In 2007, Congress clarified that the prohibition on “valuable consideration” also does not apply to an exchange of one organ for another in the context of “human organ paired donation.” See Charlie W. Norwood Living Organ Donation Act, Pub. L. No. 110-­144 (2007) (amending 42 USC § 274e(a)). In such a donation, two individuals in need of an organ (say, A and B) each recruit a willing unremunerated organ donor, but where A’s willing donor matches B (not A) and B’s willing donor matches A (not B). Under these circumstances, A and B are permitted, effectively, to trade an organ for an organ, with A’s donor giving an organ to B and vice versa. See 42 USC § 274e(c)(4). Paired donations may include more than two donor–recipient pairs, as well, permitting for extensive nonsimultaneous, extended, altruistic donor (NEAD) chains. Id.

882   Natalie Ram Despite the broad definition of organs, regulation of covered body products ranges widely. Traditional organs for transplantation, like hearts, lungs, and kidneys, are part of a heavily regulated transplantation system, including a national system for allocating scarce organs to needy recipients.14 By contrast, “the world of tissue transplantation is relatively invisible and largely unregulated.”15 Tissue banks and processors need not be nonprofit entities, nor is accreditation required.16 Indeed, as few as one-­third of tissue banks may be accredited by the relevant body, the American Association of Tissue Banks.17 And while the US Food and Drug Administration (FDA) also regulates transplantable cells and tissues,18 its inspections are typically few and far between.19 Moreover, NOTA’s coverage—and thus its prohibition on valuable consideration for body products—is not comprehensive. Blood transfusions long pre-­dated NOTA, and courts have construed its absence from the list of explicitly identified organs as a clear indication that payments for blood are permitted.20 As set forth later, a long history of payments for blood providers supports this view,21 as does the continuing practice of paying those who provide plasma, a component of whole blood.22 Federal law also explicitly contemplates that blood providers may be paid. FDA regulations require that blood for transfusion be labeled “paid donor” or “volunteer donor” to reflect the presence or absence of monetary compensation.23 Courts have also interpreted the text of NOTA’s prohibitions strictly, in part because of its exclusion of blood from the list of prohibited transactions. In Flynn v. Holder, the Ninth Circuit Court of Appeals affirmed the constitutionality of Congress’s prohibition of valuable consideration for “bone marrow,” but construed “bone marrow” not to reach peripheral hematopoietic stem cells collected and separated from whole blood.24 Although in substance bone marrow and peripheral hematopoietic stem cells fill the same need, the source and method of their collection is markedly different. Peripheral hematopoietic stem cells can be collected from blood flowing in the veins, much as is plasma. The Ninth Circuit, in other words, considered peripheral hematopoietic stem cells to be more like blood and plasma than like true bone marrow.25 In the wake of Flynn, the Department of Health and Human Services proposed expanding its regula14  See About UNOS, https://unos.org/about/(“United Network for Organ Sharing [UNOS] is the private, non-­profit organization that manages the nation’s organ transplant system under contract with the federal government”); see also the chapter by Capron in this volume (discussing clinical use of transplantable organs in the United States). 15 Oberman, supra note 7, at 905. 16  See Largent, supra note 7, at 202. 17 Id. 18  See generally 21 C.F.R. Parts 1270 and 1271. 19  See Largent, supra note 7, at 202–203. 20  See Flynn v. Holder, 684 F.3d 852, 862 (9th Cir. 2012). 21  See infra Section 3.4. 22  See Elizabeth Preston, Why You Get Paid to Donate Plasma But Not Blood, Stat, Jan. 22, 2016, https://www.statnews.com/2016/01/22/paid-­plasma-­not-­blood/. 23  21 C.F.R. § 606.121(c)(8)(v); see also Requirements for Blood and Blood Components Intended for Transfusion or for Further Manufacturing Use, 80 Fed. Reg. 29848 (2015) (codified at 201 C.F.R. pt. 606) (“FDA regulations do not preclude paid donations for blood for transfusion or for further manufacture”). 24  Flynn, 684 F.3d at 862, 864–865. 25  Id. at 864–865.

Incomplete Commodification in American Law   883 tory definition of “bone marrow” to reach peripheral hematopoietic stem cells,26 but this proposed change was never finalized. In addition to blood, gametes (sperm and eggs) are not covered either by NOTA’s express text or the regulations promulgated under it. The first public sperm banks in the United States opened their doors in the early 1970s,27 and so Congress may well have been aware of the use of such tissue when it enacted NOTA. It is far less likely that Congress would have had donor eggs in mind when enacting NOTA as the first successful birth of a child conceived using an egg outside the human body occurred only in 1978,28 with the successful use of eggs from third parties occurring even a few years after that.29 Regardless, neither in 1984 nor at any time thereafter has Congress or the Department of Health and Human Services suggested that sperm and eggs are governed by NOTA’s prohibition. Other sources of law have largely left payments for body products unregulated beyond the scope of NOTA’s payment restrictions. Federal law and regulations have imposed or suggested a variety of health and safety requirements for blood and gametes, but these have not regulated the ability to pay for these body products.30 State governments also have not acted extensively to regulate payments for body parts otherwise permitted under federal law. A few states have required disclosure that for-­profit entities may be a part of transplantable tissue processing and distribution31 but have not barred such profit-­seeking directly. A small number of states have acted to limit or bar payments specifically for human eggs and embryos intended for reproductive use.32 This lack of state action may be because federal law might preempt conflicting measures. The Supreme Court has previously indicated, however, that state statutes pertaining to body products otherwise regulated under federal law are not necessarily preempted.33

26  See Change to the Definition of “Human Organ” Under Section 301 of the National Organ Transplant Act of 1984, 78 Fed. Reg. 60810-­01 (2013). 27  See Kara W. Swanson, Banking on the Body 219 (2014). 28  See id. at 229. 29  See Rene Almeling, Sex Cells: The Medical Market for Eggs and Sperm 4 (2011); Katherine M. Johnson, The Price of an Egg: Oocyte Donor Compensation in the US Fertility Industry, 36 New Genetics & Soc’y. 354, 355 (2017) (“Egg donation . . . emerged in the early 1980s as one of several treatments for infertility”). 30  See Yaniv Heled, The Regulation of Genetic Aspects of Donated Reproductive Tissue—The Need for Federal Regulation, 11 Col. Sci. & Tech. L. Rev. 244 (2010) (describing existing federal and state regulatory schemes for donated sperm, eggs, and embryos); see generally 21 C.F.R. Subch. F (regulating “biologics,” including blood); 21 C.F.R. § 1271 (regulating human tissue). 31  See, e.g., Cal. Health & Safety Code § 7158.3(b); Wis. Stat. Ann. § 157.06(25m). Both California and Wisconsin law permit a tissue donor or their family member to decline to consent to the use of donated tissue by a for-­profit entity. See Cal. Health & Safety Code § 7158.3(b)(2); Wis. Stat. Ann. § 157.06(25m)(b)(2). 32  See, e.g., Ind. Code Ann. § 35-­46-­5-­3; La. Rev. Stat. Ann. § 9:122; see also John A. Robertson, Egg Freezing and Egg Banking: Empowerment and Alienation in Assisted Reproduction, 1 J. L. & the Biosciences 113, 124 & n. 47 (2014). 33  See Hillsborough Cty., Fla. v. Automated Med. Labs., Inc., 471 US 707 (1985) (declining to find preemption of county regulations of plasma providers and collection centers).

884   Natalie Ram In sum, while US law often bars payment and other “valuable consideration” in exchange for human organs and body products, that prohibition is conspicuously incomplete. When it comes to blood, sperm, and eggs, few legal limits on compensation exist.

2  Incomplete Commodification Despite NOTA In the face of NOTA’s clear prohibition of profit-­seeking in transplantable tissue, one might expect that there would be little to say about profits and markets for such tissues. The truth, however, is far less clear-­cut. Indeed, it is in the context of NOTA-­covered tissues that some of the starkest disparities in compensation and profit-­seeking occur. The result is an incomplete commodification of human tissues—with n ­ oncommodification for tissue providers and their families, on the one hand, and for-­profit tissue processors and market-­based tissue allocations, on the other. As with some organs, both living and recently deceased individuals may serve as sources for transplantable tissues, though most such tissues are recovered post­hu­ mously.34 Many more individuals are candidates for posthumous tissue recovery than for organ recovery. That is so for two primary reasons. First, tissue may be recovered from a body up to 24 hours after death.35 Second, rejection of transplanted tissues is not typically a concern, and so tissue donors and recipients need not be immunologically matched.36 Moreover, many more tissues than organs can be recovered from a single body.37 “Whereas one person can save up to eight lives through organ donation, one person can enhance the lives of fifty or more people through tissue donation.”38 In this setting, families of the recently deceased are often solicited to donate their loved one’s remains for tissue and other recovery. Although the costs of tissue retrieval may be substantial,39 families are not made to shoulder these expenses. But neither are they privy to any monetary gains that may be extracted through the harvesting, processing, distributing, or use of their loved one’s body products. NOTA bars such revenue-­ sharing for families of the deceased.40 Those monetary gains may, however, be substantial. Nonprofit organ procurement organizations (OPOs) or tissue banks charge tissue processors “standard acquisition charges” in exchange for the tissue they recover from posthumous bodies.41 These charges vary widely, with a “typical donor produc[ing] $14,000 to $34,000 in sales for the nonprofits.”42 After treating and rendering tissue transplantable, a tissue processor

34  See Largent, supra note 7, at 193. 35  Id. at 197. 36  Id. at 199. 37  Id. at 195. 38 Id. 39 Oberman, supra note 7, at 909. 40  See id. at 910. 41  See Largent, supra note 7, at 197–198. 42 Oberman, supra note 7, at 909 (internal quotation marks omitted).

Incomplete Commodification in American Law   885 may in turn reap as much as $220,000 from a single cadaver as tissues are distributed to doctors, hospitals, and other users.43 The uses of these valuable tissues also vary widely. Transplantable organs are typically life-­saving, and nearly two dozen people die each day waiting for transplants.44 Human tissue may also play a significant role in healing, though it is not often “life-­saving.”45 Human bone may be used for spinal fusions or to repair fractures, heart valves and arteries may be transplanted, veins may be used in cardiac bypass procedures, skin grafts can replace burned skin, and corneas may restore clear vision.46 But human tissue may also be used for less vital purposes.47 For instance, instead of treating burn victims, human skin “can be processed into products that are useful for cosmetic purposes such as enhancing lip size, reducing wrinkles, or enlarging penises.”48 Allocation of human tissue among and between these myriad uses follows the dictates of a market in which “[p]otential profits rather than (roughly) prioritarian considerations” rule.49 Many for-­profit tissue processors and nonprofit tissue banks have created alliances or exclusive partnerships “in order to stabilize or increase their supply needs.”50 Some for-­profit tissue processors have gone further, establishing nonprofit tissue banks to supply their needs.51 More broadly, “the close working relationship between non-­profit and for-­profit agencies dealing in human tissue has led suppliers to favor more regular buyers, or those able to reimburse at higher rates, over others.”52 One result is that burn units often have difficulty securing enough skin for transplantation, even as “private plastic surgeons enjoy an ample supply for their elective surgeries.”53 Moreover, the close relationships between nonprofit and for-­profit entities underscore that “nonprofit” does not mean free from the market. Rather, as Michelle Oberman astutely observes, “[t]he distinction between nonprofit and for-­profit is little more than semantic when it comes to the existence of a market, complete with profit incentives, in human tissue.”54 In sum, where NOTA applies, it has not prevented spectacularly lucrative markets in human tissue from developing; it has merely prevented donors and their families from sharing in that wealth. Commodification is not absent from the acquisition, receipt, or transfer of these tissues, as NOTA suggests and perhaps intended. Rather, commodification is merely incomplete, leaving uncompensated those individuals whose bodies are the source of valuable material. There are two additional ways in which NOTA-­covered body products merit consideration under the rubric of incomplete markets, one legal and largely unregulated and the other illegal but difficult to monitor. The first consists of human remains that are “donated” for scientific research or education.55 In this setting, families of the recently 43 Id. 44 Largent, supra note 7, at 200. 45 Id. 46  See id.; Oberman, supra note 7, at 910. 47 Id. 48  Oberman, supra note 7, at 910. 49 Largent, supra note 7, at 199. 50 Oberman, supra note 7, at 912. 51 Id. 52  Id. at 911. 53 Id. 54  Id. at 912. 55  See Brian Grow & John Shiffman, In the US Market for Human Bodies, Almost Anyone Can Dissect and Sell the Dead, Reuters, Oct. 24, 2017, https://www.reuters.com/investigates/special-­report/ usa-­bodies-­brokers/.

886   Natalie Ram deceased are often solicited to donate their loved one’s remains to “advance medical studies” in exchange for cremation services.56 Those who agree, largely families of limited economic means who cannot afford the substantial costs of a funeral or ordinary cremation, do receive a thing of value: cremation services of the parts of the body that are not taken by body brokers.57 But the value of that service pales in comparison to what body brokers can recover from a single human body. Many brokers subdivide human bodies into multiple parts that can be sold or leased to medical schools and other interested parties for thousands of dollars apiece.58 Indeed, with medical schools and other training programs often requiring several similar body parts, orders for “27 shoulders” or “5 arms” are not out of the norm.59 Meanwhile, most body brokers in the United States are for-­profit entities, generating multimillion dollar revenues.60 The vagaries of this market are largely beyond the scope of this chapter, however, as this is not a market in human body products for clinical use. Rather, these bodies are provided and used for “education” and “research” purposes and thus fall outside the legal regulations discussed here. Finally, a string of public exposés and lawsuits reveals that body products, including NOTA-­covered organs and tissues, are sometimes taken from human remains for transplantation without the consent (and even against the express wishes) of family members of the deceased.61 In one case, the Sixth Circuit Court of Appeals recognized that family members retain a quasi-­property interest in the remains of their loved ones. The court held further that such an interest can be violated where a coroner removes the deceased’s corneas as transplantable “gifts” contrary to the family’s wishes.62 In the case of Alistair Cooke, a British-­American journalist and broadcaster whose bones were posthumously removed without his family’s knowledge, the individual at the heart of the body business generated millions of dollars in profit by unlawfully forging signatures on consent forms and selling transplantable body products to other intermediaries in the body product market.63 In many instances, as in the Brotherton and Cooke cases, those who traffic in human remains and their parts do so in violation of the law, and their conduct can be difficult to detect because they traffic in the dead. But as lawful markets in NOTA-­ covered tissues for clinical use indicate, the appropriation of human remains without consent to generate substantial profits may, unfortunately, be more a difference in degree than a difference in kind.

56 Id. 57 Id. 58 Id. 59 Id. 60 Id. 61  See, e.g., Brotherton v. Cleveland, 923 F.2d 477 (6th Cir. 1991) (reinstating wife’s § 1983 claim where deceased husband’s corneas were removed for transplantation purposes despite wife’s refusal to consent to such removal and use); Tom Peterkin, Alistair Cooke’s Body Snatcher to be Jailed, Telegraph, May 21, 2008, https://www.telegraph.co.uk/news/worldnews/northamerica/usa/1999617/Alistair-­Cookes-­ body-­snatcher-­to-­be-­jailed.html (describing the unconsented-­to harvesting and sale of body parts from human remains without consent). 62  Brotherton, 923 F.2d 477. 63  See Peterkin, supra note 61; Haroon Siddique, Alistair Cooke Bodysnatcher Suspect Expected to Admit Guilt, Guardian, Jan. 22, 2008, https://www.theguardian.com/uk/2008/jan/22/usa.world.

Incomplete Commodification in American Law   887

3  Incomplete Commodification Beyond NOTA In the absence of price limits or prohibition, one might expect that the United States would develop free markets in body products that fall outside of NOTA’s ambit. Indeed, in some instances, relatively robust market mechanisms have taken root, most prominently in the case of human sperm and, to some extent, human eggs. More frequently, however, a lack of formal payment limitations has not given rise to a true market for those who supply the desired raw materials. Rather, non-­legal norms or guidelines have created price ceilings or price barriers for many body product suppliers—even where intermediaries between suppliers and ultimate users of body products have cultivated profitable markets. This section describes these possible markets in turn and draws out how the trend away from unregulated market systems in the context of human eggs and even whole blood may be, in part, attributable to long-­held discomfort about women and markets.

3.1  Human Sperm Human sperm is largely marketed, acquired, and sold in the traditional language and tools of the market. Men who provide sperm to sperm banks are paid, and there are indications that those payments are not merely compensation for the “time” or “burden” required to produce usable sperm.64 Payments for sperm providers can be substantial. Prominent sperm banks advertise that men can earn more than $1,500 per month for as little as five hours of (possibly enjoyable) physical effort per month.65 Significantly, payments to sperm providers are typically calculated on a per sample basis, with compensation ranging from about $50 to $100 per sample.66 Some banks provide an additional payment for each vial of saleable 64  But see ASRM, 2000 Financial Incentives for Oocyte Donors, supra note 4, at 219 (emphasizing that payments to egg providers should “reflect the amount of time expended and the burdens of the procedures performed”). 65  Cal. Cryobank, Sperm Donor Compensation, http://www.spermbank.com/why-­donate/ sperm-­donor-­pay; see also NW Cryobank, Top 5 Reasons to Donate Sperm, https://nwsperm.com/ why-­donate/top-­5-­reasons-­to-­donate-­sperm (identifying “Get Compensated” as the Number 2 reason to donate sperm and explaining, “[w]e reimburse qualified donors for their time and expenses with compensation of up to $1,000/month”) 66  See Rene Almeling, Selling Genes, Selling Gender: Egg Agencies, Sperm Banks, and the Medical Market in Genetic Material, 72 Am. Sociological Rev. 319, 320 (2007); see also Almeling, supra note 29, at 2 (reciting an average of $100 per sample); ASRM, 2000 Financial Incentives for Oocyte Donors, supra note 4, at 220 (“In 2000, the average payment to sperm donors was $60–$75”); Dina Spector & Hilary Brueck, 9 Ways to Make Money by Selling Your Body to Science, Bus. Insider, Apr. 1, 2019, https://www.businessinsider.com/ways-­to-­make-­money-­from-­medical-­research-­and-­ donations-­2013-­12 (“Men are paid anywhere from $35 to $125 per donation”).

888   Natalie Ram sperm a given sample will fill.67 The sperm bank determines whether a given sample is “acceptable” for payment purposes, with this determination turning on factors like sperm count and quality.68 Thus, a man’s effort is not directly rewarded; he is only paid for what he produces. Sperm providers may be eligible for other rewards as well. They often are wooed with “periodic incentives such as movie tickets or gift certificates for extra time and effort expended by participants.”69 They may receive “hefty ‘finder fees’” for referring other successful sperm providers.70 And they typically receive free medical benefits like “semen analysis,” sexually transmitted disease (STD) screening, and testing for certain genetic and communicable diseases.71 Compensation serves a vital role in supporting the supply of human sperm for third-­ party reproductive use, particularly in the United States. In a review of 29 studies exploring sperm donation, the authors concluded, “studies with potential donor populations from the USA often report higher percentages of men with a financial motivation.”72 Recent survey data also indicate that most people believe and support that the primary motivation of sperm providers is monetary and that payments to sperm providers should reflect necessary incentives to recruit enough providers.73 Economic language permeates the relationship between sperm providers and sperm banks. Sperm banks advertise that a sperm provider can “[b]e your own boss!”74 Although sperm banks note that sperm providers are “help[ing] people fulfill their dreams of starting a family,”75 sperm banks fully “expect [sperm providers] to be financially motivated, and the staff speak directly about responsibility rather than couching it in terms of altruistic motivations.”76 Market forces shape the business of sperm banks in other ways as well. Sperm banks collect and make available to their purchasers large amounts of information about potential sperm providers, enabling buyers to fine-­tune the kinds of characteristics they hope their future child will have. One large sperm bank, on its website for sperm ­purchasers, enables these potential buyers to search by “ancestry,” hair and eye color, education, “ethnic background” and “skin tone,” and “hair type,” as well as by a host of 67  See David Plotz, The Genius Factory: My Short, Scary Career as a Sperm Donor, Slate, June 7, 2005, http://www.slate.com/articles/life/seed/2005/06/the_genius_factory.html (reporting that Fairfax Cryobank personnel indicated, “[y]ou will get paid $50 per usable specimen, for starters. Then you will get $5 for every vial from the specimen. The average is 10 to 14 vials per specimen. When a vial is released from quarantine after six months, you will get another $5. So the average payment is $209 per deposit”). 68 Almeling, supra note 66, at 320. 69  Cal. Cryobank, supra note 65. 70  See Almeling, supra note 66, at 325. 71  See Cal. Cryobank, supra note 65; NW Cryobank, supra note 65. 72  U. Van den Broeck et al., A Systematic Review of Sperm Donors: Demographic Characteristics, Attitudes, Motives and Experiences of the Process of Sperm Donation, 19 Hum. Reprod. Update 37, 44 (2013). 73  Malinda S. Lee et al., Limitations on the Compensation of Gamete Donors: A Public Opinion Survey, 107 Fertility & Sterility 1355, 1357. 74  Cal. Cryobank, supra note 65. 75 Id. 76  Almeling, supra note 66, at 327.

Incomplete Commodification in American Law   889 “lifestyle” factors, including astrological sign, religion, “favorite subject,” and “favorite pet.”77 This facilitation of detailed discrimination between potential sperm sources reinforces that sperm providers are valued (and recruited and paid) for the precise parameters of the sperm they can produce. There is at least one aspect, however, in which sperm banks do not reflect a fully realized free market. In some sperm banks, all sperm samples are bought and sold for the same rate, regardless of the specific characteristics of a particular sperm provider. Thus, Professor Rene Almeling reports that, for the sperm banks she interviewed, all sperm providers serving a particular bank are paid the same rate, and all sperm vials cost ­clients the same amount, too.78 But this same-­price practice is not universal. At least one report observes that Fairfax Cryobank, a large sperm bank with branches in multiple US states, charges clients a premium for sperm from providers with more desirable profiles.79 It is unclear whether those “more desirable” sperm providers are also compensated at higher rates. Moreover, research consistently indicates that base compensation rates paid to sperm donors do not appear to vary widely between and within institutions.80 Nonetheless, the market for sperm is one that operates free from both formal and informal price constraints. Neither government entities nor professional organizations have suggested that regulation of monetary compensation for sperm providers is ­desirable or necessary.

3.2  Human Eggs In many respects, the market in human eggs resembles the market in human sperm. Egg providers are usually paid.81 Eggs agencies82 typically collect extensive data about egg providers so as to facilitate maximum choice for their clients.83 Rates of compensation are significant. Various studies peg average compensation for a cycle of egg retrieval at $4,000–$6,000.84 Much of this premium may reflect the 77  Fairfax Cryobank, Donor Search, https://fairfaxcryobank.com/search/; see also Plotz, supra note 67. 78 Almeling, supra note 66, at 332. 79  See Plotz, supra note 67 (reporting that, at Fairfax Cryobank, “[f]or a premium, mothers can buy sperm from donors who have doctoral degrees or are pursuing them”). 80  See Almeling, supra note 29, at 2. 81  See id. 82  Until recently, it was not possible to reliably freeze and thaw unfertilized human eggs for reproductive use. Indeed, ASRM recognized that egg freezing “should no longer be considered experimental” only in 2013. Practice Cmte. of the Am. Soc’y for Reproductive Med., Mature Oocyte Cryopreservation: A Guideline, 99 Fertility & Sterility 37 (2013). Prior to the advent and acceptance of egg freezing, individuals seeking to acquire human eggs from another for their reproductive use typically worked with egg “agencies” to connect with egg providers. See Almeling, supra note 66. 83  See Almeling, supra note 66, at 326 (describing how both sperm banks and egg agencies collect and communicate extensive health, appearance, and “social” data about providers to “sell” their product to clients). 84  See id. at 320 (“Compensation to egg donors varies substantially, both within particular agencies and in different regions of the United States, but the national average is around $4,200”); Johnson,

890   Natalie Ram c­ onsiderable time, discomfort, and risk associated with ovarian stimulation for egg retrieval. As the American Society of Reproductive Medicine (ASRM) has observed, “[w]omen undergoing retrieval purely to provide oocytes to others are exposed to physical and psychological burdens they would not otherwise face.”85 These burdens include a heightened risk of unintended pregnancy, as well as health risks associated with the drugs used to induce superovulation and complete egg retrieval.86 In some ways, the market in human eggs may be more robust than the market in human sperm. Unlike sperm providers, who are often paid the same fee regardless of how “desirable” their sperm may be, “[c]ompensation to egg donors varies substantially, both within particular agencies and in different regions of the United States.”87 Some of this variation may be the result of greater variability in the egg provider population. According to a recent study of US egg agencies, only 20% had “selective donor recruitment criteria, such as requiring a college degree or recruiting donors of a specific race/ ethnicity.”88 Sperm banks, by contrast, often boast about their low “acceptance rates”— the number of men who are accepted to be sperm providers at all.89 But other factors affect egg provider compensation as well. An egg provider may command a higher fee for a second or subsequent egg retrieval cycle than for her first.90 That fee may be higher still if a previous egg retrieval cycle resulted in a pregnancy.91 Moreover, fees may vary based on the characteristics of the egg provider, with more highly educated women netting higher rates of compensation.92 One egg agency website bluntly explains, “We do not have a cap on fees but the amount is usually set by qualifications for egg donors. These may include educational level, SAT/ACT Test scores, personal achievements such as athletic endeavors, family history, talents, advanced degrees, special awards, and modeling photos.”93 And compensation rates can reach levels unimagined for other body products, including reports of payments of $50,000 or even $100,000.94 Compensation for egg providers differs from that for sperm providers in other ways, too. Women are typically paid some or all of their previously negotiated fees regardless supra note 29, at 363 (“Average donor compensation was $5890”). Both of these studies occurred while the ASRM’s price guidelines were in place, and so it is unclear whether these rates will change substantially in their absence. 85 ASRM, 2000 Financial Incentives for Oocyte Donors, supra note 4, at 219. 86 Id. 87 Almeling, supra note 66, at 320. 88 Johnson, supra note 29, at 364. 89  See Tamar Lewin, 10 Things to Know About Being a Sperm Donor, N.Y. Times, Nov. 3, 2016, https:// www.nytimes.com/2016/11/08/health/sperm-­donor-­facts.html (“Your odds of getting into Harvard or Stanford are higher than your chances of being accepted as a donor at the major sperm banks”); Fairfax Cryobank, https://fairfaxcryobank.com/ (“Fairfax Cryobank offers a large selection of high quality sperm donors; only 1 in 200 applicants make it through our rigorous screening process to become donors”). 90 Almeling, supra note 66, at 332. 91 Id. 92  Id. at 333. 93  Egg Donor Program, Donor Compensation, https://www.eggdonation.com/becoming-­an-­egg-­ donor/donor-­compensation/. 94  See Abbie Boudreau et al., “Premier” Donor Eggs Command High Prices for Desirable Genes, ABC News, Nov. 5, 2015, https://abcnews.go.com/Health/big-­business-­egg-­donation-­sparks-­controversy-­ pricing/story?id=34967399.

Incomplete Commodification in American Law   891 of how many eggs they produce.95 Contrary to sperm providers, who are paid for the volume of usable sperm they produce, egg providers are compensated more directly for their time and effort. Moreover, because women have historically contracted to produce eggs for a particular recipient, the relationship between the ultimate egg buyer and the egg seller is more “intimate.”96 Women whose eggs do not result in a pregnancy may be asked to accept a lower fee in the future,97 attributing the “failure” in part to the egg provider. Despite the significant sums involved, compensation for egg providers occupies a precarious role in the fertility industry. Unlike men, women are not advised that acting as a gamete provider will allow them to “[b]e [their] own boss!”98 Indeed, economic language is downplayed, rather than emphasized, where egg providers are concerned. Egg agencies “adorn their advertisements with images of plump babies and appeal to the joys of ‘helping’ infertile couples; some do not even list the amount donors will be paid.”99 In place of financial gain and economic exchange, egg agencies emphasize altruism and may reject potential egg providers if they seem too interested in the financial aspects of their role.100 More significantly, lawmakers and professional organizations have singled out egg provider payments as problematic. Since 1986, Louisiana law has stated that “[t]he sale of a human ovum, fertilized human ovum, or human embryo is expressly prohibited.”101 Indiana limits payments to “a woman donor of an ovum” to lost wages, travel and medical expenses, and “recovery time in an amount not to exceed four thousand dollars ($4,000).”102 No similar limitations exist for male sperm providers. More broadly, in 2000, the ASRM promulgated a report recommending maximum payments for women providing eggs for use in assisted human reproduction. According to the ASRM, “at this time sums of $5,000 or more require justification, and sums above $10,000 go beyond what is appropriate.”103 ASRM arrived at this sum by calculating an average hourly wage for men providing sperm for use in assisted human reproduction and multiplying that sum by the average number of hours required for egg provision.104 ASRM emphasized that “the provision of financial or in-­kind benefits does not necessarily discourage altruistic motivations” and that payments to egg providers should “reflect the amount of time expended and the burdens of the procedures performed.”105 Yet ASRM did not inquire into the basis for payments for sperm providers, nor did ASRM suggest “appropriate” limits for what these men should be paid. ASRM also did

95 Almeling, supra note 66, at 320 (“[W]omen . . . are paid regardless of how many eggs they produce”). 96  Id. at 333.    97 Id. 98  Cal. Cryobank, supra note 65. 99 Almeling, supra note 66, at 326. 100  Id. at 326, 328 (emphasizing altruism and observing that “‘girls who just want to lay their eggs for some quick cash’ are rejected”). 101  La. Rev. Stat. Ann. § 9:122. 102  Ind. Code Ann. § 35-­46-­5-­3. 103 ASRM, 2000 Financial Incentives for Oocyte Donors, supra note 4, at 219. 104 Id. 105 Id.

892   Natalie Ram not vary the amounts stated in its guideline between 2000 and 2016, when it agreed to remove specific payment amounts altogether as part of a class action settlement.106 Although the sums contemplated under the ASRM price guidelines are not insignificant, they may well have acted as a constraint on the market for human eggs. Most directly, ASRM sought to enforce this guideline in conjunction with the Society for Assisted Reproductive Technology (SART), an affiliated professional organization to which most American assisted-­reproduction clinics belong.107 SART compiles and publishes a list of egg agencies that comply with ASRM practice and ethics guidelines.108 Individuals interested in acquiring eggs from a third party for reproductive use rely on that list to identify agencies that “are acting in line with professional standards and are ‘safe’ choices.”109 Agencies that did not comply with ASRM’s price guidelines faced removal from SART’s list of approved agencies. Thus, while the ASRM price guidelines did not have the force of law, it carried significant weight in the industry. The ASRM price guidelines may also have indirectly affected compensation to egg providers, acting as an anchor around which acceptable rates were negotiated.110 “Anchoring” describes the unconscious human tendency toward “overreliance on an initial numerical reference point that causes absolute judgments to assimilate toward the initial value.”111 Thus, even where a suggested monetary payment does not have the force of law—such as ASRM’s price guidelines—its mere suggestion may affect the ultimate monetary agreements made by industry actors. In 2011, women who had previously sold their eggs under the guideline filed a lawsuit alleging that the payment cap violated antitrust laws by interfering in the market for human eggs and female labor.112 In February 2016, the parties settled, with ASRM and SART agreeing to remove the language suggesting a compensation cap.113 It is too soon to know whether the elimination of the ASRM price guidelines will affect compensation paid to egg providers. Public opinion accords with ASRM’s focus on fairness, rather than economic incentives, for egg providers. The same survey results that showed broad public support for monetary payments to “incentivize” sperm providers indicated support for payments to egg providers as “fair compensation for the medical risk undertaken” or “fair for the time and effort undertaken.”114 Fewer than half of survey respondents supported an 106  See Jacob Gershman, Fertility Industry Group Settles Lawsuit Over Egg Donor Price Caps, Wall St. J., Feb. 3, 2016, https://blogs.wsj.com/law/2016/02/03/fertility-­industry-­group-­settles-­lawsuit-­ over-­egg-­donor-­price-­caps/. 107  See SART, What is SART?, https://www.sart.org/patients/what-­is-­sart/(“SART? includes more than 90% of the ART [assisted reproductive technology] clinics in our country”). 108  See Johnson, supra note 29, at 360. 109 Id. 110  Cf. Daniel M. Isaacs, Baseline Framing in Sentencing, 121 Yale L. J. 426, 439 (2011) (describing the heuristic of anchoring and explaining how even non-­mandatory numerical reference points can shape outcomes in sentencing). 111  Id. at 439 (internal quotation marks and alterations omitted). 112  See Gershman, supra note 106. 113 Id. 114  Lee et al., supra note 73, at 1357 (reporting 65% support for financial compensation of egg providers as “fair compensation for the medical risk undertaken” and 62% support for financial compensation as “fair for the time and effort undertaken,” while only 44% supported financial compensation as an incentive to recruit more egg providers).

Incomplete Commodification in American Law   893 economic “incentive” justification for human egg providers.115 Thus, industry actors, professional regulatory bodies, and the public at large evince a common discomfort surrounding compensation for egg providers, even as each supports some measure of remuneration. This discomfort affects how egg providers are recruited and how their compensation is described and calculated. Yet, none of this discomfort is evident in the market for human sperm. As others have observed, this disjunction in attitudes about sperm and egg providers maps neatly to “gendered cultural norms.”116

3.3  Blood Human blood is perhaps the most interesting example of an incomplete market in the absence of a legal prohibition on compensation. As set forth earlier, there is no legal prohibition on paying people to give or transfer blood.117 In fact, there is a well-­established market in blood plasma in which those who give blood for this purpose are paid.118 And there is a market in moving blood from one place to another, to the tune of $4.5 billion each year.119 Blood arbitrage is big business in America.120 Meanwhile, nonprofit organizations charge hospitals upward of $250 per pint of whole blood.121 Hospitals that pass such charges on to patients may inflate them to $1,000 or more.122 But, by and large, those whose bodies are the source of that whole blood are not paid. Despite the lack of a legal prohibition on payment, there simply is no modern market for blood whose sources are paid.123 American hospitals will pay for blood from uncompensated donors, but not for blood from “paid donors.”124 Indeed, there persists a belief in America that paying people to give blood would degrade the quality of the blood supply without sufficient countervailing gains in quantity—a theory of “crowding out.”125 It was not always this way. Payments for blood used to include those whose blood flowed through the market. In the 1920s and 1930s, the “professional donor” emerged as 115 Id. 116  See Almeling, supra note 66, at 331. 117  See supra Section 1. 118  See Preston, supra note 22; Darryl Lorenzo Wellington, The Twisted Business of Donating Plasma, Atlantic, May 28, 2014, https://www.theatlantic.com/health/archive/2014/05/ blood-­money-­the-­twisted-­business-­of-­donating-­plasma/362012/. 119  See Carlyle, supra note 5. 120  Id. (describing General Blood, a for-­profit blood arbitrage business). 121  See id. (“A pint of blood might cost a hospital $210 in Wisconsin but $265 in New Jersey”); Matthew L. Wald, Blood Industry Shrinks as Transfusions Decline, N.Y. Times, Aug. 22, 2014, https:// www.nytimes.com/2014/08/23/business/blood-­industry-­hurt-­by-­surplus.html (“Nonprofit organizations collect whole blood from unpaid donors, but hospitals may pay $225 to $240 a unit, according to executives in the business, which covers a variety of costs, including testing”). 122 Wald, supra note 121. 123  See Preston, supra note 22 (“In practice, nobody really pays for blood, said Mario Macis, an economist at the Johns Hopkins Carey Business School who has studied incentives for blood donation”). 124 Id.; see also 21 C.F.R. § 606.121(c)(8)(v) (requiring that blood for transfusion be labeled “paid donor” or “volunteer donor” to reflect the presence or absence of monetary compensation). 125  See Richard Titmuss, The Gift Relationship: From Human Blood to Social Policy (1970).

894   Natalie Ram a heroic figure: a “husband-­provider who was so successful in the business [of selling his blood] that he could earn a fairly steady income in the big cities, without the need to engage in any other wage-­earning activity.”126 In the postwar years, the “replacement donor” emerged, accruing blood credits redeemable for future blood needs.127 Where the professional donor earned income for his family by giving blood, “the working husband and father during the Cold War could provide for his family by purchasing blood credits with his donation.”128 The replacement donor was every responsible family man. By giving blood as a “replacement donor,” he was “showing the same fiscal responsibility and ability to care for his family as he showed by maintaining a savings account.”129 Both of these figures, the professional and replacement donor, were not inherently male. After all, both men and women can give blood. But, in each instance, these remunerated individuals, giving blood either for pay or for credit, were “presumptively male.”130 Blood banks recruited replacement donors from fraternal organizations and work places; that is, settings where men dominated.131 Tellingly, early efforts to phase out compensation for blood providers, whether in cash or in kind, coincided with the first broad efforts to recruit women as blood providers, too. As the United States prepared to (and then did) enter World War II, blood banks working with the American Red Cross called on both men and women to do their part—and donate blood without compensation.132 The Red Cross impressed upon blood donors that their blood was “a gift . . . of a more personal nature than any contribution of money or articles could be.”133 As uncompensated blood “gifting” became the norm in the war years, women “could now be ‘champion blood donors.’”134 Indeed, women provided “about half of the blood collected” that the Red Cross collected for the war effort.135 By contrast, where patients owed blood debts and blood providers were compensated in cash or in kind, women were less frequently sought out and were less likely to give.136 To be sure, the modern default of uncompensated blood donation owes its status to more than simply a shifting conception of whose blood should be recruited for use. While the Red Cross and traditional blood banks initially diverged in their view about whether blood recipients should pay for the volume received, later conflicts focused attention on whether blood providers should be paid for the volume produced. Lawsuits seeking to hold blood banks and related medical actors liable for transfusions gone wrong and growing fears about infection spread through desperate blood sellers also

126  Swanson, supra note 27, at 43 (internal quotation marks omitted). 127  Id. at 108–109, 113. 128  Id. at 114. 129 Id. 130  Id. at 113. 131 Id. 132  Id. at 74–75. 133  Id. at 74 (internal quotation marks omitted). 134  Id. at 75 (quoting Harold Putnam, Victory Forum: Champion Blood Donors, Daily Boston Globe, Oct. 9, 1942). 135  Id. at 75. 136  Id. at 114.

Incomplete Commodification in American Law   895 disfavored the traditional model.137 By 1975, nearly all monetary compensation—for blood providers—had disappeared from the US blood supply.138 By the time the FDA enacted its requirement that blood be labeled “paid” or “volunteer,”139 that label was all but a foregone conclusion. Still, the transformation of the American blood supply from paid to unpaid coincided with the transformation of the prototypical American blood supplier from male to gender neutral. Here again, the need to recruit women helped to reframe blood donation as an altruistic act. Though a traditional discomfort surrounding women and markets is not the whole of the story, it is inextricably linked with it.

4 Conclusion Throughout the American system of body product exchange, incomplete commodification abounds. Where the law bars compensation, many still get rich, though the actual sources of body products—primarily posthumous individuals and their families—are notably largely uncompensated. Even where American law does not bar payment, free markets often do not emerge for the suppliers of valuable body products. Particularly where a body product is sourced from or associated with female bodies, altruism and fairness—rather than market-­driven incentives—take on a primary role. There may be good reasons to limit or avoid market mechanisms where body products are at issue.140 Glenn Cohen has synthesized the bulk of arguments against the commodification of body products into four types: Corruption, or the concern that markets in body products “will do violence to or denigrate our views of how goods are properly valued.”141 Crowding out, or the concern that “allowing payment for goods may change its social meaning in a way that discourages altruistic giving.”142 Coercion, exploitation, and undue inducement, or concerns that payments for body products may lead those of limited economic means to participate in efforts not otherwise consistent with their well-­being or desires.143 And unfair distribution, or the concern that permitting payments for body products will result in an unfair distribution of body products according to ability to pay rather than need.144

137  See Ram, supra note 6, at 499–501 (summarizing Swanson’s account of the shift from professional and replacement donors to a system of unpaid blood donation). 138  See Swanson, supra note 27, at 154–155. 139  21 C.F.R. § 606.121(c)(8)(v). 140  See generally Symposium, Organs and Inducements, 77 Law & Contemp. Probs., no. 3, 2014, at i. 141  I. Glenn Cohen, Regulating the Organ Market: Normative Foundations for Market Regulation, 77 Law & Contemp. Probs., no. 3, at 73. 142  Id. at 74. 143  Id. at 75–79. 144  Id. at 79.

896   Natalie Ram But there is reason to doubt the sincerity of many of these rationales. Such rationales are often inconsistent with markets for any actor in body product exchange. That is, insofar as these rationales have force, they ought to limit or prevent the development of markets surrounding body products altogether. That is what NOTA sought to achieve in its prohibition of “valuable consideration” for any person involved in the acquisition, receipt, or transfer of a transplantable organ.145 But various intermediaries in the incomplete markets for human tissues, both within and beyond NOTA’s statutory and regulatory terms, defy such a broad prohibition on money-­making. Whether they are nominally nonprofit or admittedly for-­profit, many of these entities make bank. Tissue processors working with NOTA-­regulated body products are nonetheless part of a multibillion-­dollar industry.146 So, too, are sperm banks and egg agencies.147 And blood banking and arbitrage supports a $4.5 billion a year industry.148 In other words, markets in body products exist. But these markets are incomplete. Sources of body products often occupy a singular role, as both the most essential and least well-­compensated actor in body product exchange. And that role is typically more keenly enforced when female bodies are at issue.

145  42 USC § 274e. 146  See Katz, supra note 8, at 943. 147  See Donna Rosato, How High-­Tech Baby Making Fuels the Infertility Market Boom, Money, July 9, 2014, http://time.com/money/2955345/high-­tech-­baby-­making-­is-­fueling-­a-­market-­boom/. 148  See Carlyle, supra note 5.

chapter 42

Of M a r k ets a n d Pr i nciples The European Perspective Stéphanie Hennette Vauchez

Blood, tissue, cells. How are they retrieved, and how do they circulate in Europe? Asking the question at the European level allows us to unearth the ongoing tug of war between two main regulatory paradigms in the field of human body parts and cells: a human/fundamental rights-­inspired paradigm on one hand, and a market-­inspired one on the other hand. It also allows us to make more sophisticated (and indeed, more complex) a familiar trope of the comparative work in the field of health and biomedical law that readily contrasts a European ‘human rights’ model of regulation to a North American ‘market’ one. It is all the more important to refine and relativize this common portrait that often carries along a host of assumptions that supposedly are inherent to both human rights and market logic. For instance, many American authors readily read the human rights approach as more prone to conveying and perpetuating traditional gender stereotypes, whereas the market would be more morally neutral in that respect.1 Conversely, the common European understanding of the American model focuses on the undue penetration of brutal capitalist logic in all spheres of human life and the issues of axiological degradation that ensue—not to mention the many dimensions of inequality it triggers.2

1  One example can be found in: M. Ertman, ‘The Upside in Baby Markets,’ in M. Goodwin (ed.) Baby Markets, Money and the New Politics of Creating Families (Cambridge University Press 2010), p. 23: ‘market mechanisms provide unique opportunities for law and culture to recognize that people form families in different ways. If state or federal law, rather than the laws of supply and demand, determines who can have children using reproductive technologies, then many single and gay people will be excluded from this important life experience.’ 2  See, e.g., C. Lafontaine, Le corps-­marché. La marchandisation de la vie humaine à l’heure de la bioéconomie (Seuil 2014).

898   Stéphanie Hennette Vauchez This chapter does not seek to dispute the fact that the market is more extended and pervasive in the United States than it is in Europe. The most vivid illustration of this true difference is provided by the status of the actual actors that evolve in the field of ­biomedicine and, as far as this particular chapter is concerned, in that of blood, tissue, and cells circulation. These actors are indeed mostly private and for profit in the United States, whereas they are either public or heavily regulated in Europe and requested to respect a largely not-­for-­profit framework. Together with its sister chapter,3 however, this contribution seeks to mitigate and refine both the contrast between the two regulatory models and the readily made assumption that Europe befits the human rights one and the United States the market one. While Natalie Ram’s chapter tells an American story of incomplete commodification, this one focuses on the reverse movement in Europe, one that sees the logic of the market creeping into the circulation of human blood and cells. Both chapters underline the gender dynamics that are at stake in these movements that they depict and analyze. This chapter operates in three steps. First, it recounts a (short) structural history of legal regulation of the circulation of human blood and cells in Europe. It then turns to an analysis of the applicable rules and seeks to unveil the extent to which they contribute to defining a ‘European way of market’—but a market nonetheless. Finally, the argument turns to the specific example of the circulation of reproductive cells as this seems to pose the greatest quantitative and qualitative—and gendered—challenges to existing forms of regulation.

1  A Short Structural History of European Biomedical Law The legal regulation of biomedicine emerged in many European countries as a primarily professional issue: historically, medical boards and codes of ethics have largely set the frame in matters of abortion, assisted reproduction, end-­of-­life decisions, and other issues. Although this professional mode of regulation still plays a considerable role in some countries,4 biomedicine was progressively institutionalized. This process first took the form of the multiplication of ethics committees (in the 1980s) and, later, that of legislative regulation as parliaments stepped in significantly throughout many European countries. From a substantive standpoint, human rights have been the dominant framework of these forms of legal regulation of biomedical issues.5 The famous book authored by 3  Ram, this volume. 4  Assisted medical reproduction, for instance, is hardly regulated by state or federal laws in Germany but essentially by the federal medical board. 5  T. Hervey, J. McHale, Health Law and the European Union (Cambridge University Press 2004), p. 66: fundamental rights ‘have framed the discourse’ over biomedical issues in general, in Europe.

Of Markets and Principles: The European Perspective   899 Richard Titmuss in 1970 based on his comparison between models of blood transfusion regulation in the United States and Britain positively analyzed the prevalence of axiology in the British model (voluntary donation, not-­for-­profit overall organization of the system). Following the lines of Titmuss in turning the technical study of the regulation of blood donation into a matter of social theory, it is tempting to underline the ways in which legal and political actors have tended to insist on the very high moral stakes that are associated with the regulation of biomedical issues in European societies. Abortion, embryonic research, organ transplantation or assisted reproduction seem to never only be about just that: deciding over those issues is always also framed as implying the self-­ definition of national identities, a conception of humanity, etc.6 Against this context in which European countries have tended to approach biomedical issues as ‘fundamental’ issues whose regulation implies nothing short of national identity and the future of humanity, human rights have appeared to provide a principled paradigm that was well equipped to do just that. Human rights thus constitute the dominant paradigm of most European domestic legal frameworks in the field of biomedicine. Be they constitutional or legislative, they are articulated around central normative concepts such as the right to health or the human dignity principle. This is further echoed at the supranational level. The 1997 Council of Europe Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine (Oviedo Convention) is emblematic of this principled approach, as is the EU Charter of Fundamental Rights that was adopted in 2000 and addresses biomedical issues in its Article 3 under the heading ‘Dignity.’7 The supranationalization of human rights standards, however, has not really challenged the domestic diversity of regulatory options pursued under the human rights paradigm. The European Court of Human Rights, for instance, is remarkably deferential when it comes to biomedical issues: For example, it rules that the European Convention on Human Rights (ECHR) ‘cannot . . . be interpreted as conferring a right to abortion.’8 The Court also acknowledges that, as the question is highly sensitive, states retain a wide margin of appreciation in the regulation of gametes procurement and assisted reproductive technologies9: as a consequence, Austria may outlaw third-­party gametes donation. Is it then impossible to grasp a European regulatory model or trend? Is there nothing to be said beyond the notion that particular domestic normative choices are derived 6  The national(ist) dimension of the debate on human embryonic stem cells for instance was very conspicuous in Germany, where research on such cells was generally forbidden save on imported ones. 7  The Oviedo Convention was adopted by the Member States of the Council of Europe, an organization founded in 1949 for the preservation of peace and the promotion of human rights and whose backbone in the 1950 European Convention of Human Rights. The EU Charter of Fundamental Rights was signed in 2000, and signals the elevation of fundamental rights within the narrower European Union, an organization that was initially essentially economic. 8  ECHR, GC, 16 Dec. 2010, A, B and C v Ireland, no 25579/05. 9  ECHR, GC, 3 Nov. 2011, S and H v Austria, no 57813/00.

900   Stéphanie Hennette Vauchez from a bundle of fundamental principles such a human dignity and the integrity of the person? This was probably the case until recently. Since the beginning of the twenty-­first century, however, important changes, both empirical and normative, have taken place. Empirically, the social practice of biomedicine has undergone important evolutions, among which, most importantly, is its globalization. Blood, tissue, and cells now circulate transnationally; there is a global biomedical economy that rests on the many clinical and therapeutic uses of human material.10 This, in turn, has triggered a greater relevance of a level of regulation that had remained rather neglected and understudied in Europe by scholars of biomedicine: that of the European Union. The affirmation of the EU in the field of biomedicine is all the more important in that it has contributed to tilting the substance of the regulatory paradigm because the EU is not predominantly about human or fundamental rights but rather, as a primarily economic organization, about the completion of the internal market. The stepping in of the EU has thus led to a two-­step movement. First, EU legal actors have affirmed the theoretical and practical relevance of the legal categories of the market (and, in particular, of the freedoms of circulation of persons [patients] and services [including medical services]) in the field of biomedicine.11 It is mainly throughout the freedoms of circulation jurisprudence that the Court of Justice of the European Union (CJEU) has established the right of persons as patients to travel and seek medical care in EU countries other than their countries of residence if or when such treatment was not nationally available to them. The Court also established that medical services were services and that there was no reason to consider that they were services of a special nature (even abortive services12), hence the full applicability of free movement principles. As a result, cross-­border care and importation/exportation of human blood, tissue, and cells became coextensive to the freedoms of circulation on which the EU’s internal market is built.13 This affirmation of the relevance of market-­ oriented legal categories was further consolidated by early legislative interventions such as the Patent14 and Clinical Trials15 directives adopted in 1998 and 2001. Both these texts have Article 114 of the Treaty on the Functioning of the European Union (TFEU) as a legal basis, which means that they were adopted and justified as necessary to the com10  C. Waldby, R. Mitchell, Tissue Economies: Blood, Organs and Cell Lines in Late Capitalism (Duke University Press 2006). 11  Kyriaki-­Korina Raptopoulou, EU Law and Healthcare Services. Normative Approaches to Public Health Systems (Kluwer 2015). 12  CJEU, 1991, Grogan v the Society for the Protection of Unborn Children, C-­159/90. The Court further established that the free movement of services encompasses the free movement of information pertaining to the said services, including in countries others than the ones they are provided in. 13  These fundamental freedoms of circulation do not, however, serve the interests of those who wish to donate human material, for the person who is entitled to freedom of circulation is essentially the patient (i.e., a person who receives care rather than one who provides or donates blood, cells, or organs). 14  Directive 98/44/EC on the legal protection of biotechnological inventions. 15  Directive 2001/20/EC on the approximation of the laws, regulations, and administrative provisions of the Member States relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use.

Of Markets and Principles: The European Perspective   901 pletion of the internal market, essentially because the diversity of national legislations is framed as an obstacle to the market. Second, evolutions structural to the EU towards newer raisons d’être (including fundamental rights and public health) have generated a new approach to biomedical issues, one that can be portrayed as a hybrid between the market and a version of human rights-­inspired considerations. First and foremost, the adoption of the EU Charter of Fundamental Rights in 2000 confirmed the commitment of the EU to fundamental freedoms, and, in fact, the 2000 Charter does not ignore biomedical issues.16 Furthermore, after the Maastricht Treaty elevated each European citizen’s right to healthcare, the EU also acquired competences in the field of health. Article 168 of the TFEU now endows the EU with (minimal) harmonization competences geared towards public health. An important and dynamic legislative movement ensued, as the Blood,17 Tissue,18 and Organs19 directives adopted in 2003, 2004, and 2008, respectively, are all based on Article 168. These texts are illustrative of a form of hybridization of human rights and the market as regulatory paradigms in that, although they are predominantly concerned with a logic of sanitary risks (individual and collective), they also are geared towards enhancing market operationality. As Hervey and McHale have it, despite the market paradigm for the legislation, ‘in European traditions, parts of the human body are not “products” and their regulation cannot therefore be undertaken in terms of marketor trade-­based risk to consumers. Rather, ethical or human rights frameworks are applied.’20 Later, the 2011 directive on cross-­border care also expressed this hybridization and specifically European twist to the market, from its very making (the initial version prepared by the administrative unit of the European Commission concerned with the internal market [DG Market] was progressively amended in a direction that led to a greater implication of the Commission’s health and consumer protection unit [DG Health & Consumer Protection]) to its contents (incorporation of issues greater than market completion, including public health and patients’ rights issues). Tellingly, the directive has a dual legal basis, made of both Article114 (the EU’s internal market competence) and Article 168 of the TFEU. This short structural history of the legal regulation of biomedicine in Europe thus illustrates the tug of war between human rights and the market as regulatory paradigms and the eventual hybrid settlement. The following section looks at the substantive content of the norms applicable to blood, tissue, and cells in Europe. It is 16  Article 3. 17  Directive 2002/98/EC setting standards of quality and safety for the collection, testing, processing, storage, and distribution of human blood and blood components. 18  Directive 2004/23/EC on setting standards of quality and safety for the donation, procurement, testing, processing, preservation, storage, and distribution of human tissues and cells. This directive has been followed by several complementary ones: Directive 2006/17/EC (technical requirements), Directive 2006/86/EC (traceability), Directive 2015/565/EC (coding), and Directive 2015/566/EC (imported material). 19  Directive 2010/45/EU on standards of quality and safety of human organs intended for transplantation. 20  T. Hervey, J. McHale, European Union Health Law, 2nd ed. (Cambridge University Press 2015), p. 385.

902   Stéphanie Hennette Vauchez contended that, as elements detachable from the embodied person, they do indeed circulate within something like a market—albeit one that is contained to limit profit and commodification.

2  The Circulation of Blood, Tissue and Cells: A European ‘Way of Market’? In Europe, most of the relevant legal texts pertaining to the legal status of human blood and tissue converge in establishing that although they are to circulate within a market, this circulation ought to operate in a contained way. They are marketized in that they are procured and exchanged, including internationally (imported and exported), as essential components of a number of medical services. Although remuneration for these is possible, their marketization remains contained by the principle of voluntary unpaid donation (VUD) and the norm of a not-­for-­profit sector that are elevated as founding principles governing the retrieval, circulation, and exchange of blood and tissue in Europe. This legal standard operates for blood, tissue, and cells retrieved from living and deceased donors, although this chapter will insist that it is being mitigated and accommodated to some forms of market creep in the case of living donations. To start, the legal rules for procurement of blood, tissues, and cells parallel the rules for organ procurement. Living donors of body parts must give consent, and they must do so voluntarily, without remuneration (though they may receive compensation for the costs incurred in donating). While living donors can provide a limited range of body parts beyond organs—blood, bone marrow, and gametes—deceased donors can supply a variety of parts, including heart valves, bone, skin, tendons, ligaments, corneas, and veins. For deceased donors, consent to donation is presumed in some countries unless the decedent or family has opted out of donation. Presumed consent countries include France, Greece, Sweden, Slovenia, or Portugal. Still, even under presumed consent regimes, family consent often is obtained in practice. In other countries, consent for posthumous donation must be given by the donor (while alive) or the family after death. As with living donation, posthumous donation must be voluntary, with no payments to either the donor before death or to the next of kin after the donor dies. In some countries, there are further limitations on the kinds of cells that can be retrieved from deceased donors. In France, for instance, only skin, corneas, and cortical bones are retrievable. With regard to the voluntary nature of donation, the most solemn articulation of these principles certainly stems from Article 21 of the important Council of Europe Oviedo Convention of 1997, which reads: ‘the human body and its parts shall not, as such, give rise to financial gain.’ The Convention’s official explanatory report ­clarifies that

Of Markets and Principles: The European Perspective   903 under this provision organs and tissues proper, including blood, should not be bought or sold or give rise to financial gain for the person from whom they have been removed or for a third party, whether an individual or a corporate entity such as, for example, a hospital. However, technical acts (sampling, testing, pasteurisation, fractionation, purification, storage, culture, transport, etc.) which are performed on the basis of these items may legitimately give rise to reasonable remuneration. For instance, this Article does not prohibit the sale of a medical device incorporating human tissue which has been subjected to a manufacturing process as long as the tissue is not sold as such. Further, this Article does not prevent a person from whom an organ or tissue has been taken from receiving compensation which, while not constituting remuneration, compensates that person equitably for expenses incurred or loss of income.  (for example as a result of hospitalisation)

Nor does it prohibit a tissue establishment from charging for the costs of processing heart valves, skin, bone, or other tissues for use in transplantation or grafting onto a patient. Article 21 of the Oviedo Convention thus encapsulates the two main pillars of the European legal regime for blood and tissues: VUD (no direct remuneration of donors) and a not-­ for-­ profit economy (limits on the profitability of subsequent exchanges and acts of circulation). But VUD is by no means an invention of the Council of Europe. These principles echo, first and foremost, national regulatory traditions. The notion that this regime of procurement should be favored also appeared early on in the EU, for instance in a 1993 opinion of the European Group of Ethics (EGE) on blood, in which it insisted that nobody should be allowed to make a profit from a donor’s blood and that when blood was used for the manufacture of derivatives, neither the supplier nor the manufacturer should be allowed to charge more than the actual costs incurred.21 The EGE later transposed this analysis to other substances of human origin. Moreover, VUD did not remain a vague ethical principle; it was later transposed in the directives that were adopted in  the 2000s pertaining to blood,22 tissues,23 and organs,24 all of which echo these principles. Directive 2002/98 on blood seeks to insert itself into a history of ‘modern blood-­ transfusion practice’ that is recounted as having been founded on the principles of voluntary donor services, anonymity of both donor and recipient, benevolence of the donor, and absence of profit on the part of the establishments involved in blood transfusion services.25 The Directive’s Article20 thus rules that ‘Member States shall take the necessary measures to encourage voluntary and unpaid blood donations with a view to ensuring that blood and blood components are in so far as possible provided from such

21  European Group of Ethics, 12 March 1993, Opinion no 2: Products derived from human blood or human plasma. 22  Directive 2002/98/EC. 23  Directive 2004/23/EC. 24  Directive 2010/45/EU. 25  Recital 20. See also Recital 23: ‘Voluntary and unpaid blood donations are a factor which can contribute to high safety standards for blood and blood components and therefore to the protection of human health.’

904   Stéphanie Hennette Vauchez donations.’26 Although the Directive does not make VUD mandatory, it is now ‘recognized in all Member States, albeit enforced to different degrees,’27 as indicated in the 2016 report on the implementation of the Directive. A comparable regulatory framework was subsequently adopted for human tissue and cells28—a category that the 2004 directive applies not only to ‘classical’ replacement tissue (e.g., skin, bones, cardiovascular tissue) but also to ‘hematopoietic peripheral blood, umbilical-­cord (blood) and bone-­marrow stem cells, reproductive cells (eggs, sperm), foetal tissues and cells and adult and embryonic stem cells.’29 Nonbinding, but persuasive and relevant to interpretation, Recital 3 of Directive 2004/23 specifies that the availability of such tissue and cells is dependent on donation; Recital 4 subsequently burdens all Member States with the task of organizing campaigns around the theme ‘we are all donors.’ More specifically, Recital 18 makes it clear that VUD is both a principle and a philosophical choice,30 while Article 12 of the directive further states that that ‘Member States shall endeavor to ensure voluntary and unpaid donations of tissues and cells.’ The 2016 report on the implementation of the Directive provides yet another justification of this principle: that of efficiency with respect to the main goals of the Directive (improving and maintaining high standards of quality and safety of tissues and cells). It 26  The EU seems to work with the COE definition of VUD: Recommendation No R(95)14 of the Committee of Ministers to Member States on the protection of the health of donors and recipients in the area of blood transfusion, Appendix, Article 2 states ‘if the person gives blood, plasma or cellular components of his or her own free will and receives no payment for it, either in the form of cash or in kind which could be considered a substitute for money. This would include time off work other than that reasonably needed for the donation and travel. Small tokens, refreshments and reimbursements of direct travel costs are compatible with voluntary, non-­remunerated donation.’ 27  Commission Staff Working Document on the implementation of the principle of voluntary and unpaid donation for human blood and blood components as foreseen in Directive 2002/98/EC on setting standards of quality and safety for the collection, testing, processing, storage, and distribution of human blood and blood components and amending Directive 2001/83/EC: SWD(2016)130final, 21 April 2016. 28  The 2004/23 directive on Tissues and Cells of 31 March 2004 (as well as further amendments through directives 2006/17 and 2012/39) applies to all human cells intended for human application (as opposed to cells used for research purposes), including reproductive (sperm and eggs) and foetal and embryonic cells and tissues (Poland, whose 2005 national act of transposition explicitly excluded foetal and embryonic sperm, was found in breach of its obligations of transposition: CJEU, 11 June 2015, Commission v Poland, C-­29/14). Another state is facing an infringement procedure. The 2004/23 directive does not apply in cases where human tissues and cells are not intended for human use. When human tissues and cells are used in the manufacture of medicinal products that are either a gene therapy medicinal product, a somatic cell therapy medicinal product, or a tissue-­engineered product, Regulation (EC) No. 1394/2007 of the European Parliament and of the Council on advanced therapy medicinal products (‘ATMP Regulation’) applies. For such products derived from human tissues and cells, Directive 2004/23/EC and its implementing directives apply only to their donation, procurement, testing, etc. 29  Recital 7 of the directive. Adoption of directive 2004/23 was marked by numerous contentious debates over the issue of human embryonic stem cells; see T. Hervey, H. Black, ‘The EU and the Governance of Stem Cell Research’ (2005) 12(1), Maastricht J Comparative L 11. 30  ‘As a matter of principle, tissue and cell application programmes should be founded on the philosophy of voluntary and unpaid donation, anonymity of both donor and recipient, altruism of the donor and solidarity between donor and recipient. Member States are urged to take steps to encourage a strong public and non-­profit sector involvement in the provision of tissue and cell application services and the related research and development.’

Of Markets and Principles: The European Perspective   905 insists that VUD for tissues and cells should be favored for reasons pertaining both to ethics and safety standards. The reasoning is based on the premise that if donor payment were allowed, some individuals could find the monetary remuneration so important that they might hide relevant medical and/or behavioral information.31 These multiple affirmations and articulations of the VUD principle do not, however, constitute an absolute shield to the market creeping in, especially via the blurring of the lines that separate remuneration and compensation. For instance, the 2004 Directive on human tissue and cells insists that its foundational VUD approach does not exclude all forms of financial payment, and illustrations of the relaxing of the VUD standard can be found in several more detailed legal regimes concerning living donations. In fact, ‘compensation’ is allowed, and defined, in the Directive as strictly limited ‘to making good the expenses and inconveniences related to the donation.’32 Interestingly, this is one of the issues pertaining to which the 2016 Assessment report on the implementation of the Directive underlines significant degrees of variance in national transposition. The report reads: Most of these countries consider it acceptable to provide a financial compensation covering the justifiable expenses related to the donation (e.g. travel, loss of income, medical expenses). Estonia specified that both donors and recipients are entitled to subsidies for temporary incapacity from work by the Estonian Health Insurance. Italy referred to their national law which allows time off work and provides insurance coverage for stem cell donors. Lithuania reiterated that donors are entitled to medical leave for day(s) required for donating cells, tissues or organs.33

In other words, although blood, plasma, and tissue donation themselves remain unpaid (with an exception for plasma in Germany, donation of which can be remunerated up to approximately €25), there are a number of regulations that ensure that the act of donation is financially neutral for the donor: travel costs, time off work, and other forms of insurance mechanisms kick in to cover all the costs associated with donation. As this concept of ‘compensation’ grows more robust, it is sometimes considered that the founding principle of VUD is under increasing pressure. Recurring shortage of blood and tissue seems to lead national actors to develop incentives to donation and certainly 31  Report from the commission to the European Parliament, the Council, the European Economic and Social Committee, and the Committee of the Regions on the implementation of Directives 2004/23/EC, 2006/17/EC, and 2006/86/EC setting standards of quality and safety for human tissues and cells, COM(2016)223 final, 21 April 2016. 32  Article 12 of the Tissue directive. 33  Although it is beyond the scope of this chapter, the Organs directive follows the same lines. Article 13 of the directive clearly establishes that ‘Member States shall ensure that donations of organs from deceased and living donors are voluntary and unpaid,’ even though, again, ‘the principle of non-­payment shall not prevent living donors from receiving compensation, provided it is strictly limited to making good the expenses and loss of income related to the donation.’ Furthermore, ‘member States shall ensure that the procurement of organs is carried out on a non-­profit basis.’ Other provisions of the directive, especially in the text’s recitals, express the commitment of the EU to combating organ trafficking as well as undue commodification of the human body.

906   Stéphanie Hennette Vauchez generates greater policy and lobbying attempts at redrawing the contours of the admitted forms of ‘compensation.’ The Council of Europe acknowledges these contemporary challenges to VUD and calls for caution: ‘voluntary unpaid donation, long promoted as the only ethical basis for donation of bodily material, should continue to have a central role in the donation process of any type of tissue or cell. Compensation to donors should cover only justifiable expenses and loss of income, and should not act as a direct or indirect incentive or inducement.’34 As it acknowledges the development of ‘incentive’ models that potentially go further than strict ‘compensation’ models for donors,35 the Council of Europe’s 2017 Guide to the Quality and Safety of Tissues and Cells for Human Application insists that payments that do not constitute a financial gain or a comparable advantage [are not prohibited], in particular: a. compensation of living donors for loss of earnings and any other justifiable expenses caused by the removal or by the related medical examinations; b. payment of a justifiable fee for legitimate medical or related technical services rendered in connection with transplantation; c. compensation in cases of undue damage resulting from the removal of tissues or cells from living persons.36

The report nonetheless underlines that incentives so far have essentially targeted professionals and health establishments in which donations actually take place, rather than donors themselves. In Europe, it is thus hardly questionable that blood, tissue, and cells are usually donated without payment. Two caveats are in order, however. One has to do with the specificity, in that respect, of reproductive cells, an issue addressed in the final section of this chapter. The other relates to what happens after donation. Once retrieved, blood, tissues, and cells are further processed, stored, and handled by a series of actors that are both numerous and of variegated nature. Again, there is no European standard in this respect. There certainly are a great many public actors in the field who often operate on the basis that no charge is billed to the end users. But public actors may assess fees to the hospitals that use the tissues or cells in patient care, and private actors also exist. Most of the private actors are not-­for-­profit. They charge for the processed tissues and cells that they provide, but typically they do so only to cover their operating costs and do not seek financial surplus, or they seek only a surplus needed for 34  Id., p. 39. 35  As regards this distinction, the Council of Europe’s guide refers to the works of the British Nuffield Council on Bioethics that labels ‘compensation’ those interventions ‘which typically involve removal of various disincentives to act and, in doing so, remove countervailing concerns that may hinder potential donors from acting on their altruistic motivations’ and ‘incentives’ those ‘non-­altruist-­ focused interventions’ which are targeted at potential donors who have no strong motivation to help others through donation of their bodily material and who, therefore, if they are to donate, need to be provided with different reasons for action, perhaps in the form of payment or ‘incentive’ going well beyond the reimbursement of expenses.’ 36  Council of Europe EDQM, Guide to the Quality and Safety of Tissues and Cells for Human Application, 3rd ed. (2017), p. 38.

Of Markets and Principles: The European Perspective   907 investment. In some cases, a not-­for-­profit tissue establishment will cover operating costs by supplementing its fees with charitable donations. On the margins are the private for-­profit actors who do represent market-­oriented inroads into the VUD ideal. Overall, they all need to be licensed to operate in the field, and there exists an EU-­wide compendium of Tissue Establishments. Retrieval of systematic and relevant information as to the concrete economical model on the basis of which they operate is, however, challenging; the best known collection of data in this respect is the 2015 commissioned report Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU.37 The report insists that most Tissue Establishments in Europe operate with only limited cost awareness. There is a very wide variability of prices and fees that are charged throughout the EU, and it seems that they are fixed on the basis of policy choices rather than reflecting actual costs. This is turn hinders the development of intense cross-­border circulation of tissues, especially as most of them are used within wider hospital or medical-­led operations of transplantation. This matters to the European Commission because creating an EU-­wide market is one of the elements that justifies the EU legislation. In that regard, the goal of domestic self-­sufficiency, underpinned by noncommercialized models, pulls against the EU’s goal of circulation, which would be assisted by a more commercialized model. In that respect, there might also be a specificity of the subcategory of hematopoietic cells which can potentially be used directly by consumers, sourced from anywhere in the EU. The removal and banking of cord blood for potential future therapeutic use is a case in point as it is an important trigger for the development of private biobanks in Europe.38 A consumer-­led model sits under the creation of the EU’s internal market in many ordinary products: it may be that hematopoietic cells are, in effect, slowly joining that group through consumer behavior. Basically, the core principle that structures the clinical and therapeutic uses of blood, tissue, and cells in Europe is a self-­proclaimed ‘ethical’ standard of VUD: donations are unpaid and generally undirected, and donors consent not only to the donation but also to the future uses of their material. However, as the European Union seeks to favor access to and circulation of human bodily material, mitigations, if not exceptions, to the 37  European Commission, Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU, Final Report, EAHC/2012/Health/19, Contract no 20126301, 2015. 38  European Commission, Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU, op. cit., p. 132: ‘Autologous cord blood banking, or private banking has grown into a private industry, predominantly used by economically advantaged families. . . . The cost of banking is solely covered by families and ranges between EUR 2000–2500 for approximately 20 years of storage. It is estimated that the volume of private cord blood samples stored is outnumbering the number of public units, however, the use in transplantation is reported sporadically. Private cord blood banks have grown into major industries; they are often active in more than one country, and have huge inventories. . . . In the EU 28 Member States there are approximately 123 non-­public cord blood banks active, and it is known that a number of them are listed on the stock market, however precise figures on actual market value are unknown. Apart from the storage of cord blood, private banks also offer paid-­for cryopreservation of stem cells from dental pulp, adipose/lipo or fat tissue and stem cells from adult bone marrow and blood (see also http://parentsguidecordblood.org).’

908   Stéphanie Hennette Vauchez said principles are multiplying. Minors are increasingly being attracted into the realm of potential legal donors. The rule according to which donors are generally to be informed of the kinds of uses of their material (e.g., therapeutic, clinical) is also under pressure: in some countries it now falls on donors themselves to check whether there are changes in the planned use of the material. Moreover, the pressure on the unpaid dimension of donations in Europe is increasing. As far as pressure on the goal of VUD, distinctions need to be made, however, according to the nature of the material: while blood remains strictly separated from any kind of commercial paradigm, the principles approach seems more mitigated as far as other tissues and cells are concerned. As we have seen, blood operates in a particular way, historically, normatively, and empirically. While national self-­sufficiency was the goal when blood donation really took off before World War II, by the end of the 1970s, blood circulation had internationalized and the market became populated with both public and private actors. But ‘notwithstanding the changing reality of the European blood market, however, the commitment to voluntary unpaid donation remained central to national, as well as European, blood policy.’39 The comparative study of domestic regulatory frameworks shows that blood donation remains in principle unpaid in all European countries, and it is the expressed policy preference of Directive 2000/23. As far as other tissues and cells are concerned, although hostility to monetization similarly permeates the legal and institutional framework, it does seem that the stand­ ard of VUD is experiencing greater pressure. Replacement tissues (e.g., cardiovascular, musculoskeletal, ocular, skin), which are in large part removed from deceased donors, lead to no payment to donors. But they then circulate an arcane institutional pathway until they reach their final destination, and the financial/economic aspects of the exchanges that take place at that stage remain difficult to ascertain. This aspect of the circulation of hematopoietic tissues and cells (e.g., bone marrow, peripheral blood, cord blood) seems better known—and significant indeed40—if only because it has triggered an important development in the very category of private biobanks. But it is reproductive cells in particular that play an important role in the challenges VUD is facing: in Europe as elsewhere, the emerging field of gametes circulation and exchange is resembling a market—much more than that of blood or other tissues and cells. Reproductive cells do indeed seem to pose a particular challenge to the hybrid model of regulation that this chapter has claimed governs the circulation of blood and other tissue and cells. The hybrid model implies that although the legal tools and mechanisms of the market have affirmed their hold, the approach remains principled, and notions of commodification and profit are tentatively contained. As far as gametes are concerned, not only are there more private actors operating within the system, but the legal stand­ ard of compensation also is at greater pains to contain the monetization that is taking 39  A.-M. Farrell, ‘Is the Gift Still Good? Examining the Politics and Regulation of Blood Safety in the European Union’ 2006 Med L Rev 155. 40  The 2015 European Commission report indicates an average cost of €12,000–25,000 euros per hematopoietic donor product.

Of Markets and Principles: The European Perspective   909 place. The pressure in demand seems to be tilting the overall functioning of the system towards a greater creeping in of the market. The gendered dynamics that make the circulation of gametes specific and more favorable to market creep must be underlined.

3  The Pressure of Desire: The Circulation of Gametes Like all topics related to the procurement, banking, circulation, and exchange of substances of human origin, the regulatory landscape in Europe for gametes is extremely diverse and accounts for a host of different domestic normative options. For instance, while there is no sperm bank in Ireland, Denmark prides itself in being home to the world’s largest existing one. Belgium imports 63% of all its sperm but exactly that same share is used to treat foreign patients.41 Regardless of their content, however, all these domestic rules seem to find themselves under the increasing pressure exerted by the pragmatic view that markets where individuals may freely give away, choose (select), and obtain reproductive material and services are more desirable than human rights-­ inspired legal regulation. That this idea is gaining traction needs to be unpacked in terms of the gender dynamics that feed it because both the prevalence of legal rules restricting access to assisted reproduction technology (ART) services to heterosexual couples and the gendered biases in the legal treatment of ‘gametes’ in many of the existing legal regulations explain much of the pressure that is being exerted on the existing regulations. Increasingly, domestic regulatory models that had been elaborated in the last decades of the twentieth century are being challenged by claims pertaining to the illegitimacy of rules limiting access to gametes and ART to heterosexual couples that remain strongly in place in countries such as France, Italy, Germany, Austria, or Slovenia. Progressively, and although not directly as a result from transnational European law,42 a number of countries have abandoned this heteronormative model and opened up access to ART to same-­sex couples (e.g., Greece, Spain, the United Kingdom, Belgium, the Netherlands). These evolutions are very important in two main ways. First, they express the increasingly compelling force of the notion that reproductive issues are best left to individual 41  G. Pennings, ‘Import and Export of Gametes,’ in S. Lundin, M. Petersen, CKroløkke, E. Muller (eds.) Global Bodies in Gray Zones (Stias 2016). 42  The European Court of Human Rights has not ruled that legal regulations restricting access to ART to heterosexual couples were in violation with the ECHR; a case against France was recently deemed inadmissible (ECHR, Charron and Merle-­Montet v France, no 22612/15). It must be noted however that although the ECtHR has ruled that the absence of any form of civil union accessible to same-­sex couples was a disproportionate infringement on the right to private and family life (Article 8: ECHR, 21 June 2015, Oliari and others v Italy, no 18766/11 and 36030/11), it never ruled that marriage ought to be accessible to both opposite-­sex and same-­sex couples.

910   Stéphanie Hennette Vauchez autonomy. This notion runs directly counter to the gendered norms that are expressed and indeed embedded in traditional legal institutions such as marriage, adoption, and (historically) many legislative rules governing ART in Europe. Second, they reinforce the diversity in European domestic legislations and this, in turn, encourages cross-­ border circulation of patients and services.43 Increasingly, same-­sex couples travel to get married abroad,44 lesbian women travel to access sperm, infertile women as well as homosexual men travel to access ova, and courts are ordering that sperm samples be exported to countries where posthumous insemination is legal for widows to be able to access this particular form of ART.45 To the extent that although ‘very limited data is available to indicate flows of gametes,’ ‘there is considerable cross-­border exchange of sperm.’46 Such circumvention tourism47 is in fact supported and strengthened by European legal rules and categories. At the level of the European Union, as explained earlier, both the free movement principles and later legislation pertaining to cross-­border care explicitly acknowledge the fact that the European market allows patients to seek and obtain abroad care that is otherwise illegal or unavailable in their country of residence and/or nationality. Interestingly, the interpretation of the ECHR by the Strasbourg court is also accommodating the logic of the freedoms of circulation to its human rights lens. This is particularly conspicuous in the 2010 A, B and C v. Ireland and the 2011 S and H v. Austria Grand Chamber rulings, in which the Court relies on the notion that applicants ‘have the right to lawfully travel abroad’48 to seek the kind of care that the petitioned national law barred them from accessing in their countries of residence (i.e., abortion in 43  Although the circulation of gametes itself is on the rise (it is reported that Cryos [the largest European sperm bank based in Denmark] currently delivers 50% of their order directly to customer’s homes (‘home insemination’) rather than to fertility clinics: Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU, op. cit., p. 165). 44  The ECtHR has ruled that Italy disproportionality infringed upon the right to private and family life when refusing to recognize a same-­sex couple’s marriage contracted abroad because Italian law anterior to 2016 did not provide any legal protection or recognition of their union: ECHR, 14 Dec. 2017, Orlandi and others v Italy, no 26431/12; 26742/12; 44057/12 and 60088/12). 45  See, e.g., the Diane Blood case, in which a court in the United Kingdom relied on EU free movement principles to order a British clinic to export her deceased husband’s sperm to Belgium in order for her to benefit from postmortem insemination in contradiction to UK law: Ruling by the English Court of Appeal: Human Fertilisation and Embryology Authority, ex parte Blood [1997] 2 All ER 687. A recent case in France led to similar results, absent explicit references to EU law: Conseil d’Etat, Assemblée, 31 May 2016, Gomez, no 396848. The issue is disputed though, with lower courts denying such requests especially when claimants have no links to the foreign country in which they claim the gametes be exported (see TA Toulouse, 13 Oct. 2016; TA Nancy, 14 Dec. 2018, no 1803326). 46  Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU, op. cit., p. 170–171. 47  I. Glenn Cohen, ‘Circumvention Tourism’ (2012) 97 Cornell L Rev 1309. 48  ECHR, GC, 10 Dec 2010, A, B and C v Ireland, 25579/05, §XX: ‘accordingly, having regard to the right to lawfully travel abroad for an abortion with access to appropriate information and medical care in Ireland, the Court does not consider that the prohibition in Ireland of abortion for health and well-­being reasons, based as it is on the profound moral views of the Irish people as to the nature of life and as to the consequent protection to be accorded to the right to life of the unborn, exceeds the margin of appreciation accorded in that respect to the Irish State.’

Of Markets and Principles: The European Perspective   911 Ireland and gamete donation in Austria,49 respectively). Besides the restrictive rules on access to ART for single or homosexual intended parents, other rules that have been historically motivated by the preservation of gendered understandings of the sexual family,50 such as that of donor anonymity, are also under pressure. To a certain extent then, it appears that the logics of the market (cross-­border care, circulation of patients and service, full applicability of rules and categories such as free movement principles, exportation and importation of gametes, etc.) are indeed contributing to the calling into question of a number of gender norms. It would, however, be hasty to jump to the conclusion that the market is liberating and/or emancipatory in that sense. Empirical inquiry does indeed call for greater caution. First, several important works (especially in the United States) have established that the markets are far from operating free of gender norms.51 In fact, both the justifications and the amounts for payment of sperm and eggs, for instance, reveal a form of discomfort that predominantly affects female reproductive material’s marketization. This is one of the reasons that, despite the absence of any strong legal prohibition on the marketization of gametes, the situation remains one of incomplete commodification.52 Second, in Europe, legal rules are currently being driven by two tendencies that raise a number of questions in terms of their inherent gender dynamics, albeit ones that differ from the North American situation. Overall, the trend in Europe is to relax rules pertaining to gamete procurement and circulation. Because this trend allows for greater autonomy in reproductive choices, it can also be viewed as one that dismantles former gendered understandings of reproductive rights. Spain, Belgium, or the Czech Republic are affirming themselves as hubs of egg supply and monetary exchange. Routes for reproductive resources are well and truly appearing. The ‘good international reputation’ of countries such as Spain for egg donation, for instance, has been presented as one explanation for the decline in demand experienced in Belgium between 2003 and 200753; whereas Spain and the Czech

49  ECHR, GC, 3 Nov 2011, SH v Austria, 57813/11, §XX: ‘the fact that the Austrian legislature, when enacting the law on artificial procreation which enshrined the decision not to allow the donation of sperm or ova for in vitro fertilization, did not at the same time prohibit sperm donation for in vivo fertilization . . . is a matter that is of significance in the balancing of the respective interests and cannot be considered solely in the context of the efficient policing of the prohibitions. It shows rather the careful and cautious approach adopted by the Austrian legislature in seeking to reconcile social realities with its approach in principle in this field. In this connection the Court also observes that there is no prohibition under Austrian law on going abroad to seek treatment of infertility that uses artificial procreation techniques not allowed in Austria and that in the event of a successful treatment the Civil Code contains clear rules on paternity and maternity that respect the wishes of the parents.’ 50 M. Fineman, The Neutered Mother, The Sexual Family and Other Twentieth Century Tragedies (Routledge 1995). 51  See the fascinating inquiry of R. Almeling, Sex Cells: The Medical Market for Eggs and Sperm (University of California Press 2011). 52  Ram, this volume. 53  G. Pennings et al., ‘Cross-­Border Reproductive Care in Belgium,’ p. 3117.

912   Stéphanie Hennette Vauchez Republic seem to be particularly attractive to German people seeking egg donations.54 In fact, egg donation is becoming the primary reason that couples travel for fertility treatment.55 Not to mention the important development of ‘web shops’ and unofficial markets in which private actors (individuals and couples seeking gametes as well as more or less institutionalized suppliers and brokers) are taking hold.56 As far as sperm goes, Denmark is allegedly turning into the ‘sperm capital of the world.’ Not only is the country home to the world’s largest sperm bank, but it also exports massively to more than 70 countries, thus leading to what some call ‘a new Viking invasion.’57 The Cryos sperm bank offers individuals and couples the ability to select their donor on the basis of a number of criteria, ranging from height and weight to eye color, blood type, and education. Belgium is a close contender to Denmark on the sperm market and proves especially attractive to French women, who represent 80% of the foreign patients who come to access sperm.58 In other words, gametes represent a large share of the human cells that are exported and imported in Europe, and this creeping in of the market is providing European men and women with ‘new ways of creating families.’59 To be sure, some countries remain conservative in their normative choices.60 In France until 201161 (and although sperm donation has been routinely practiced and regulated since the end of the 1970s), egg donation was only very restrictively allowed: women could only become donors if and when they already had given birth to at least one child. At any rate, egg and sperm donation can only take place in nonprofit structures that benefit from administrative authorizations: in fact, public hospitals. The constraints on egg donation were relaxed in 2011, under the pressure of increasing demand and the development of ethically questionable practices62; and young women can now donate and even potentially store part of their ‘donation’ for potential future autologous use. Their donation, however, remains for free: although an important official report of

54  R. Storrow, ‘The Proportionality Problem in Cross-­Border Reproductive Care’ in I. Glenn Cohen (ed.) The Globalization of Health Care. Ethical and Legal Challenges (Oxford University Press 2013), p. 125. 55  G. Pennings, ‘Import and Export of Gametes,’ op. cit. 56  For a breathtaking account on undercover sperm markets in France: C. Dumont, Supergéniteurs, Enquête sur le don de sperme sauvage en France (Michalon 2016). 57  https://www.theguardian.com/society/2012/nov/02/worlds-­biggest-­sperm-­bank-­denmark 58  G. Pennings and al., ‘Cross-­Border reproductive care in Belgium,’ p. 3116. 59  M. Goodwin ed., Baby Markets: Money and the New Ways of Creating Families (Cambridge University Press 2010). 60  For instance, in Italy, both sperm and egg donation remain prohibited; in Austria, ARTs are only applicable in autologous settings. 61  Law of 6 August 2011. As a counterpart, and subject to the condition that their donation is abundant enough, women are offered the option of autologous conservation of some of their eggs for possible future need; see Article L. 1244–2 Code of Public Health. 62  Women who needed to benefit from egg donation progressively developed the following strategy: apply to a fertility clinic as a potential recipient and back up the application by bringing in a new donor to the facility.

Of Markets and Principles: The European Perspective   913 2011 recommended a policy shift towards a more generous understanding of the costs incurred by egg donors,63 they currently cannot be remunerated. Moreover, even in countries where gametes are increasingly available, the various limitations and containing rules that make up the ‘European way of market’ have not disappeared. In fact, the raw comparison between prices for gametes in Europe and in the United States proves striking: whereas it is routinely documented that sperm samples go for an average of US$100 per sample and eggs range from US$4,000 to US$6,000 overseas, prices remain much lower in Europe. The legal categories of ‘compensation’ and ‘remuneration’ (for donors) suggest that compensation should involve a lower amount: remuneration, after all, is more than simply covering expenses and in­con­ven­ ience. But realities on the ground suggest that it is not so simple. Take egg donation: a donor in the UK would receive £750 (around €870) classified as ‘compensation.’ In Spain a donor would receive around the same amount (€800) but as ‘remuneration’ for the same service/donation. And in Belgium, the egg donor would receive up to €2,000 Euros. Spain and the UK, each in their own ways, are the two countries where the remuneration/compensation of egg donors is most documented and conspicuous. In Spain, it costs €900 on average to access donated eggs64; in the UK, egg donors are not remunerated and the Human Fertilization and Embryology Authority indicates that egg donor’s expenses are estimated at £750.65 In Belgium, the sum may rise to €2,000.66 Sperm donation, when it is remunerated, remains subjected to mostly nominal amounts: in the UK, each sample is compensated around £35,67 and, even at Cryos, the leading Danish sperm bank, the amount remains capped at around £65.68 In many countries, even when legal rules express the notion that there can be no remuneration, the amounts at stake sometimes constitute a form of de facto circumvention of the rule. In other words, the notion that forms of ‘payment’ for reproductive cells remain contentious still has a strong hold in Europe. In fact, the various moves towards commodification that are particularly conspicuous as far as gametes are concerned has essentially operated within the margin allowed by a profusion of euphemisms and semantic tricks, whereby words like ‘compensation’ or ‘incentives’ have been doing a lot of work. Nowhere are sperm and eggs openly marketized or sold or associated with prices. The costs associated with the procurement of eggs and sperm are always presented as mere compensations and indemnities, and legal rules often heavily insist on the fact that it is the donors’ time and effort rather than the actual product that are being compensated. On the demand side, however, people who are eager to access reproductive material readily speak in terms of price and would seem ready to pay higher prices, too. But legal 63 IGAS, Etat des lieux et perspectives du don d’ovocytes en France, op. cit. 64 IGAS, Etat des lieux et perspectives du don d’ovocytes en France (Février 2011). 65  https://www.hfea.gov.uk/treatments/explore-­all-­treatments/ using-­donated-­eggs-­sperm-­or-­embryos-­in-­treatment/ 66  G.Pennings, ‘Import and Export of Gametes,’ op. cit. 67  https://www.hfea.gov.uk/treatments/explore-­all-­treatments/ using-­donated-­eggs-­sperm-­or-­embryos-­in-­treatment/ 68  Economic Landscapes of Human Tissues and Cells for Clinical Application in the EU, op. cit., p. 174.

914   Stéphanie Hennette Vauchez regulators hold on to a model of noncommodification. The Council of Europe, for instance, has tried to flag the importance of what was going on as far as payment for gametes is concerned: ‘incentives are particularly worrisome in the case of gamete donors (especially oocyte donors), where they may change the donor’s perception of the relative risks and benefits of a donation that is not free of potential health hazards and psychological consequences.’69 So, overall, there is still a lot of resistance at the European level against the full marketization and remuneration of gametes. As a result, it does not take much for strong concern to be voiced if and when it is felt that gametes are being exchanged, acquired, and let go in venues that do not appear sufficiently professional and sanitized. The 2004–2005 Romanian eggs scandal is a case in point: after media reports raised concerns over the price that Romanian donors were being paid, the European Parliament voted a resolution condemning ‘payment other than compensation for cell and tissue donations,’ practices deemed to be ‘not acceptable in Europe.’70 A thorough report by the European Commission ensued that sought to take stock of national practices.71 Europe, too, thus appears to be a land of incomplete commodification. Substances of human origin (and gametes in particular) are circulating and being exchanged, and their circulation has been supported if not encouraged by the increased hold of the legal categories and rules associated with the European internal market. The contentious nature of the idea of associating prices with human material does, however, remain strong; commodification, in that respect, remains limited by the low(er) monetary compensation. Moreover, this overall relaxing of the rules pertaining to the procurement and circulation of gametes is being questioned in terms of its gender impact in ways that are strikingly different from those in the United States. As Natalie Ram’s contribution aptly evidences,72 the US market in eggs and sperm operates with stereotypical representations of both gametes and donor profiles: while sperm donation is often constructed as a job (or second job), egg donation tends to be romanticized and spoken of in the language of a gift or altruism. Furthermore, whereas sperm donors are widely understood (and in fact desired) to be motivated primarily by monetary purposes, women are expected not to be primarily driven by financial motives. In other words, if the market allows for sperm to be ‘meat,’ eggs should clearly remain viewed as ‘pets.’73

69  Council of Europe EDQM, Guide to the Quality and Safety of Tissues and Cells for Human Application, op. cit., p. 39; the report continues to insist on the particular challenges that exist in that respect with regard to gametes: ‘In addition, gamete donation for treatment purposes presents further ethical implications because it involves the potential generation of a new human being.’ 70  Res. P6_TA(2005)0074. 71  European Commission, Report on the Regulation of Reproductive Cell Donation in the European Union, 2006. 72  Ram, this volume. 73  M. A. Case, ‘Pets or Meat’ (2005) 80 Chicago-­Kent L Re 1129.

Of Markets and Principles: The European Perspective   915 Conversely, in Europe, price caps and limitations seem to be playing out on both sperm and eggs. This does not mean, however, that there are no questions framed in terms gender equality that are triggered by these contemporary evolutions in the field of gamete procurement. The greater question seems to be that of the protection of women who are unduly represented as ‘gamete donors’ in the same ways that men are. A number of authors express anxieties related to the invisibilization of the greater threats to women’s health and greater invasiveness of egg retrieval procedures than the ones posed by sperm donation. These authors tend to frame them as capitalist forms of pressure that marketization exerts on women’s reproductive material. In other words, the market appears to be both a source of liberation from gender norms that have traditionally informed the legal regulation of ART (access of same-­sex couples, increased reproductive autonomy) and a cause for concern because of the pressure resulting from strong demand for female reproductive material. In their fascinating Tissue Economies book, Catherine Waldby and Robert Mitchell insist on the importance of a pragmatic approach to what is actually happening as far as circulation of blood, tissue, and cells and the extent to which the overall framing of regulatory paradigms can be deceiving: ‘gift systems and commodity systems for managing human tissues are often cast . . . as mutually exclusive and morally incompatible social forms.’74 Their book aims at complicating and disorganizing this dichotomy. This chapter contends that observations in the evolutions of European governance of blood, tissue, and cell circulation, in comparison with those that feature North American developments on the subject, are concrete illustrations of both the intertwined nature of gift and commodification and the ambivalent relationship of the market to gender equality.

Further Reading O Cayla, S Hennette Vauchez, ‘Bioéthique et Constitution,’ in M. Troper ed., Traité international de droit constitutionnel (Dalloz 2012) t. 2. A Dubuis, ‘L’appréhension juridique de la notion de patient,’ in Anne Laude, Didier Tabuteau dir., Les droits du patient européen (Société de législation comparée 2009) 17. S Halliday, ‘A comparative approach to the regulation of human embryonic stem cell research in Europe’ (2004) 12 Med L Rev 40. S Hennette Vauchez, ‘Bioéthique et genre: cadrage théorique, enjeux européens,’ in Anne Françoise Zattara-Gross ed., Bioéthique et Genre (LGDJ 2013) 21. S Hennette Vauchez, ‘The Society for the Protection of Unborn Children v. Grogan (ECJ, 1991): Rereading the case and retelling the story of reproductive rights in Europe,’ in F.  Nicola, B.  Davies eds., EU Law Stories. Contextual and Critical Histories of European Jurisprudence (Cambridge University Press 2017) 393. M-A Hermitte, Le sang et le droit. Essai sur la transfusion sanguine (Seuil 1996). 74  C. Waldby, R. Mitchell, Tissue Economies: Blood, Organs and Cell Lines in Late Capitalism, op. cit., p. 9.

916   Stéphanie Hennette Vauchez T Hervey, J McHale, European Union Health Law, 2nd ed. (Cambridge University Press 2015). S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton University Press 2005). L Marguet, Le droit de la procréation en France et en Allemagne. Etude sur la normalisation de la vie. PhD dissertation (Université Paris Nanterre 2018). R Titmuss, The Gift Relationship: From Human Blood to Social Policy (Allen & Unwin 1970).

ETHICAL AND LEGAL I M PL IC AT IONS O F A D VA N C E S I N GENETICS

chapter 43

I n troduction to Gen etics a n d the L aw Maxwell Mehlman, Mette Hartlev, and Sonia Suter

1 Introduction Advances in genetics and genomics have created both hopes and concerns and raise a number of legal issues in both the United States and Europe. For example, how do we protect people with genetic risks of disease from unfair discrimination? How should home genetic testing kits be regulated? And how should we regulate clinical genetic testing and gene editing, including germline modifications? In some areas, the laws of both jurisdictions are similar, but, as this chapter will show, there are some notable differences as well. To understand the legal issues in genetics, it is helpful to have some understanding of the different types of genetic analysis that can be used and for which purposes. Genetic analysis has evolved considerably over the decades as scientists have refined the ability to identify variants in the genes or genome. Certain variants in genes or abnormal numbers of chromosomes can lead to disease. Genetic analysis therefore involves efforts to assess genes, chromosomes, or the genome to identify variants that might have clinical implications. Genetic analysis can be used for different purposes and in different contexts. It can confirm or establish the diagnosis of a genetic condition in adults and children. Initially, genetic testing was limited to identifying a few single-­gene disorders, like sickle cell anemia, cystic fibrosis, or Duchenne muscular dystrophy, or chromosomal numeric anomalies, like Down syndrome (trisomy 21). Increasingly, genetic testing can identify susceptibility to future illnesses, like cancer or neurological disorders, some of which develop later in life. Today it is possible to test for more than 2,000 genetic conditions.

920   Maxwell Mehlman, Mette Hartlev, and Sonia Suter One form of genetic testing, carrier testing, can determine whether unaffected individuals carry a recessive gene for conditions like sickle cell anemia, Tay Sachs, or cystic fibrosis that could affect the health of their children. Genetic testing also can identify variants that influence medical treatment or predict responses to pharmacological therapy, such as whether someone has warfarin sensitivity, a low tolerance for the anticoagulant warfarin, or clopidogrel resistance, where the antiplatelet drug clopidogrel is less effective. During pregnancy, the fetus can be tested through amniocentesis or chorionic villus sampling, and, after in vitro fertilization, the embryo can be evaluated through preimplantation genetic testing. Finally, in most industrialized countries, newborns undergo testing for certain genetic disorders—such as phenylketonuria, which if not treated properly can lead to mental impairment—to offer early intervention and treatment for affected infants. The nature of genetic testing also varies greatly. Some tests assess the number of chromosomes to identify chromosomal numeric anomalies such as trisomy 21. Others examine DNA in a gene to look for mutations. In some instances, the testing looks for several mutations because the disease can be caused by multiple different mutations—in the case of cystic fibrosis, more than a thousand different mutations can cause the condition. Finally, some genetic tests identify genetic defects by analyzing the products of the gene, such as the RNA that codes for a protein or the protein itself, as with some forms of carrier testing. An important development in genetic analysis is next-­generation sequencing, simultaneous or “high throughput” sequencing of the whole genome or exome (sequences within the genome that code for proteins), which is becoming ever cheaper and faster. Today, next-­generation sequencing is mostly used to search for mutations when someone has a medical condition without a clear diagnosis, but it may become much more widely used in the future. In virtually all these contexts, the doctrine of informed consent is important. Indeed, the goal of helping patients make adequately informed medical decisions arguably plays a central role in genetics more than in any other medical specialty. Genetic’s professionals have long espoused the importance of trying to help patients make informed decisions consistent with their values and life preferences. Often, the professionals strive to be as nondirective as possible so as not to bias the patient’s decisions about whether to undergo genetic analysis and for what purpose, although there is continued debate about the feasibility and value of non-­directiveness. While the law in the United States and Europe does not generally require non-­ directiveness, the legal doctrine of informed consent applies in all jurisdictions, imposing on physicians and genetic counselors the legal obligation to disclose material information that could influence a patient’s decisions about genetic testing. Central to this doctrine is the notion that one has the right to learn or not learn genetic information about oneself when presented with the option of genetic testing. As we shall see in the upcoming chapters, however, as genetic analysis becomes more widespread and complex, and as more people undergo genetic testing or genome sequencing, informed consent may become increasingly difficult to achieve.

Introduction to Genetics and the Law   921 The other ethical and legal issues that arise with genetic or genomic analysis differ depending on the purpose, context, and technology used. Genetic analysis of adults implicates different concerns from testing children or reproductive testing. Similarly, genetic screening raises different issues from diagnostic genetic testing, and targeted genetic analysis differs from genome sequencing. Because of these important differences, we explore the ethical and legal issues associated with genetic/genomic analysis for adults, children, and reproductive purposes separately in our chapter. In the context of clinical genetic testing generally, one issue is whether health professionals have a legal duty to inform a patient’s family members about their genetic risks without the patient’s consent. With respect to children, questions arise about the scope of genetic testing that parents can request for their children, especially given professionals’ recommendations to delay predictive testing for late-­onset conditions. Issues also arise with newborn screening, including whether parental consent should be required, whether it should include whole-­genome sequencing, and whether and under what conditions the government can conduct research on the newborn screening samples. In the context of prenatal testing, legal issues arise concerning the potential liability for healthcare professionals who fail to offer such testing when medically indicated and whether parents should have the right to terminate a pregnancy based on the results of genetic testing or screening. A related issue is the acceptability, ethically and legally, of preimplantation testing of embryos to prevent disease or to select for certain traits. More general legal issues in genetics concern the methods of regulating genetic tests offered in the clinical setting and through the “direct-­to-­consumer” online genetic testing industry. Another important issue is genetic discrimination and the nature of legal protections that do or do not exist with respect to such discrimination in the context of employment and various types of insurance. A related matter is the scope of privacy protections of genetic information and the potential use of genetic information for research, sale, or other purposes by various entities, including direct-­to-­consumer companies and the military. Finally, the regulation of human gene therapy, gene therapy research, and germline modifications raise additional issues, including which actions the government will fund, permit, or prohibit. Regulatory questions also arise with respect to drug product labeling for pharmacogenetic drug–gene interactions and whether genetic testing should occur before prescribing. Although genetic testing is relevant in the forensic context, we do not explore issues associated with such testing. As our chapters will show, there are a number of similarities in the legal approaches in the United States and Europe to these issues, but there are also some areas where the legal issues arising from genomics are treated quite differently.

Chapter 44

Gen etics a n d the L aw United States Maxwell Mehlman and Sonia Suter

1 Introduction In the United States, both the federal government and the states play a role with respect to the legal and ethical issues that arise in the context of genomics. At the state level, legislatures have enacted laws in various areas, including newborn screening and nondiscrimination and privacy protections. In addition, state courts have addressed some issues concerning genetics, such as the duty to warn. At the federal level, the US Congress has enacted a specific statute, the Genetic Information and Nondiscrimination Act, which protects genetic information. Other federal statutes, which do not address genetics or genomics in particular, also have relevance in the genetics context, including laws that protect against certain forms of discrimination or that regulate laboratories. Federal agencies also play a role: for example, protecting genetic privacy or regulating genetic tests. Finally, the US Constitution is relevant to genomics, especially concerning reproductive rights, which are pertinent to reproductive genetic testing. The following sections address state and federal laws in the United States that govern legal and ethical issues concerning genetic and genomic analysis for diagnostic purposes, regulation of genetic testing, genetic discrimination and privacy, and clinical applications of genomics.

924   Maxwell Mehlman and Sonia Suter

2  Genetic and Genomic Analysis for Diagnostic Purposes 2.1  Adult Genetic Testing Genetic analysis of adults includes diagnostic, predictive, and carrier testing as well as testing to tailor medical treatment. Research and technological developments are rapidly broadening the scope of genetic analysis as well as the populations who will undergo such testing. Until recently, most adult genetic testing was limited to people seeking a diagnosis for an existing condition or who had known risk factors that would warrant genetic testing. For example, carrier testing was limited to individuals with a family history of a heritable condition, such as cystic fibrosis, or who belonged to ethnic groups in which certain recessive diseases are more prevalent, such as people of Ashkenazi Jewish descent who face higher risks of carrying mutations associated with conditions like Tay Sachs or Canavan’s disease. Carrier testing, however, is beginning to expand as labs offer testing for broad panels of conditions regardless of risk factors, sometimes identifying carriers outside the expected populations. Although this presents the risk of false positives, for the most part, this development is positive because it allows the identification of carriers who would otherwise be missed by focusing only on ethnicity. Similarly, testing asymptomatic people for susceptibility to late-­onset conditions like heritable breast and ovarian cancer is no longer limited to those at increased risk based on family history, particularly with the emergence of direct-­to-­consumer (DTC) genetic analysis. For example, 23andMe has recently begun to offer tests for susceptibility to adult-­onset diseases such as Alzheimer’s disease, Parkinson’s disease, and some forms of heritable breast cancer. This change raises questions about the impact of such testing outside the clinical context and the accuracy of risk estimates given that our understanding of these risks is largely based on research of individuals with family histories of these conditions. As we move toward personalized medicine, it is likely that the number of people undergoing genetic testing will increase because the goal of personalized medicine is to use genomic information to make individualized recommendations for preventive medicine—such as modifying lifestyle and environmental factors—and to tailor therapeutic treatment. Inevitably, this will lead to genetic testing of many more people, including those without known risk factors. As next-­generation sequencing becomes cheaper and more efficient, it may become a routine part of medicine, which would also broaden the scope of genetic analysis for most people. While many herald these potential developments, critics worry about such expansive testing for several reasons. First, a tremendous gap still exists between obtaining genetic information and being able to use that information to prevent many diseases. As a result, genetic testing may identify people at risk for conditions that are not currently prevent-

Genetics and the Law: United States   925 able, which can create anxiety and expose people to stigmatization and discrimination by employers, insurers, and other entities. Second, the shift from targeted genetic testing, which examines a single gene associated with a particular disorder, to more comprehensive genomic testing dramatically increases the amount and type of information that can be obtained. Genomic analysis can identify a wide range of variants that differ in several respects: their clinical relevance, the variability of their effect on health and age of onset (childhood or adulthood), whether prophylactic or preventive measures can reduce morbidity or mortality, and their reproductive implications. In addition, because our understanding of the genome is still incomplete, genomic analysis may identify variants of uncertain significance (i.e., variants that may or may not have clinical implications), potentially creating anxiety for patients and raising dilemmas about clinical management of the patient. In addition, the tremendous amount of information that genome sequencing can generate also potentially creates information overload for both providers and patients. Some have estimated that it would take between two and six hours to explain the various categories of information one could obtain from genomic analysis. We already have too few genetic counselors to interpret and explain the implications of genetic analysis today. As more individuals undergo genome sequencing, this problem will grow, raising serious questions about the ability of individuals to make informed decisions about genome sequencing and the capacity of healthcare professionals to understand its clinical implications. Finally, genome sequencing intended to explore a particular risk might reveal unexpected or incidental information about other risks for which the person being tested was not prepared or perhaps did not want to learn. Professional recommendations have suggested that laboratories and physicians have an ethical obligation to analyze and disclose certain variants that are clinically actionable. Because individuals have an autonomy interest in deciding whether to learn about genetic risks, however, the recommendations suggest that patients should be able to opt out of having such variants analyzed. The law has not yet addressed whether there are parallel legal obligations, although it is likely a court would find a legal duty to disclose clinically actionable variants. Given a physician’s fiduciary obligation, such a legal duty would be appropriate if it was limited to information that physicians could reasonably be expected to know. In many instances, physicians, particularly those without expertise in genetics, might not have the knowledge to identify relevant variants. While laboratories are better suited to identify clinically relevant variants than are physicians, the challenge in imposing legal obligations, even upon laboratories, is the uncertainty or sometimes even disagreement about the significance of many variants. Any legal obligations would therefore have to be limited to variants about which there was clear consensus regarding their clinical significance. Whether honoring a patient’s autonomy interests when she declines analysis of ­certain variants would provide a defense if a preventable condition associated with the variant ultimately developed is not so clear. After all, the harm of not identifying and

926   Maxwell Mehlman and Sonia Suter disclosing a preventable risk would likely be more significant and tangible to the patient than the harm of informing the patient of information she had opted against learning. On the other hand, the genetic counseling world has long placed great stock in individual choice with respect to obtaining genetic information. As a result, such a defense seems fitting if a patient fully understands the clinical implications of failing to receive such information. Questions of liability concerning genetic testing have arisen in a few contexts. For example, a few courts have explored whether providers have a duty to warn the family members of a patient found to have a genetic predisposition to a disease that can be prevented or treated. Because genetic risks in a patient have implications for related family members, three cases have found that healthcare providers owe relatives who are “within the zone of foreseeable risk”1 or who are “members of the immediate family”2 a duty of care when they discover a clinically actionable genetic variant in the patient. What this duty of care requires is a matter of some debate. One court found that an obligation to warn the patient’s relative directly would be “too heavy a burden on the physician.”3 Another suggested the scope of the duty would depend on the facts of each case, but it noted that merely informing the patient may not always fulfill the duty of care. Professional guidelines, in contrast, either prohibit disclosure to relatives or only support it in exceptional cases: when attempts to encourage the patient to disclose fail; the relative is identifiable; and the risks are highly likely, serious, and treatable. In some states, genetic privacy laws would prohibit such disclosures without consent. Because two of these cases were decided before the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule became effective in 2003, the courts did not have to contend with the possible tension between an obligation to preserve confidentiality under HIPAA and a duty to warn relatives under case law. The HIPAA Privacy Rule has some “public purpose” exceptions for disclosure of health information without consent. In addition, the Office for Civil Rights of the Department of Health and Human Services has interpreted the Privacy Rule as allowing the sharing of genetic risks with family members provided the patient “has not agreed to a restriction on such disclosure.”4 No court has yet addressed the validity of this interpretation or whether such disclosures would fall within the HIPAA exceptions, although some scholars are skeptical. When efforts to persuade a patient to inform a relative about a serious and preventable risk fail, it is problematic to bar healthcare providers from ever disclosing such information, under any circumstances. Instead, an ethical argument can be made for an interpretation of HIPAA that permits a provider in rare instances—when serious harms could not otherwise be averted and when disclosure would not cause serious 1  Pate v. Threlkel, 661 So. 2d 278, 282 (Fla. 1995); see also Molloy v. Meier, 679 N.W.2d 711, 717–719 (Minn. 2004) (relatives who would be foreseeably harmed by negligent diagnoses). 2  Safer v. Pack, 667 A.2d 1188, 1192 (N.J. Super. Ct. App. Div. 1996). 3  Pate, 661 So. 2d, at 282. This holding was consistent with Fla. Stat. Ann. § 760.40, which prohibits disclosure of an individual’s genetic information without the individual’s consent. 4  US Department of Health and Human Services Office for Civil Rights. FAQ #512. htttp:/www.hhs. gov/ocr/privacy/hipaa/faqright_to_request_a_restriction/512.html

Genetics and the Law: United States   927 harm to the patient—to inform a patient’s relatives or their physician about treatable genetic risks. Such an interpretation would not impose a duty of disclosure but would allow professional discretion to prevent serious harm in exceptional and exigent circumstances. These legal issues may become even more complex with increasing access to genome sequencing and testing outside the clinical context through DTC companies. Furthermore, because our understanding of and capacity to address genetic risks is continually evolving, additional issues concerning the duty to recontact patients may also emerge. A related issue arose in the United States in a case where a mother sued a laboratory for negligent misclassification of a variant identified through genome sequencing of her son who suffered from a seizure disorder. She claimed that, because the laboratory originally classified the variant as being of unknown significance, her son was treated in a manner that ultimately resulted in his death. Had the lab identified the variant as pathogenic, she argued, he could have been properly diagnosed and treated, thus preventing his death. In preliminary rulings, the state supreme court found that a laboratory can be sued under medical malpractice because it was “performing diagnostic testing at the request of a treating physician for the purpose of diagnosis and treatment, which is a core function of hospitals in diagnosing and treating patients,”5 and a federal court denied in part a motion to dismiss her claim for negligent misrepresentation and constructive fraud.6 This holding blurs the line between laboratory testing and the practice of medicine, raising the potential liability of laboratories in other jurisdictions. It also highlights the challenge laboratories face in standardizing the classification of genomic variants, a daunting task when labs interpret variants differently and our understanding of the genome is continually evolving. Finally, it raises issues about how laboratories should report results to physicians and what their obligations are to reclassify variants as new findings are made, especially when thousands of genes are analyzed through genome or exome sequencing.

2.2  Genetic Analysis of Children While genetic or genomic testing in children for diagnostic purposes or to tailor medical treatment is not controversial, the professional consensus is that predictive genetic testing of children for conditions that do not manifest in childhood is problematic 5  Williams v. Quest Diagnostics, Inc., 816 S.E.2d 564, 565 (S.C. 2018) (finding that this met the statutory definition of a licensed healthcare provider and therefore allowed the laboratory to be sued for medical malpractice). 6  Williams v. Quest Diagnostics, Inc., 353 F. Supp. 3d 432 (D. S.C. 2018) (finding that the mother sufficiently alleged justifiable reliance on misclassification of her son’s genetic mutation in her claim for negligent misrepresentation and constructive fraud, sufficiently stated a claim for violation of South Carolina Unfair Trade Practices Act, but failed to state a claim for civil conspiracy).

928   Maxwell Mehlman and Sonia Suter because it offers no short- or long-­term advantages and presents several risks. Some worry that it could limit the child’s future autonomy by eliminating her ability to determine whether and when she wants to learn such information. Although the limited research in this area has not identified “significant psychosocial harms” in late-­onset genetic testing of children,7 many worry that identifying susceptibility to adult conditions in children may cause parents to treat the child differently, become overprotective or unduly anxious, or create narratives that assume the child will or will not develop the condition. Identifying susceptibility to late-­onset conditions could also make the child vulnerable to stigmatization or discrimination by third parties such as insurers, schools, or other entities. Thus, although the law recognizes the right of parents to make medical decisions for their children, professional groups recommend delaying predictive or predispositional testing until the child is at least an older adolescent or unless the test is for a condition, like familial adenomatous polyposis, that presents in childhood. One form of genetic analysis in children that is widely accepted and performed in virtually every newborn is newborn screening. Such testing involves the collection of a small blood sample that is screened for certain medical conditions, many of which are inherited. Because newborn screening is a screening test, abnormal results merely indicate a heightened risk and require diagnostic testing to determine whether the child is truly affected. Newborn screening has evolved considerably since its inception more than half a century ago as a program to screen for phenylketonuria, a recessive condition that results in intellectual impairment unless the child eats a diet free of the amino acid, phen­yl­al­a­nine. For the first several decades, most states screened for no more than eight diseases, but by the turn of the twenty-­first century, technology made it possible to screen for more than forty conditions. After a period of great variability in newborn screening programs across the country, as of 2014, every state screens for a minimum of thirty conditions; some more than fifty. The most recent Recommended Uniform Screening Panel created by the Secretary of Health and Human Services listed thirty-­ five core disorders that states should include in their panels, including primary congenital hypothyroidism, cystic fibrosis, maple syrup urine disease, homocystinuria, sickle cell anemia, and other hemoglobinopathies. From the beginning, newborn screening was viewed as a public health program whose raison d’etre was to benefit newborns by identifying conditions that can be prevented or ameliorated with early intervention. To further this public health goal, almost all states essentially mandate newborn screening in some form or another. Only two states, Maryland and Wyoming, require affirmative parental consent; the rest do not, mandating newborn screening whether or not parents affirmatively consent. Nearly all states presume consent and allow parents to opt out—in many states only for religious reasons, and in others also for philosophical or personal beliefs. A few states, however, provide no option to opt out at all. Newborn screening is therefore one of the few areas 7  Jeffrey R. Botkin et al., Points to Consider: Ethical, Legal and Psychosocial Implications of Genetic Testing in Children and Adolescents, 57 Am. J. Hum. Genetics 6 (2015).

Genetics and the Law: United States   929 in medicine where medical testing can be performed without affirmative consent. The mandatory nature of newborn screening, however, has not been very controversial because it is justified by the doctrine of parens patriae, which allows the state to limit a person’s liberty to preserve human life, in this case to ensure that parents fulfill their duty to act in the best interests of their children. This rationale has been especially persuasive with respect to the original newborn screening panels, which included diseases whose early diagnosis could prevent or reduce morbidity and mortality. The expansion of newborn screening, however, cannot be defended on the same grounds because it includes some conditions for which, even with treatment, life expectancy is significantly shortened, such as Pompe disease, or where morbidity and mortality remain high, such as propionic acidemia, carnitine ­palmitoyl transferase deficiency, or long-­chain 3-­hydroxyacyl-­CoA dehydrogenase deficiency. While such screening does not directly benefit the newborn, it can provide parents with relevant information for their future reproductive decisions or help us learn more about the natural history of the disease. Such rationales construe newborn screening’s purpose much more broadly, raising questions as to whether they justify the mandatory nature of the program as persuasively as the original rationale. Although more comprehensive newborn screening panels offer many benefits, they are not cost-­free. Screening for more diseases increases the risk of false negatives, which may lead to false reassurance, and false positives or clinically ambiguous results, which may cause anxiety, confusion, and negatively affect the parent–child relationship. Even so, we will likely see newborn screening expand even further. As genome sequencing becomes more accurate, efficient, and cheap, people are beginning to discuss whether it should become part of newborn screening and personalized medicine.8 Indeed, pilot studies are under way to explore the impact of genome sequencing in the newborn context, which raises many of the same concerns regarding genome sequencing in adults. Society and professionals must therefore consider whether and what limits should be set on such screening. At this point, applying whole-­genome sequencing in newborn screening would be unwise. It would generate vast amounts of information about newborns, which itself would be overwhelming. Moreover, it would provide information about late-­onset conditions, like heritable cancers and neurological conditions; as noted earlier, genetics professionals discourage testing for such conditions in children. Finally, the uncertainties about the clinical implications of many variants and the quantity of information it would produce would exponentially increase the kinds of problems we already face with recent expansions of newborn screening, including false positives, anxiety, and diagnostic odysseys. 8  Although newborn screening indirectly identifies genetic variants across the genome by screening for a number of amino acid and organic acid disorders, it does not (yet) involve actual sequencing of the genome. Moreover, because it is aimed at only specific kinds of disorders, it does not identify (indirectly or otherwise) other kinds of variants across the genome unrelated to these kinds of disorders.

930   Maxwell Mehlman and Sonia Suter A legal issue that has emerged regarding newborn screening concerns the storage and secondary uses of the newborn screening samples, which can be used to repeat screening tests, to monitor the prevalence of various conditions, for quality assurance, and for postmortem diagnosis. In addition, the samples are useful for research, particularly because they constitute a valuable biobank representing the entire population. Very few states have regulations controlling the permissible uses of these samples or whether parents must consent to or be notified of such uses. Lawsuits,9 however, have prompted legislation in a few states to require parents to be informed about the collection and indefinite storage of newborn screening samples for potential research, with an option for parents or the adult child to request destruction of the samples. In addition, a handful of states require affirmative parental consent for storage and informed consent for future research uses of the newborn screening samples.

2.3  Reproduction and Genetic Analysis Genetic analysis for reproductive purposes can occur during pregnancy or after creation of embryos through in vitro fertilization (IVF). Amniocentesis and chorionic villus sampling, for example, can diagnose chromosomal abnormalities like trisomy 21 (Down syndrome) and single-­gene disorders in a pregnancy. Because these prenatal tests are invasive and present a potential increased risk of miscarriage, the standard of care is to offer them only when advanced maternal age, family history, carrier status, or other factors pose an increased risk of chromosomal or genetic anomalies. An alternative form of reproductive testing, preimplantation genetic testing, provides genetic information about embryos created through IVF to help parents select which embryos to implant. Preimplantation genetic testing includes two types of tests. The first, originally called preimplantation genetic diagnosis, analyzes the embryos for genetic diseases based on family history or carrier status. In rare cases, preimplantation genetic testing has been used to identify embryos that have tissue compatibility with existing children in need of stem cell donations because of serious diseases like Fanconi anemia. The second kind of test, which used to be called preimplantation genetic screening, analyzes embryos for chromosomal abnormalities in women of advanced maternal age or with a history of miscarriages. Because preimplantation genetic testing is costly, rarely covered by insurance, and involves the physically burdensome surgical process of egg retrieval, it is less desirable than amniocentesis or chorionic villus sampling for most people. For individuals with a high risk of passing on a genetic condition, or infertile couples who have already undergone IVF, however, preimplantation genetic testing offers an alternative to pregnancy termination to avoid giving birth to a child with a genetic condition. 9  See, e.g., Higgins v. Tex. Dep’t of Health Servs., 801 F. Supp. 2nd 541 (W.D. Tex. 2011); Bearder v. Minnesota, 806 N.W. 2d 766 (Minn. 2011); Kanuszewski et al. v. Mich. Dep’t of Health & Human Servs., 927 F.3d 396 (6th Cir. 2019).

Genetics and the Law: United States   931 While only some pregnant women undergo prenatal diagnosis or preimplantation genetic testing, virtually all are routinely offered prenatal screening to detect heightened risks for chromosomal anomalies. If the results are positive, diagnostic testing through chorionic villus sampling or amniocentesis is recommended. Prenatal screening has improved over the years in its ability to identify affected pregnancies and to avoid falsely flagging unaffected pregnancies. Initially, such screening analyzed alpha-­fetoprotein, a protein produced by the fetus, in maternal serum to identify heightened risks for trisomy 21 (Down syndrome), as well as spina bifida and abdominal wall defects. Maternal serum screening now analyzes certain hormones in addition to alpha-­fetoprotein. In addition, nuchal translucency screening, a specialized ultrasound that looks for fluid at the base of the fetal neck, can identify increased risks for trisomies 21, 13, and 18. Perhaps the most significant development in prenatal screening is the emergence of  cell-­free fetal DNA testing, also called noninvasive prenatal testing. This process ­analyzes cell-­free fetal DNA in maternal blood for certain chromosomal aneuploidies. Although not fully diagnostic, this test is much better able to identify affected pregnancies (sensitivity) and to accurately identify pregnancies that are not affected (specificity) than previous screening tests for chromosomal anomalies. Even with noninvasive prenatal testing, however, physicians recommend amniocentesis or chorionic villus sampling to confirm positive results because noninvasive prenatal testing has a small false-­positive rate. While clinical guidelines had previously recommended noninvasive prenatal testing only for high-­risk populations, the guidelines have become more permissive. As a result, although still primarily used for high-­risk pregnancies, noninvasive prenatal testing is increasingly being used by women with average-­risk pregnancies. By both reaching an ever-­larger population and providing highly sensitive and specific information without the risks and burdens of invasive prenatal testing, noninvasive prenatal testing is dramatically increasing parental access to information about chromosomal anomalies. For many reasons, access to prenatal or embryo genetic information is important. When the results of prenatal genetic testing are negative they can provide reassurance to parents. Conversely, when they identify fetal abnormalities, they allow parents to plan for the birth of a child with a disability or to consider alternatives such as placing the child for adoption or pregnancy termination. Preimplantation genetic testing similarly expands reproductive options by allowing parents to select embryos for implantation based on genetic information. Access to genetic analysis in the reproductive context, therefore, is directly linked to reproductive autonomy. Not only do physicians have an ethical obligation to offer reproductive testing when medically indicated, they also have a legal obligation. In most jurisdictions in the United States, physicians can be liable for wrongful birth claims if their failure to offer medically indicated prenatal tests prevents parents from obtaining information that would have led them to terminate an affected pregnancy.10 These claims do not assert that the 10  See, e.g., Garrison v. Med. Ctr. of Delaware, Inc., 581 A.2d 288 (Del. Sup. 1989) (finding actionable claim for failing to properly perform and report test results of amniocentesis in woman with increased risk of having a child with trisomy 21).

932   Maxwell Mehlman and Sonia Suter ­ rovider caused the disease. Instead, they are a variant of informed consent claims p because the assertion is that the failure to receive material medical information harmed the parents by preventing them from being able to make an informed reproductive decision. In addition, in only a handful of jurisdictions, an affected child can also bring a wrongful life suit claiming that the provider’s negligence resulted in damages associated with the suffering that comes from being born with the condition.11 While reproductive autonomy is central to reproductive testing, how best to preserve such autonomy is a matter of some debate. On one hand, many laud technological developments like noninvasive prenatal testing because they dramatically increase access to pertinent reproductive information. In addition, the growing capacity to identify a wider range of variants associated with disease and nonmedical traits will further increase reproductive choice, particularly if next-­generation sequencing becomes a part of reproductive genetic analysis. On the other hand, these advances raise several concerns. One worry is that technology and other factors, like legal pressures and cultural influences, routinize prenatal testing and push people to obtain prenatal information without carefully considering its personal value to them, thus undermining meaningful autonomy. A further risk is the possibility of information overload, particularly if reproductive testing moves from targeted testing to next-­generation genomic sequencing, which will generate vast amounts of fetal or embryo information and potentially overwhelm parents. An even greater challenge is deciding whether there should be any limits to reproductive decisions based on genetic information. Today, individuals usually seek reproductive testing to avoid passing on mutations associated with childhood illness, but sometimes they seek genetic information to make decisions based on adult-­onset conditions and nonmedical traits like sex. Some future parents even select for genetic conditions, such as heritable forms of deafness or dwarfism, through preimplantation genetic diagnosis, because they view such conditions not as disabilities, but as part of their cultural ­heritage. Their goal is to have children who resemble them and can easily integrate with their community. As reproductive genetic testing becomes more expansive, society will have to grapple with the propriety of reproductive decisions based on different kinds of genetic information and the question of who should determine the appropriate use of reproductive genetic information: parents, clinicians, or the government. This debate is currently ongoing in the United States. At one extreme, some argue that the constitutionally recognized right to abortion, first recognized in Roe v. Wade12 and reaffirmed two decades later in Planned Parenthood of Southeastern Pennsylvania v. Casey13 leads to a correlative liberty interest in determining, without state interference, 11  See, e.g., Turpin v. Sortini, 643 P.2d 954 (Cal. 1982) (recognizing cause of action for wrongful life, and only allowing recovery of extraordinary expenses associated with the hereditary affliction providers negligently failed to diagnose, but not general damages). But see, Becker v. Schwartz, 386 N.E.2d (N.Y. 1978) (finding no legally cognizable cause of action for an infant alleging a wrongful life claim for being born with trisomy 21 because “the law can assert no competence” as to whether “it is better never to have been born at all than to have been born with even gross deficiencies”). 12  410 U.S. 112 (1973). 13  505 U.S. 833 (1992).

Genetics and the Law: United States   933 what kind of prenatal testing or preimplantation genetic testing to undergo and what reproductive decisions to make based on genetic information. An intermediate view holds that professional autonomy should allow clinicians to refuse to participate in reproductive decisions they view as harmful. Finally, at the other extreme, legislation in a few states limits reproductive decisions based on genetic information by prohibiting abortions based on sex or genetic anomalies, like Down syndrome, sometimes justifying these laws as protections against discrimination based on these traits. While the Supreme Court has not evaluated the constitutionality of these reason-­based statutes, lower federal courts have been divided, with some enjoining and others upholding these laws.14 The Sixth Circuit recently overturned a preliminary injunction of an Ohio law. Pre-Term Cleveland v. McCloud, No. 18-3329 (6th Cir. filed Apr. 13, 2021) (en banc). Which view of reproductive autonomy in the context of reproductive genetics will ultimately prevail in the United States is uncertain given increasingly aggressive legislation in some states limiting or banning women’s access to abortion generally and the changing composition of the US Supreme Court. As the United States becomes more polarized with respect to reproductive rights, the most likely outcome is that some states will be very protective of reproductive autonomy and others will more significantly limit reproductive rights.

3  US Regulation of Genetic Testing Government regulation of genetic testing in the United States is divided into the regulation of test products by the Food and Drug Administration (FDA) as medical devices under the Federal Food, Drug, and Cosmetic Act, and the regulation of the laboratories that perform the tests by the Centers for Medicare and Medicaid Services in the Department of Health and Human Services (HHS) under the Clinical Laboratory Improvement Amendments of 1988.

3.1  Regulation of Test Products In regard to test products, the nature and degree of FDA regulation depends on whether the test product is a complete test kit sold commercially to a laboratory, called an in vitro diagnostic test kit; a genetic test that is developed and processed by the laboratory itself, called a “home brew” or laboratory-­developed test; or a testing service provided directly to consumers, called DTC testing. 14  After the Seventh Circuit found a law banning abortions based on the diagnosis of a disability, race, or sex unconstitutional (Planned Parenthood of Ind. & Ky. v. Comm’r, Ind. State Dep’t of Health, 888 F.3d 300 (7th Cir. 2018)), the Supreme Court denied the state of Indiana’s petition for a writ of certiorari with respect to this issue (Box v. Planned Parenthood of Indiana and Kentucky, Inc., 139 S. Ct. 1780 (2019)).

934   Maxwell Mehlman and Sonia Suter The FDA regulates in vitro diagnostic (IVD) genetic test kits as medical devices. There are three classes of medical devices, depending on their degree of risk, with Class III being the highest risk. The FDA has classified approximately 100 IVD test kits as Class II or III medical devices, such as test kits that examine the genetic makeup of cancerous tumors to enable more targeted forms of chemotherapy. Class I device manufacturers must follow good manufacturing practices; manufacturers of Class III devices additionally must obtain premarketing approval from the agency; Class II devices do not need premarketing approval, but, in addition to meeting good manufacturing practice requirements, they must adhere to special regulatory requirements. Most genetic tests are laboratory-­developed tests, rather than IVD test kits or DTC testing. Historically, the FDA has exercised extremely limited regulatory authority over laboratory-­developed genetic tests, in part because until recently there were few of them and in part due to concern that laboratory-­developed tests may not qualify as “medical devices” under FDA law. The only FDA requirement has been that chemicals used in testing, called reagents, that a laboratory purchased rather than formulated itself are subject to Class I good manufacturing practice requirements. The growth in laboratory-­ developed testing, including genetic testing, sparked concerns about safety and efficacy, so, in 2014, the FDA issued two draft guidance documents proposing to phase in additional regulatory requirements over nine years.15 Then, in 2017, the FDA proposed a modified regulatory approach under which existing tests, but not future tests, would be exempt from new regulation.16

3.2  Regulation of Laboratories Whereas test products are regulated by the FDA, most laboratories that perform genetic tests used in clinical decision-­making are regulated under the Clinical Laboratory Improvement Amendments of 1988 (CLIA) by the Centers for Medicare and Medicaid Services (CMS), the federal agency that oversees Medicare. CMS exempts New York and Washington from CLIA, however, and defers to strict state regulatory systems in those states. CLIA requires the laboratories to comply with good laboratory practices, conduct proficiency testing, and meet certain personnel requirements. In addition, CMS conducts routine inspections to assure that tests are valid analytically, meaning that they 15  US Food and Drug Administration, Draft Guidance for Industry, Food and Drug Administration Staff, and Clinical Laboratories: Framework for Regulatory Oversight of Laboratory Developed Tests (LDTs), Oct. 3, 2014, https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/ GuidanceDocuments/UCM416685.pdf; US Food and Drug Administration, Draft Guidance for Industry, Food and Drug Administration Staff, and Clinical Laboratories: FDA Notification and Medical Device Reporting for Laboratory Developed Tests (LDTs), Oct. 3, 2014, https://www.fda.gov/ downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM416684.pdf 16  US Food and Drug Administration, Discussion Paper on Laboratory Developed Tests (LDTs), Jan. 13, 2017, https://www.fda.gov/downloads/medicaldevices/productsandmedicalprocedures/ invitrodiagnostics/laboratorydevelopedtests/ucm536965.pdf

Genetics and the Law: United States   935 can accurately detect and evaluate the DNA sequences of interest. However, CLIA does not require laboratories to demonstrate clinical validity, meaning that they can show that the test actually diagnoses a genetic condition or a genetically affected response to treatment. These gaps in CLIA oversight are one of the reasons the FDA is considering increasing regulatory oversight of laboratory-­developed tests, including review of ­clinical validity.

3.3  Regulation of Direct-­to-­Consumer Testing The most controversial type of genetic testing is DTC testing. DTC tests, including the necessary saliva collection bottles, can be purchased online and in drugstores. Consumers obtain their test results online. While DTC genealogy or ancestry testing was available earlier, DTC health-­related genetic testing was first offered in 2007 by the company 23andMe, which obtained DNA from saliva samples that customers mailed to the company. Originally, the company offered tests for 107 conditions and traits, including BRCA cancer mutations, Crohn’s disease, cystic fibrosis, hemochromatosis, prostate cancer, Parkinson’s disease, and type 1 and 2 diabetes, and for non-­disease traits as “avoidance of errors,” “breastfeeding and IQ,” “measures of intelligence,” and “memory.” In 2013, the FDA issued warning letters to DTC companies stating that they were selling unapproved tests, and the companies halted health-­related testing. Then, in 2015, the FDA reviewed data submitted by 23andMe and authorized it to sell a test to identify carriers of a rare, autosomal-­recessive blood disorder called Bloom syndrome. In 2017, the FDA approved ten more DTC health-­related tests, followed, in March, 2018, by three breast cancer tests. The agency classified these DTC tests as Class II medical devices and required DTC testing companies to give consumers information about the limits of the testing, in particular that it was not intended to diagnose a disease or condition, disclose anything about the consumer’s current state of health, or be used to make medical decisions. Another issue with DTC genetic testing is that the FDA considers genetic tests ordered by physicians to be laboratory-­developed tests, which the agency currently does not heavily regulate, rather than DTC tests, even if the tests are purchased directly by consumers. Some DTC testing companies provide online access to physicians who order the tests but never communicate or interact with consumers. A further concern is the companies’ use of test results. 23andMe, for example, invites its customers to agree to allow the company to share their geno- and phenotype information with commercial interests such as drug companies. The company promises that, unless the customer agrees otherwise, researchers will be given access only to de-­identified information, but consumers must take the company’s word that its coding and security systems would withstand an attempt to link test results to specific individuals. Moreover, customers have no control over how their genetic information is handled and secured by the commercial interests with which it is shared. This raises concerns about genetic discrimination and privacy, discussed in the next section.

936   Maxwell Mehlman and Sonia Suter

4  Genetic Discrimination and Privacy Because genetic information has the potential not only to diagnose disease but also to predict susceptibility to future illness, it can be of great interest to third parties who benefit from learning about an individual’s health risks. As a result, genetic testing raises concerns about potential stigmatization or discrimination by employers, insurers, schools, the military, or even potential future partners. Although these concerns have captured the public imagination since the early 1990s, when some studies asserted that genetic discrimination was a serious problem, the evidence for such discrimination then and today is equivocal. Nevertheless, as our understanding of genetic information improves and as genetic testing becomes more widespread, the risks of genetic discrimination may grow in the future. Whether or not genetic discrimination is a serious risk today, the public fears it. These worries have serious implications for healthcare and research if they prevent individuals from undergoing genetic testing for their own healthcare or to participate in genetics/ genomics research. The use of genetic information for medical decision-­making will be thwarted if people refuse genetic testing for fear of discrimination. As a result, both to combat potential genetic discrimination and to allay public fears, state legislatures and the federal government have enacted statutes to prohibit genetic discrimination and protect genetic privacy. Often these protections exist in combination when laws both prohibit certain entities from engaging in discriminatory uses of genetic information and prohibit their access to such information.

4.1  Genetic Anti-­Discrimination Legislation We begin with anti-­discrimination laws. In the United States, these efforts began at the state level with statutes prohibiting discrimination based on genetic information. Nearly all states enacted legislation dealing with genetic discrimination. Although the laws vary widely, most such legislation focuses on genetic discrimination in the employment and health insurance context. Typically, the laws prohibit employers and insurers from using genetic information to make, respectively, employment decisions—such as hiring, retention, compensation, etc.—and insurance decisions—such as whether to provide insurance, at what rates, and for what conditions. For some time, there was no genetics-­specific legislation at the federal level with respect to either employment or insurance discrimination, although two federal statutes offered some form of protection in these areas. Because the patchwork of state and federal legislation was incomplete in protecting against genetic discrimination, in 2008, after thirteen years of failed attempts, the US Congress finally passed, and President Bush signed into law, the Genetics Information Nondiscrimination Act (GINA), which prevents discriminatory uses of genetic information in health insurance and employment. The scope of federal protections is somewhat different in these two areas.

Genetics and the Law: United States   937 With respect to health insurance, initially only HIPAA, enacted in 1996, provided federal protections against genetic discrimination. It, however, only addressed employer-­ sponsored group health plans, with no protection for other forms of health insurance. When GINA was enacted, Title I prohibited health insurers from using genetic information to determine eligibility for health insurance, to set premium or contribution rates, or to impose preexisting condition exclusions. GINA fortified HIPAA’s protections with respect to group health plans and extended protections against genetic discrimination into the individual insurance market, which had not been protected at the federal level or in all states. GINA also strengthened protections by defining “genetic information” more broadly than many state laws. For example, some state genetic nondiscrimination laws define genetic information as the results of genetic tests. Because one can learn a great deal about genetic risk based on knowledge that a family member has a heritable condition, GINA expanded the definition to include family history. Both GINA and state genetics laws, however, apply their protections to information about predisposition to disease, as opposed to information about manifested diseases. The theory is that it is unfair to discriminate against someone who is not yet and may never become sick. Thus, under GINA, a health insurer could make insurance decisions based on information about manifested breast cancer in a woman who carries a BRCA mutation associated with heritable breast cancer. But it could not discriminate based on information about another woman with the same mutation if she did not have breast cancer. Not only is this dichotomy morally problematic—does the latter deserve access to health insurance any more than the former?—it also presents difficult line-­drawing problems. How do we determine whether a condition has manifested or not, particularly because the development of a disease can be a gradual process? Two years after GINA was enacted, however, the Patient Protection and Affordable Care Act (ACA) closed the gap between protections against health insurance discrimination for individuals with manifested genetic conditions and those merely at increased risk. The ACA guarantees access to health insurance by prohibiting refusal of coverage or underwriting based on preexisting conditions, which includes genetic information about both manifested disease and disease predisposition. The ACA therefore avoids the potential inequities with respect to health insurance inherent in genetics-­specific laws. In addition, it essentially makes Title I of GINA irrelevant. How secure these ACA protections are, however, remains to be seen given that the statute has been weakened by legal challenges as well as congressional and executive actions. One example of efforts that undermined these protections was the previous Trump administration’s promulgation of a regulation that expands “short-­term” health insurance plans that are not required to comply with the ACA. Recently, the Biden administration revoked by excluding people with preexisting conditions and offering more limited benefits than ACA plans, their premiums were significantly lower than ACA plans. President Biden, however, issued an executive order within days of taking office suggesting his administration would undo the prior administration’s efforts to undermine the ACA. An even greater threat to the ACA, however, is the litigation initiated by twenty state governors and attorneys general alleging that the statute is unconstitutional. In 2018, a federal court ruled in their favor on the grounds that the ACA’s requirement that individuals have

938   Maxwell Mehlman and Sonia Suter health insurance coverage is unconstitutional.17 The Fifth Circuit also found the individual mandate unconstitutional, but it remanded to the lower court to decide whether that rendered the entire statute invalid.18 The fate of the ACA is now in the hands of the US Supreme Court, which heard oral arguments just a week after the presidential election and will soon issue a ruling.19 In the employment context, the American with Disabilities Act (ADA), enacted in 1990 and amended in 2008, prohibits employment discrimination based on disabilities, which includes manifested genetic diseases that substantially limit a major life activity. The ADA and its amendments, however, do not specifically address genetic discrimination. Some courts have found that individuals with minimal symptoms are covered if employers regard them as disabled, although it is unclear whether individuals merely at an increased risk of genetic disease would be covered on the same grounds. GINA’s Title II, however, directly addresses genetic discrimination in employment, prohibiting the use of genetic information to make employment decisions like hiring, discharging, determining compensation and other conditions of employment, etc. Like Title I, Title II uses a broad definition of genetic information, thereby providing broader protections against genetic discrimination in employment than most state laws. Unfortunately, because GINA protects against predisposition to illness as opposed to manifested illness, individuals in the early stages of a disease may find themselves vulnerable to employment discrimination in spite of GINA and the ADA. Mild symptoms or biomarkers of their illness might not meet the criteria for protection under the ADA, but the presence of some symptoms would leave them unprotected by GINA. Thus, this combination of federal laws does not prohibit all possible forms of genetic discrimination in employment. Although the United States has fairly robust federal protections against genetic discrimination in employment and health insurance, very few laws offer protection in the context of life, long-­term care, disability, mortgage, and auto insurance; education; commercial or real property transactions; etc. No federal law and only some state laws address these potential areas of genetic discrimination, each of which raises different policy concerns. Where state legislation exists, it varies considerably across jurisdictions with respect to which entities are covered and what combination of protections are used. Some legislation prohibits certain insurers from requiring genetic tests, performing genetic tests without informed consent, or obtaining genetic test results without the source’s consent. Florida is the first state to have enacted legislation that bans life, long-term care, and disability insurers from using genetic information for insurance decisions, and the only state that bans life insurers from doing so. Often, however, the statutes do not prohibit these insurers from using genetic information for underwriting decisions or only allow the use of genetic information for such purposes if the decisions are actuarially justified (i.e., based on sound risk assessments). In those instances, individuals who have undergone genetic testing must disclose results if asked to avoid invalidating the terms of any insurance contract they may try to obtain. Given that we still have much to learn about the disease risks of various mutations 17  Texas v. United States, 340 F. Supp. 579 (N. Dist. TX 2018). 18  Texas v. United States, 945 F.3d 355 (5th. Cir. 2019). 19  California v. Texas, 140 S. Ct. 1262 (2020); Texas v. California, 140 S. Ct. 1262 (2020).

Genetics and the Law: United States   939 in ­different populations, such as those with family histories and those without, it may be difficult to determine which insurance decisions are truly actuarially sound.

4.2  Genetic Privacy Legislation Another way to prevent discrimination and other harms based on genetic information is through laws protecting genetic privacy. Such legislation has value in its own right given the very personal nature of genetic information. Genetic information is tied to our identity. It links us to family members and can shape our life plans. Furthermore, our genes (in combination with environmental influences) shape who we are and will become in terms of our physical appearance, temperament, capacities, and health. Genetic privacy therefore is intrinsically valuable as well as consequentially important in helping to prevent discrimination. In the United States, some genetic privacy protections exist at the federal level. GINA, for example, not only prohibits employers and health insurers from using genetic information for employment and insurance decisions, but it also prohibits them from accessing genetic information, a form of genetic privacy protection. Neither entity may request or require individuals or their family members to undergo genetic testing or request, require, or purchase the individual’s genetic information, with a few narrow exceptions for employers. Because it can be difficult to prove discriminatory purpose, these prohibitions prevent health insurers and employers from accessing the information in the first place, preempting the possibility of discrimination. Such federal genetic privacy protections, however, are limited because they only apply to employers and health insurers. In addition, a few courts have incorrectly interpreted GINA’s definition and been unwilling to find violations of GINA when employers obtained an employee’s genetic information.20 The HIPAA Privacy Rule, promulgated by the Department of Health and Human Services under HIPAA’s mandate, offers additional federal protection of genetic privacy. It requires written authorization for use or disclosure of all personal health information, including genetic information,21 that is not used for medical treatment, payment, or healthcare operations, and it limits uses and disclosures to the minimum required to 20  See, e.g., Poore v. Peterbilt Bristol, L.L.C., 852 F. Supp. 2d 727 (W.D. Va. 2012) (finding that family history is not “genetic information” under GINA, in direct conflict with GINA’s definition of genetic information). See also Sonia M. Suter, GINA at Ten Years: The Battle over “Genetic Information” Continues in the Courts, 5 J.L. & Biosci. 495 (2019), https://doi.org/10.1093/jlb/lsz002, for a comparison of the different interpretations of “genetic information” that courts have used. 21  HIPAA, which was promulgated before GINA was enacted, protects “identifiable health information,” 45 C.F.R. § 160.103 (2014). When GINA became law, it required the Department of Health and Human Services to modify the Privacy Rule to expressly include genetic information within its protected health information and to define it according to the language of GINA. This was consistent with the Department’s prior issued guidance that genetic information was considered protected health information under the Privacy Rule. Modifications to the HIPAA Privacy, Security, Enforcement, and Breach Notification Rules Under the Health Information Technology for Economic and Clinical Health Act and the Genetic Information Nondiscrimination Act; Other Modifications to the HIPAA rules, 78 Fed. Reg. 5566–01, 5661–5662 (Jan. 25, 2013).

940   Maxwell Mehlman and Sonia Suter achieve those goals. Protections under the HIPAA Privacy Rule are limited, however, because they only apply to health plans, healthcare providers, healthcare clearinghouses, and any business associates that help these entities carry out their healthcare activities and functions. This means that no federal law protects the privacy of genetic information in the possession of DTC online companies like 23andMe. Most states also have genetic privacy legislation, although the nature of protection varies. Many laws require consent (sometimes written and/or informed) for the collection, retention, and/or disclosure of genetic information, usually with exceptions for law enforcement purposes, research on de-­identified information, newborn screening, etc. A handful of jurisdictions protect genetic privacy by declaring that genetic information is the property (sometimes modified as exclusive, real, personal, or sole property) of the person from whom the information was obtained or the person whose genome was analyzed. It is not fully clear what the full implications are of defining genetic information as property, particularly given the various meanings that property has under the law. These statutes do, however, underscore the notion that the information is the individual’s to control. At least one court has held that such laws do not conflict with cases suggesting that individuals do not have property rights in their excised tissue.22 Finally, some state privacy protections prohibit certain entities such as school districts, specific kinds of insurers, or employers from collecting or obtaining genetic information without consent.

4.3  Genetic Discrimination in the Military Since 1993, the US military has collected DNA from all service members to enable the identification of remains. It also screens them for genetic blood diseases such as sickle cell and glucose G6PD deficiency. The introduction of whole-­genome and whole-­exome sequencing in civilian medicine has sparked interest in their use as diagnostic and screening tools by the military. The Air Force, for example, is conducting a pilot study in which primary care providers are offering whole-­exome sequencing to healthy service members,23 and, in the future, the military might seek to test potential recruits for genetic risk factors that might affect their performance, especially in combat. Genetic screening and testing in the military raises special concerns about potential discrimination because GINA does not apply to the military.24 The armed forces have adopted equal opportunity policies that state that it is unlawful to discriminate based on 22  Perenboom v. Perlmutter, No. 2013-CA-015257 (Fla. Cir. Ct. Apr. 7, 2017) (finding that allowing a conversion claim for unconsented taking of another’s genetic information is consistent with cases like Moore v. Regents of University of California, 793 P.2d 479 (1990), because genetic information is different from genetic material). 23  National Institutes of Health, National Library of Medicine, ClinicalTrials.gov, “Enabling Personalized Medicine Through Exome Sequencing in the U.S. Air Force (MilSeq)” (2017), https:// clinicaltrials.gov/ct2/show/NCT03276637 24  Susannah Baruch & Kathy Hudson, Civilian and Military Genetics: Nondiscrimination Policy in a Post-­GINA World, 83 Am. J. Human Genetics 435, 439 (2008).

Genetics and the Law: United States   941 genetic information,25 and some military geneticists maintain that military service cannot be adversely affected by test results that reveal genetic risk factors until the risks actually materialize.26 However, genetic test results are likely to be placed in service members’ military records to which their commanders may have access. As greater numbers of variants are discovered that might affect military service, considerations of individual and unit effectiveness and the national interest may override concerns about privacy and discrimination. Once service members leave the military, they are protected by GINA. Originally, Department of Defense policy denied disability compensation and benefits to service members who experienced injury or illness during their time of service if the condition was “congenital or hereditary.” However, the National Defense Authorization Act of 2008 provided that service members with more than six months of service are entitled to compensation and benefits so long as there is no clear and unmistakable evidence that they had a hereditary or congenital disease at the time of enlistment or if the disease was aggravated by their service.27

4.4  Limits of Genetic Nondiscrimination and Privacy Legislation Although the United States has many laws that address genetic discrimination and privacy, such legislation is inadequate in a few respects. First, as noted, they are not as robust as some might desire, leaving many areas of potential discrimination unprotected. Second, their definition of genetic information is incomplete and fails to acknowledge developments in science and emerging technologies. Genetic variants are not the only predictive health information. Evidence of environmental effects on gene expression—epigenetics—or microbiome data, for example, are also highly relevant to future health risks, but such information is not protected under genetic legislation. Indeed, the definitional challenges of defining genetic information to distinguish it from other health information raises conceptual and moral questions as to whether genetic information is sufficiently unique to warrant special legal protections not always accorded to other medical information. For example, like some genetic information, non-­genetic medical information, such as HIV infection, high cholesterol levels, and hypertension, can identify predisposition to illness. Similarly, non-­genetic medical 25  E.g., Air Force Instruction 36–2706, Equal Opportunity Program, Oct. 5, 2010, 11: https://www. af.mil/Portals/1/documents/eeo/afi-­36-­2706.pdf 26 See Handout to Potential Participants, “Enabling Personalized Medicine Through Exome Sequencing in the U.S. Air Force (MilSeq)” (“A variant found in genetic testing cannot result in any action unless that individual suffers symptoms during their time of service and those symptoms limit the member’s ability to carry out their duties”). 27  US Department of Defense, DoD Directive Type Memo Implementing 2008 NDAA. Revised DoD Instruction E3.P4.5.2.2. on “Hereditary and/or Genetic Diseases” (2008), https://m.realwarriors.net/ ides/M/docs/DTM_on_NDAA_2008.pdf

942   Maxwell Mehlman and Sonia Suter information can be a basis for discrimination by employers and insurers if there are no legal prohibitions against that. In addition, like our genes, many other health risk factors, such as in utero exposure and environmental conditions, are not in our control. Finally, some of the genetics legislation in the United States raises tensions between the goal of precision medicine—to encourage the use of genomic data to make lifestyle changes—and the goal of genetic privacy. Prohibiting access to genetic information by entities like employers, for example, limits their ability to adjust environmental risks or make other accommodations to maximize the well-­being of the individual. In short, as is often the case with legislation that addresses emerging technologies, the law has not always succeeded in keeping up with the science or fully responded to the various policy concerns that arise in the field of genomics medicine.

5  Clinical Applications of Genomics The goal of genomics research is not only to identify genes associated with illness for diagnostic purposes, but also to use the technology either to treat, prevent, or ameliorate disease. The three areas in which genomic research is focusing on clinical interventions are gene therapy, modifications of the germline, and precision medicine.

5.1  Gene Therapy The development of recombinant DNA technology, which combines DNA from different organisms and can be used, among other things, to enable bacteria to produce therapeutic agents, led to the first efforts to regulate genetic modifications for therapeutic purposes. In 1974, the National Institutes of Health (NIH) established the Recombinant DNA Advisory Committee (RAC) and required it to review and approve all NIH-­ funded gene therapy experiments to ensure that they followed NIH biosafety guidelines. To reduce their liability risk, private research sponsors voluntarily sought RAC approval for their experiments. The FDA asserted regulatory authority over gene therapy in 1984.28 Importantly, the agency stated that it would not impose any special regulatory requirements for gene therapy trials. The agency regulates gene therapy primarily as a “biologic,” as defined in the Federal Food, Drug, and Cosmetic Act.29 Within the FDA, biologics are regulated by 28  US Food and Drug Administration, Statement of Policy for Regulating Biotechnology Products, 50 Fed. Reg. 50,878 (1984). 29  42 USC 262(i) (defining a biologic as “a virus, therapeutic serum, toxin, antitoxin, vaccine, blood, blood component or derivative, allergenic product, protein (except chemically synthesized polypeptide), or analogous product or arsphenamine or derivative of arsphenamine (or any other trivalent organic arsenic compound), applicable to the prevention, treatment, or cure of a disease or condition of human beings”).

Genetics and the Law: United States   943 the Center for Biologics Evaluation and Research (CBER). Questions might be raised about whether gene therapy should be treated as biological or other types of products, which the FDA has legal authority to regulate, or the practice of medicine, over which the agency does not have jurisdiction. Nevertheless, the only court challenges have concerned agency efforts to regulate certain stem cell therapies, which were unsuccessful. The FDA requires sponsors of gene therapy experiments to obtain an approved Investigational New Drug application (IND) from CBER. The sponsor must include evidence of safety and efficacy in a Biologics License Application (BLA) to CBER, and it can market its gene therapy when the BLA is approved. In 2017, FDA approved BLAs for the first three gene-­therapy products in the United States.30 In response to industry complaints about duplicative regulatory requirements, the NIH in 1997 renounced regulatory oversight by the RAC over gene therapy and relegated it to a purely advisory role. In 2016, NIH limited the RAC to advising only about exceptional matters related to gene therapy. Finally, in 2018, NIH proposed to eliminate RAC review of gene therapy protocols altogether and instead give it an advisory role on emerging biotechnologies generally.31 The introduction of CRISPR-­cas9 gene editing technology, which makes altering DNA easier, cheaper, and more precise, has raised hopes for new gene therapies. At the same time, some concern has been expressed that it could encourage do-­it-­yourself experimenters to conduct risky genetic modification experiments without adequate regulatory oversight.

5.2  Germline Modifications Gene therapy can involve modification of or transfer of DNA into somatic cells, cells that are not involved in reproduction. A specific form of gene therapy, called germline genetic modification, however, involves modifications or introductions of DNA in reproductive cells (eggs, sperm) or in early-­stage embryos, which can be passed on to future generations. The prospect of germline modification has raised a number of concerns, including safety concerns about unanticipated adverse effects on future generations, fears of altering the gene pool, worries that it is a slippery slope that will lead to enhancement as opposed to merely disease prevention, and concerns about social justice, equality, and discrimination. In contrast, it may offer unique benefits to parents who carry recessive disease genes and wish to have genetically related children who are disease free. In response to germline experiments in 2014 and 2015 on monkey and human embryos, the US Congress precluded the FDA from considering any INDs for research “in which a human embryo is intentionally created or modified to include a heritable

30  Id. at 1. 31  Francis S. Collins and Scott Gottlieb, The Next Phase of Human Gene-­Therapy Oversight, N. Eng. J. Med., doi: 10.1056/NEJMp1810628 (August 15, 2018).

944   Maxwell Mehlman and Sonia Suter genetic modification.”32 Moreover, NIH will not fund human germline modification experiments.

5.3  Precision Medicine Potential clinical applications of genomic medicine do not only involve manipulation of the genome. Genomic information can also be used for what has been called precision medicine. The classic Western approach to diagnosing and treating medical disorders is group-­based: patients with similar characteristics are grouped into the same disease categories, and clinical studies of interventions focus on effects in groups of subjects, rather than in individuals. Although the result has sometimes been called “one-­size-­fits-­all” medicine, it actually is “one-­size-­fits-­some” since many patients are not helped by or suffer adverse effects from treatment. Advances in genetics have led to a more refined, patient-­specific understanding of disorders and responses to treatment, such as more targeted treatments for cancer. In 2007, the US Department of Health and Human Services (HHS) launched the “Personalized Health Care Initiative” to foster more research and attention to individual differences.33 The initiative articulated an ambitious goal: “Personalized health care is envisioned as a system in which doctors, pharmacists, and other health care providers customize treatment and management plans for individuals.”34 After concerns were raised beginning in 2013 that truly individualized medicine was unrealistic, the “personalized medicine” approach was changed to “precision medicine.” Rather than promising individually tailored medical care, precision medicine aims at employing new genetic and other knowledge to produce a better, more “precise” understanding of diseases and treatments. The critical task for precision medicine is to create large databases containing genotypic and phenotypic patient information that can be analyzed for useful therapeutic insights and correlations. In 2015, President Obama announced a Precision Medicine Initiative, a federal effort to establish precision medicine databases, and, in 2018, the NIH launched the “All of Us” project with the goal of collecting and analyzing DNA and medical records on 1 million individuals. One aspect of precision medicine is pharmacogenetics or -genomics, where genetic information is obtained about patients and used to prescribe drugs more safely and effectively. Currently, the FDA requires drug manufacturers to include pharmacogenetic information in the labelling of approximately 150 drugs and to recommend genetic testing prior to prescribing about a dozen of them.

32  Consolidated Appropriations Act, 2016, Public Law no. 114–113 (2015); www.congress.gov/ bill/114th-­congress/house-­bill/2029/text 33 DHHS, Personalized Health Care: Opportunities, Pathways, Resources (Sept. 2007), https://www. genome.gov/pages/policy/personalizedmedicine/dhhsreportpersonalizedhealth.pdf 34  Id. at 4.

Genetics and the Law: United States   945 Another legal issue raised by precision medicine is preserving the privacy and confidentiality of the personalized medical information contained in precision medicine databases, as discussed in Part 4.

Bibliography ACMG Board of Directors, ACMG Policy Statement: Updated Recommendations Regarding Analysis and Reporting of Secondary Findings in Clinical Genome-Scale Sequencing, 17 Genetics Med. 68 (2015). American Medical Association, Disclosure of Familial Risk in Genetic Testing, Code Med. Ethics § 2:131 (2014–2015). American Society of Clinical Oncology, American Society of Clinical Oncology Policy Statement Update: Genetic Testing for Cancer Susceptibility, 21 J. Clin. Oncol. 2397 (2003). American Society of Human Genetics, Social Issues Subcommittee on Familial Disclosure, Professional Disclosure of familial Genetic Information, 62 Am. J.  Human Genetics 474 (1998). Jonathan S. Berg et al., Deploying Whole Genome Sequencing in Clinical Practice and Public Health: Meeting the Challenge One Bin at a Time, 13 Genetics Med. 499 (2011). Peter Benn et al., Obstetricians and Gynecologists’ Practice and Opinions of Expanded Carrier Testing and Noninvasive Prenatal Testing, 34 Prenatal Diagnosis 145 (2014). Paul R. Billings et al., Discrimination as a Consequence of Genetic Testing, 50 Am. J. Human Genetics 476 (1992). Ellen W. Clayton et al., The Law of Genetic Privacy: Applications, Implications, and Limitations, 6 J.L. & Biosci. 1 (2019), https://doi.org/10.1093/jlb/lsz007 Ellen W. Clayton et al., Managing Incidental Genomic Findings: Legal Obligations of Clinicians, 15 Genetics Med. 628 (2013). Francis Collins, A Brief Primer on Genetic Testing: World Economic Forum, National Human Genome Research Institute, Jan. 24. 2003, at https://www.genome.gov/10506784/a-briefprimer-on-genetic-testing/ Judith Daar, A Clash at the Petri Dish: Transferring Embryos with Known Genetic Anomalies, 5 J.L. & Biosci. 219 (2018). M. De Castro et al., Genomic Medicine in the Military, 1 NPJ Genomic Med. 15008 (2016). Asude A. Durmaz et al., Evolution of Genetic Techniques: Past, Present, and Beyond, BioMed Research Int’l (2015), Article ID 461254, https://doi.org/10.1155/2015/461524 Joel Feinberg, The Child’s Right to an Open Future, in Whose Child? Children’s Rights, Parental Authority and State Power (William Aiken & Hugh Lafollette eds., 1980) GenomeWeb, Reference Labs See Average-Risk Patients Fueling Continued Rise in Noninvasive Prenatal Testing, GenomeWeb, Dec. 11, 2017, at https://www.genomeweb.com/clinical-labmanagement/reference-labs-see-average-risk-patients-fueling-continued-rise-noninvasive#.W7jrXWhKhaR Robert C. Green et al., ACMG Recommendations for Reporting of Incidental Findings in Clinical Exome and Genome Sequencing, 13 Genetics Med. 565 (2013). Guttmacher Institute, Abortion Bans in Cases of Sex or Race or Genetic Anomaly, Aug. 1, 2020, at https://www.guttmacher.org/state-policy/explore/abortion-bans-cases-sex-or-raceselection-or-genetic-anomaly

946   Maxwell Mehlman and Sonia Suter Mark A. Hall, Legal Rules and Industry Norms: The Impact of Laws Restricting Health Insurers’ Use of Genetic Information, 40 Jurimetrics 93 (1999). Hastings Center Special Report, The Ethics of Sequencing Newborns: Reflections and Recommendations (Josephine Johnston et al., eds. 2018). Human Fertilisation & Embryology Authority, Pre-Implantation Genetic Diagnosis (PGD), at https://www.hfea.gov.uk/treatments/embryo-testing-and-treatments-for-disease/ pre-implantation-genetic-diagnosis-pgd/ Human Fertilisation & Embryology Authority, Pre-Implantation Genetic Screening (PGS), at https://www.hfea.gov.uk/treatments/embryo-testing-and-treatments-for-disease/ pre-implantation-genetic-screening-pgs/ Jamie S. King, Not This Child, 60 UCLA L. Rev. 2 (2012). Larry Levitt et al., Why Do Short-Term Health Insurance Plans Have Lower Premiums Than Plans that Comply with the ACA?, KFF, Oct. 31, 2018, at https://www.kff.org/health-reform/ issue-brief/why-do-short-term-health-insurance-plans-have-lower-premiums-than-plansthat-comply-with-the-aca/ Maxwell J. Mehlman, Mark A. Rothstein, & Sonia M. Suter, Genetics: Ethics, Law and Policy (2020 5th ed.). National Human Genome Research Institute, Welcome to the Genome Statute and Legislation Database, at https://www.genome.gov/policyethics/legdatabase/pubsearch.cfm National Institutes of Health, The NIH Director, Statement on NIH Funding of Research Using Gene-Editing Technologies in Human Embryos (2015), www.nih.gov/about/director/04292015_ statement_gene_editing_technologies.htm. National Institutes of Health (NIH), Genetic Testing: How It Is Used for Health Care, NIH Fact Sheets, at https://report.nih.gov/nihfactsheets/ViewFactSheet.aspx?csid=43 Nuffield Council on Bioethics, Noninvasive Prenatal Testing: Ethical Issues (2017). Robert Pear, Trump Officials Broaden Attack on Health Law, Arguing Courts Should Reject All of It, N.Y.  Times, Mar. 29, 2019, at https://www.nytimes.com/2019/03/25/us/politics/obamacare-unconstitutional-trump-aca.html?action=click&module=RelatedCoverage&pgtype= Article®ion=Footer Jessica Roberts, Protecting Privacy to Prevent Discrimination, 56 William & Mary  L.  Rev. 2097 (2015). John Robertson, Children of Choice: Freedom and the New Reproductive Technologies (1994). Mark A. Rothstein, GINA at Ten and the Future of Genetic Nondiscrimination Law, Hastings Ctr. Rep., May–June, 2018, at 5. Mark A. Rothstein, Reconsidering the Duty to Warn Genetically At-Risk Relatives, 20 Genetics Med. 286 (2018). Mark A. Rothstein, GINA’s Beauty is Only Skin Deep, 22 Gene Watch 9 (2009). Sonia  M.  Suter, Did You Give the Government Your Baby’s DNA? Rethinking Consent in Newborn Screening, 15 Minn. J. L. Sci. & Tech. 729 (2014). Sonia  M.  Suter, Disentangling Privacy from Property: Toward a Deeper Understanding of Genetic Privacy, 72 George. Wash. L. Rev. 737 (2004). Sonia  M.  Suter, Genomic Medicine: New Norms Regarding Genetic Information, 15 Houston J. Health L. & Pol’y 83 (2015). Sonia M. Suter, Legal Challenges in Genetics, Including Duty to Warn and Genetic Discrimination, in Genetic Counseling: Clinical Practice and Ethical Considerations (Laura Hercher et al. ed., 2020).

Genetics and the Law: United States   947 US Food & Drug Administration, FDA Allows Marketing of First Direct-to-Consumer Tests that Provide Genetic Risk Information for Certain Conditions, Apr. 6, 2017, at https://www. fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm551185.htm US Food & Drug Administration, FDA Authorizes, with Special Controls, Direct-to-Consumer Test that Reports Three Mutations in the BRCA Breast Cancer Genes, Mar. 6, 2018, at https:// www.fda.gov/news-events/press-announcements/fda-authorizes-special-controls-directconsumer-test-reports-three-mutations-brca-breast-cancer

chapter 45

Gen etics a n d the L aw Europe Mette Hartlev

1  Introduction to the European Legal and Ethical Framework In Europe, the legal and ethical aspects of gene technology are both a national concern for European states and a European concern addressed by both the Council of Europe (CoE) and the European Union (EU). The CoE and the EU provide a legal framework for the development of national law in the European countries that are members of the EU (27 states) and/or the CoE (47 states). Despite regulation at the European level, there may be significant national differences depending on the room left for national discretion by European regulation. The European legal and ethical frameworks are to a large degree intertwined, and the development of patients’ rights in Europe is aligned with the development of international human rights law and the notion of a human rights-­based approach to health. This explains why human rights law has a major impact in the area of genetics and the law in Europe. The CoE has been engaged in bioethical issues since the beginning of the 1980s and has adopted a number of recommendations and resolutions concerned with developments in biomedical science and technologies. Most notable is the adoption of the CoE’s Convention on Human Rights and Biomedicine (Oviedo Convention) in 1997, which focuses on a number of selected human rights aspects of developments in biomedicine and has been signed and ratified by 29 of the 47 CoE Member States. The convention is only applicable to those states which have signed and ratified the convention. The overall aim of the Convention is, as stipulated in Article 1, to ‘protect the ­dignity and identity of all human beings, and to guarantee everyone without

950   Mette Hartlev ­ iscrimination, respect for their integrity and other rights and fundamental freed doms with regard to the application of biology and medicine’. In addition, the supremacy of the human being is stressed in Article 2, according to which the ‘interests and welfare of the human being shall prevail over the sole interest of society or science’. The Convention includes a number of general legal principles applying to medical interventions on human beings, including informed consent and the right to privacy. The Convention also includes more specific rules and regulations for predictive genetic tests, interventions on the human genome, and genetic discrimination. The Convention is supplemented by a number of additional protocols, including a protocol on genetic testing for healthcare purposes and a protocol on medical research, both of which are of particular relevance for genetic testing.1 The Oviedo Convention builds on a number of earlier adopted legal instruments, of which some specifically targeted genetic testing and others included provisions relevant for genetic testing. The Oviedo Convention should also be seen in relation to the European Convention on Human Rights (ECHR), which includes provisions of particular relevance for the application of biomedicine, including genetic technologies on human beings, such as the protection of the right to private and family life stipulated in Article 8 and the prohibition against discrimination in Article 14. The European Court of Human Rights (ECtHR) has provided rulings in several cases involving human rights aspects of medical technologies, including genetic technologies.2 The EU is primarily oriented at ensuring economic and political cooperation and has not adopted a similar comprehensive legal framework for patients’ rights and ethical aspects of new technologies as has the CoE. But the EU has adopted provisions to promote health. In particular, Article 168 of the Treaty of the Functioning of the European Union (TFEU) demands a high level of health protection to be pursued in Union policies, thus stressing the need to develop better and more sustainable healthcare systems. In addition, the Charter of Fundamental Rights of the European Union includes some provisions relevant for the application of gene technology. The Charter stresses the importance of respect for human dignity and the obligation to obtain an informed consent in connection with interventions in the medical field and respect the rights to ­privacy and data protection. It also specifically prohibits eugenic practices and emphasizes the right not to be discriminated against. The importance of a high level of health protection and access to preventive care and a right to benefit from medical treatment is also stressed. In addition to the TFEU and the Charter, secondary legislation on clinical trials, medical devices, and in vitro diagnostics devices is potentially of importance for issues

1  CETS nos. 203, 195. 2  CoE publishes on a regular basis a list of case law from the ECtHR where bioethical issues have been addressed. See more on this at https://www.coe.int/en/web/bioethics/ European-­Court-­of-­Human-­Rights

Genetics and the Law: Europe   951 relating to genetic testing.3 As genetic testing raises issues of privacy and data protection, the EU General Data Protection Regulation (GDPR) is also of particular relevance. Although the EU has not adopted comprehensive legal regulation in this area, it has for many years been involved in ethical and legal issues related to the application of science and new technologies in broader areas. As observed by Sheila Jasanoff, biotechnology policy in the EU ‘became a site of interpretive politics, in which important elements of European identity were debated along with the goals and strategies of European research’.4 An important actor in this field is the European Group on Ethics in Science and New Technologies (EGE), which is an assembly of independent experts. The EGE gives independent and interdisciplinary advice to the European Commission on ethical questions related to science and new technologies and can take up issues either on its own initiative or by the request of the Commission. The EGE has adopted and published several opinions in areas related to genetic testing, including prenatal testing, biobanking, genetic testing in the workplace, synthetic biology, information and communication technology, and new health technologies and citizen participation.5 Although the opinions of the EGE have no formal legal status, they appear to have an impact on the development of EU regulation in the area of new biotechnologies. Hence, the role of the EGE has been described as the ‘role of a broker’ serving to mitigate between the actors striving to promote a biotechnology-­enabling regulatory environment and those actors who are sceptical towards such a regulatory environment.6 Together, various pieces of CoE and EU regulation provide a legal and ethical framework for national laws regarding application of genetic technologies on human beings in Europe. Some of the regulatory mechanisms provide clear legal boundaries whereas others leave room for national discretion and variation.

2  Genetic Testing and Screening The use of genetic testing in healthcare services is generally rising, including the use of more comprehensive genetic analyses such as whole genome sequencing (WGS) and genome wide association studies (GWAS). Whereas genetic testing was previously mostly used in families with a known hereditary history or as a diagnostic tool to establish the cause of a health condition, it has become increasingly mainstream medical 3  Directive 2001/20/EC to be replaced by Regulation 536/2014 (clinical trials); Directive 2007/47/EC to be replaced by Regulation 2017/745 (medical devices); Directive 98/79/EC to be replaced by Regulation 2017/746 (in vitro diagnostic medical devices). 4  Sheila Jasanoff, Designs on Nature. Science and Democracy in Europe and the United States (Princeton University Press 2005) 92. 5  See a list at https://ec.europa.eu/info/publications/ege-­opinions_en. 6  Helen Busby, Tamara Hervey, and Alison Mohr, ‘Ethical EU Law? The influence of the European Group on Ethics in Science and Technology’ (2008) E L Rev 33, 803.

952   Mette Hartlev practice to make use of carrier and predictive genetic testing. This raises issues regarding the application of traditional patient rights and fundamental societal rights, such as the rights to information and self-­determination, protection of privacy, and the right not to be discriminated against. As outlined in the introduction to this chapter, genetic testing can take place in various contexts, for various purposes, and with different methods. These differences are important from both an ethical and a legal perspective. In general, the right to information and the obligation to obtain informed consent apply in all situations where genetic technologies are used on human beings. This European part of the chapter addresses informed consent and genetic counselling and specific issues related to genetic testing of adults, children, foetuses, and embryos. Genetic research will also be discussed.

2.1  Regulating genetic tests As genetic testing is becoming an increasingly used diagnostic tool in healthcare services, legal regulation for carrying out genetic testing and screening has become a crucial issue. The traditional medical law regulation put significant emphasis on the patient’s informed consent as a justification for medical interventions, but it has been questioned whether it is sufficient to rely exclusively on informed consent for performing genetic testing or whether interests other than patient autonomy should be taken into consideration (e.g., if there should be restrictions on which genetic tests should be available, and who should be entitled to perform them). The CoE’s Oviedo Convention and its Additional Protocol on Genetic Testing for Health Purposes provide a general legal framework for genetic testing in Europe with the overall purpose of defining and safeguarding fundamental rights of individuals subjected to genetic testing for health purposes. The Convention is only concerned with predictive genetic testing, which is considered as more problematic than diagnostic genetic testing due to uncertainty regarding interpretation of test results and because of the adverse impact predictive testing may have on the individual in other areas, such as insurance and employment. According to Article 12 of the Convention, predictive genetic testing may only be performed for health purposes or for scientific research linked to health purposes. Consequently, this rules out predictive testing required as a condition to obtain an insurance policy and (in most situations) an employment contract as well as a number of direct-­to-­consumer genetic testing services performed outside the healthcare setting (see further in Sections 2.6 and 3). The Additional Protocol is concerned with a broader range of genetic tests (both diagnostic and predictive tests, as well as testing for healthy carrier status, pharmacogenetic tests, and genetic screening programs). It also provides a more comprehensive legal framework for genetic testing because it elaborates on several issues related to the application of the Oviedo Convention to genetic testing, such as the provision of information and genetic counselling prior to consenting and respect for private life.

Genetics and the Law: Europe   953 Compared to the Convention, only a few countries (so far only six) have both signed and ratified the Additional Protocol, but the protocol may still serve as inspiration for national legislation in this area. The EU does not specifically regulate genetic testing of adults, and EU regulation in this area is scarce. However, both the fundamental rights enshrined in the Charter of Fundamental Rights of the European Union and secondary EU legislation may have importance for such testing. According to Article 3 of the Charter, informed consent is required for all interventions in the medical field, and eugenic practices are prohibited. Test kits used for testing should also fulfil certain safety and quality standards. The regulation of genetic testing reflected in the CoE and EU legal framework provides both a binding legal framework and soft law inspiration for legal regulation by the European countries. Some countries have adopted separate laws specifically addressing genetic testing (e.g., Norway, Spain, France, Austria, Germany, Sweden and Switzerland), whereas other countries have specific provisions on genetic testing in other national laws or regulate genetic testing through the general legal framework for provision of healthcare services. In terms of restricting the use of genetic testing, some countries specifically restrict genetic testing in general to healthcare purposes (e.g., France, Norway, and Switzerland), whereas others only restrict the use of predictive tests to healthcare purposes (e.g., Germany). Switzerland requires an authorization to perform genetic testing for medical purposes, and other countries demand special expertise from the persons who prescribe a test or who are involved in providing information or genetic counselling to patients (e.g., specialization in genetics). There are also examples of national legislation that limit specific kinds of genetic tests to certain conditions. As an example, the 2007 Swiss Act on Human Genetic Testing does not allow prenatal testing for conditions which do not impair health, and a number of countries prohibit prenatal and preimplantation genetic testing for establishing the gender of the foetus or embryo. Genetic screening is also subject to special regulation and restrictions in, for example, the 2003 Norwegian Biotechnology Act and the 2009 German Human Genetic Examination Act. The legal limitations reflected in the national laws and the European legal framework for genetic testing demonstrate that genetic testing is a medical technology that touches on a broader range of ethical and societal issues than other kinds of treatment and research.

2.2  Informed Consent and Genetic Counselling Informed consent for medical interventions is a general principle in health law, one that is reflected in both national laws of the European countries, in Article 3(2) of the EU Charter of Fundamental Rights of the European Union, and in the Oviedo Convention. Individuals should be properly informed before a medical intervention so that they can make an enlightened decision.

954   Mette Hartlev The right to information raises special issues in regard to genetic testing both with respect to the right to know and the right not to know, as well as with dealing with unexpected findings and managing genetic information in a family context. Genetic counselling—both before and after a genetic test—is generally considered to be good practice both among geneticists and in the legal framework. The Oviedo Convention stresses the importance of both the right to information and genetic counselling in connection with use of genetic testing. The right not to be informed is also acknowledged. In general, Article 5 of the Convention requires that an intervention in the health field may only be carried out based on the individual’s free and informed consent, and this requires that the individual receive in advance of the intervention ‘appropriate information as to the purpose and nature of the intervention as well as on its consequences and risks’. The right to information also follows from Article 10 of the Convention, according to which ‘Everyone is entitled to know any information collected about his or her health’. At the same time, Article 10 also stresses that a wish not to be informed—the right not to know—should also be respected. In the context of genetic testing and screening, this implies—according to Article 16 of the Additional Protocol on Genetic Testing for Health Purposes—that the tested individual is entitled to know any information collected about him or her derived from the test and that the results of the test should be returned in a comprehensible form unless the individual wishes not to be informed.7 Genetic testing for heritable breast cancer can be used as an illustrative example. Such tests may be offered both as diagnostic tests, where the patient has been diagnosed with breast cancer, or as predictive tests (e.g., based on a family history of breast cancer). Before deciding on having a test, the patient must be offered sufficient information to be supported in making an autonomous decision regarding whether or not to undergo genetic testing. Apart from information about the test and the physical intervention, the patient should also be offered information regarding the potential outcome of the test (e.g., whether it will provide a clear diagnostic answer and what the consequences—in terms of making decisions regarding removal of breasts and ovaries—would be). The broader consequences for the patient and genetic relatives (e.g., children or siblings) should also be addressed. Furthermore, patients should also be informed about the right not to know (e.g., about the hereditary character of test results). After being tested, patients are entitled to receive information about the results, but may also decide not to be informed and remain ignorant. Results of genetic tests may be very complex to interpret and understand for patients and therefore give rise to uncertainties regarding risks for future health conditions, the severity and onset of the conditions, the impact on family members, and on reproductive decisions. A predictive genetic test for heritable cancer will, for example, not provide a clear test result, but solely information regarding the risk of developing breast cancer (70–80% depending 7  In exceptional cases it is, however, possible to restrict the right to information and the right not to be informed by law if this is in the interest of the person concerned: see Article 10.3 of the Oviedo Convention and Article 16.4 of the Additional Protocol, CETS no. 203.

Genetics and the Law: Europe   955 on the genetic variant). The test may also reveal information regarding variants of unknown significance in genes other than those related to breast cancer (so-­called secondary findings). Consequently, the patient is often left with test results involving a degree of uncertainty, which can both produce anxiety and make it difficult to make decisions. Apart from the patient, this information could also be of significance for relatives and thus raises the issue of whether information should be provided to those and by whom (see Section 4). To support patients both in making decisions regarding testing and in how to respond to the results, it is stressed in Article 7 of the Additional Protocol that genetic counselling prior to and after the test is of utmost importance.8 Current technologies used for genetic testing (e.g., WGS or GWAS) have significantly increased the amount and type of information that can be obtained and thereby produce risks of so-­called unexpected, incidental, or secondary findings, which could both be health and non-­health related. Genetic testing for breast cancer could, for example, reveal information regarding a susceptibility to developing Alzheimer’s disease, and it could also produce unexpected results regarding family bonds, including paternity. Secondary findings may also occur in clinical encounters other than genetic testing, and the physicians’ duty of care will in many situations speak in favour of informing patients about unexpected findings if this could be vital for preventing or treating serious disorders. However, genetic tests may also reveal secondary findings for which there is no prevention and treatment available, or the finding may be of unknown significance. In these situations, it is less obvious how to respond, as test results may bring damage to the tested individual or to other family members. The response to this dilemma has been different in the United States and Europe. In the United States, the American College of Medical Genetics and Genomics (ACMG) recommends screening of a specific list of genes (currently 56 genes) that are associated with late-­onset conditions or have a long asymptomatic phase and to report such findings back to the patient unless they have opted out. In contrast, the European Society of Human Genetics (ESHG) recommends that tests should be designed to limit the risk of producing secondary findings and that such findings should be reported back to the patient only if they concern a serious health problem and treatment or prevention is possible. The CoE framework does not take an explicit position on this issue. Article 8 of the Explanatory Report to the Additional Protocol emphasizes that the potential implications of the results must be communicated to the individual beforehand, and—in this context—the protocol also specifically mentions information regarding biological links in a family. This will give the patient an opportunity to make a decision before the test regarding return of such results. The Explanatory Report leaves it to national law to settle whether unexpected findings should be available and communicated to the person concerned. In this regard, national law should reflect a proper balance between the

8  Paragraphs 71–85 of the Additional Protocol, CETS no. 203, include very detailed and comprehensive guidance as to provision of information and genetic counselling.

956   Mette Hartlev interest in having access to information and the damage such information may bring to the person or other family members. Some countries require that the patient should be informed beforehand about the possibility of secondary findings and thereby be given a chance to decide whether he or she would want to be informed. As examples both the Swiss Act on Human Genetic Testing (Article 14.3) and the Spanish Act on Biomedical Investigation (Article 47) make it mandatory to inform patients about the possibility of secondary findings before the test. In Denmark, it is also mandatory to inform patients about secondary findings when having a comprehensive genetic analyses, and, according to Article 2.5 of Executive Order 359/2019 to the Health Act, it is furthermore required that patients make an explicit, written decision before the test regarding whether or not to be informed about such findings. The German Act on Human Genetic Examination (Article 8) gives the patient the right to decide not to be informed and also to demand the destruction of test results. The legal situation in the United Kingdom is not crystal clear. However, after the Montgomery case,9 it seems that there could be a legal duty to report secondary findings if these are considered to be of significance to the patient. In the UK, Genomics England (100,000 genomes project) has adopted an approach in between the ACMG and the ESHG, where a more limited list of genes is investigated based on patient consent. Apart from the issue of right to information and to genetic counselling, the legal protection of adults without decision-­making capacity is also important. This is a w ­ ell-­known issue in any medical intervention, and the Oviedo Convention stipulates (Article 6) that medical interventions may only be carried out for the direct benefit of the person concerned and only with an authorization from his or her legal representative (proxy) or an authority, person, or body provided for by law. Proxy consent to genetic testing can, however, give rise to special concerns, as the proxy may have an individual or family-­ related interest in the test and therefore also be the beneficiary of the test. This calls for special attentiveness to the ‘best interest’ requirement. The Additional Protocol on Genetic Testing for Health Purposes addresses this special situation (Article 13 together with Article 10) and leaves room for testing of a person without decision-­making capacity if the test is necessary for obtaining a preventive, diagnostic, or therapeutic benefit for a family member of the tested person and the test does not expose the person to more than minimal risks and burdens. In this context, the German Act on Human Genetic Examination can serve as an example. According to Article 14 of the Act, genetic testing of persons without capacity to consent may only be carried out if it is necessary for and in the best interest of the person concerned. It is, however, also possible to perform genetic testing in case of a planned pregnancy of a genetically related person, if it is necessary to establish whether there is a risk that a certain genetically caused condition could appear in the offspring, and if testing can be done without risk and burden to the person concerned.

9  Montgomery v Lanarkshire Health Board [2015] UKSC 11.

Genetics and the Law: Europe   957

2.3  Genetic Testing of Children and Adolescents Genetic testing of children and adolescents raises special concerns regarding respect for children’s right to health as well as their right to self-­determination, information, and privacy. In some situations, a genetic test is necessary to diagnose a child and to provide prevention and treatment and thus is clearly in the child’s best interest. In other situations (e.g., predictive testing of late-­onset disorders), testing may be more problematic because it could deprive children of the right to make their own decisions and thus shape their own future. Knowledge about the child’s susceptibility to develop a genetically related condition could influence the way parents relate to the child and could also raise concerns regarding the risk of discrimination in various areas of life. Testing of children could also be motivated by interests of siblings, parents, or other family members, as in the situation addressed in Section  2.2 regarding testing of adults without decision-­making capacity. There is no unified age limit in Europe for obtaining legal capacity in regard to healthcare decisions, so there may be different age limits applying in national laws. In the context of the UN Convention on the Rights of the Child (CRC), a child is defined as a human being below the age of 18 years, unless maturity is attained earlier according to national law. The Oviedo Convention also leaves it to national law to determine the age of decision-­making capacity. According to Article 6 of the Convention, decisions to perform a genetic test on a child or adolescent without legal capacity to consent should always be authorized by his or her legal representative or an authority, person, or body provided for by law. When making such decisions, it is important to take the fundamental rights of children into consideration. Several provisions in the CRC have importance for genetic testing of children. In general, the best interests of the child should always be a main consideration in all aspects related to the child (Article 3), and the child’s right to health should be respected (Article 24). When making decisions, the child has a right to be heard and to influence decisions regarding its life (Article 10), and the child’s right to privacy (Article 16) should also be respected. As a whole, the rights enshrined in the CRC reflects a fundamental ethical principle of ensuring the child’s right to an open future. Parents are seen as custodians of children’s rights until they reach the age of having actual and legal capacity to make decisions regarding their own life. This implies that parents should not make decisions on behalf of the child which may be postponed until the child achieves the capacity to make her or his own decisions. The Oviedo Convention reflects these fundamental principles. It stipulates that any intervention on persons who do not have the capacity to consent should always be carried out for his or her direct benefit (Article 6). This is further clarified in Article 10 of the Additional Protocol, which requires that testing of persons not able to consent should be deferred until the person attains the capacity to consent, unless delayed testing will be detrimental to her or his health or well-­being. This means that testing may only be justified during childhood if this will allow for preventive and therapeutic

958   Mette Hartlev measures to be taken, or if this may spare the child from highly invasive regular examinations. Apart from leaving the decisions to the child when she or he reaches the age of maturity, this also serves to ensure protection of the child’s right to privacy as parents’ interest in private information is limited to what is necessary to fulfil their role as guardians. This is an important aspect of the child’s ethical right to an open future. On the other hand, the right to an open future also requires decisions to be made during childhood whenever this is required to ensure the child’s right to health and the best prospects for the child’s future. Both the Oviedo Convention and the Additional Protocol require that the legal representative must be given prior information regarding the purpose and nature of the test and the implications of the results and that the child or adolescent should also be provided with prior information based on his or her capacity to understand the information. This is in line with CRC Article 10, where the child’s right to information and right to be heard is stressed, and also with Article 24 of the Charter of Fundamental Rights of the European Union, which underscores children’s right to express their view and have their opinions taken into consideration in accordance with their age and maturity. As is clear, the CoE regulatory instruments together with the EU Charter of Fundamental Rights provide a rather comprehensive and child-­rights compliant framework for genetic testing of children and adolescents, and a number of European countries have adopted legislation which addresses the special issues arising from genetic testing of minors. Some countries restrict testing of minors to tests which are of significant importance to the child’s health during childhood. The Norwegian Biotechnology Act (Article 5-­7) prohibits predictive testing of children under the age of 16 for late-­onset conditions unless testing during childhood is necessary to avoid or reduce health damage. Other countries have no special regulation specifically targeting children but regulate testing of persons without the capacity to consent, which also includes children. This regulation restricts testing to what is considered to be in the patients’ best interest and may also include other regulatory elements. The German Act on Human Genetic Examination (Article 14) provides a comprehensive regulation of genetic testing of persons unable to consent. Such testing is only allowed if this is necessary to avoid, prevent, or treat a genetically caused illness or health condition of the subject person and under the condition that the test will not burden or harm the tested person physically or mentally. The person must be informed in an understandable manner before the test is performed, and informed consent from the guardian or legal representative must be provided. Less comprehensive rules are, for example, found in the Swiss Act on Human Genetic Testing (Article 10) and French Public Health Code (Article L 1131-­1), which emphasize that genetic testing must be in the best interest of the person concerned (the child) and that informed consent must be provided by the guardian or legal representative. Even though national laws are concerned with the best interest of the child, they generally lack a more comprehensive framework which takes the multitude of legal issues outlined in the introduction to this part of the volume into consideration. However, the

Genetics and the Law: Europe   959 legal framework is normally supplemented by professional guidelines and practices in genetics, which may account for other pertinent issues, such as how to preserve privacy rights of the child if parents request more comprehensive genetic testing than necessary or obtain information regarding secondary findings.

2.4  Newborn Screening Newborn screening programs have been in place in a number of European countries since the early 1960s. The screening covers a number of early-­onset conditions—mostly genetic conditions—with a wide variety in test panels from country to country. The legal and organizational setup of the program also differs significantly between countries. In some countries testing is mandatory, whereas parents’ consent is required in other countries. Some countries store blood samples from children for unlimited duration, whereas samples are destroyed after some years in other countries. The World Health Organization (WHO) criteria for public health screening programs10 have been widely applied to newborn screening. These criteria were developed 50 years ago and were modified in 2008 to reflect the increasing use of genomic technologies in screening. However, it is still debated whether the criteria are outdated in the light of scientific development. In general, the criteria should ensure that screening programs are only initiated when the screened conditions are considered serious health problems, where there is a safe and validated screening test, and where there is an effective treatment or intervention available which—with early detection and intervention— could improve the health outcome. In addition, it is also expected that the screening program is clinically, socially, and ethically acceptable to both health professionals and the public and that proper information about the test and the treatment is provided for potential participants. These criteria are clearly reflected in the Article 19 of the Additional Protocol for Genetic Testing for Health Purposes which requires that genetic screening programs for health purposes must be (a) recognized for their health relevance for the whole population or for a section of the population concerned, (b) the scientific validity and effectiveness of the program has been established, (c) appropriate preventive or treatment measures are available to the person concerned, (d) on an equitable basis; and (e) adequate information about the program is provided to the section of the population concerned, which should include information regarding the existence, purpose, and means of accessing the program and also of the voluntary nature of participation. A few European countries have special legal regulation for genetic screening programs. The German Act on Human Genetic Examination (Article 15) allows for genetic screening provided the objective is to determine genetic traits which are of importance for preventing, treating, or avoiding an illness or health problem. The program must 10  James MG Wilson and Gunnar Jungner, ‘Principles and Practice of Screening for Disease’, Public Health Paper no. 34, World Health Organization (1968).

960   Mette Hartlev also be approved by the German Genetic Commission, and the standard regulation regarding informed consent and genetic counselling seems to apply to genetic screening. Similar rules follow from the Swiss Act on Genetic Testing (Article 12), which also requires authorization from a competent federal authority, which can be granted provided early treatment or prevention is possible, the test produces reliable results, and appropriate genetic counselling is provided. The Swiss Expert Commission for Human Genetic Testing must be consulted before authorization is granted, and it may also be necessary to consult the Swiss National Advisory Commission on Biomedical Ethics. The Norwegian Biotechnology Act (Article 5-­6) presents a different model. According to the Act, the King of Norway (in practice the Government) can give instructions regarding genetic screening programs which may derogate from the standard rules regarding written informed consent and genetic counselling. The differences between the German and Swiss Act on the one side and the Norwegian Act seem to reflect a stronger focus on the individual and individual rights in German and Swiss society compared to a more community-­oriented approach favoured by Norway as a Nordic welfare society. Currently, newborn screening programs use targeted genetic tests to identify infants with specific genetic conditions included in the screening panel. However, since WGS and GWAS testing has become cheaper and thus more common as a diagnostic tool, it has been debated whether WGS/GWAS should be performed routinely on all newborns. The ethical aspects of routine genetic sequencing of newborns have been explored and discussed, highlighting both advantages and risks.11 Whereas there is wide support for using WGS and GWAS to diagnose severely sick infants, the ethical, legal, social, and economic aspects seem to speak against use of genome and exome sequencing as a routine in newborn screening. In this context, it has also been questioned whether such a screening program would comply with the WHO criteria for screening programs. Consequently, a cautious approach is needed.

2.5  Preimplantation and Prenatal Genetic Diagnosis Genetic testing is also widely used to identify risk of genetic conditions in embryos and foetuses. Previously, this kind of testing was mostly available in case of a family history of hereditary diseases. However, today routine screening of foetuses for chromosomal disorders is practised in a number of European countries in addition to more targeted analyses in situations where more specific indications for genetic analyses exist. Such prenatal testing (PND) is performed at various stages during pregnancy to test whether the foetus is affected by specific genetic or chromosomal disorders. The prenatal testing program also relates to the legal options provided by national laws for abortion in cases where the child is at risk of having a severe condition. Genetic analyses of embryos are 11  See e.g., Friedmann et al., ‘Genomic Newborn Screening: Public Health Policy Considerations and Recommendations’, BMI Medical Genomics (2017) 10, 9.

Genetics and the Law: Europe   961 also increasingly used. Preimplantation genetic diagnosis (PGD) is performed on embryos prior to implantation to ensure either that the embryo is not affected by a specific suspected genetic disorder (typically family related) or—in rare cases—to ensure genetic compatibility (HLA compatibility) between the child to be and a sibling in need of tissue donation (so-­called saviour sibling situations). Access to genetic analyses of embryos and foetuses relates to the mother/parents’ reproductive autonomy and right to private and family life. Knowledge about the future child’s genetic makeup may give parents important information and in some situations also allow them to make decisions regarding termination of a pregnancy or embryo selection. However, this is an area of European controversy with significant differences in national laws (as discussed in Chapter—(Rothmann)). There is no unified position in Europe in regard to the ethical and legal status of the embryo and the foetus. The legal frameworks for termination of pregnancies, use of artificial reproductive technologies, and selection of embryos as part of reproductive services differ considerably in Europe. Most European countries have introduced regulation allowing some kind of use of PGD, although with significantly different criteria applying for testing. Likewise, there are considerable differences among European countries in regulation of PND and whether national law allows for abortion in cases where the foetus has a genetic or other condition. A few countries completely prohibit PND as well as abortion in these situations, whereas others allow for abortion within a certain period of the pregnancy and based on specific criteria regarding the seriousness and/or onset of the disorder. Some countries (e.g., Germany, and Norway) have addressed PGD and/or PND as part of general regulation of genetic testing, while others have included PGD and PND as part of regulation of artificial reproduction or termination of pregnancies. There are also countries where PGD and/or PND is not regulated in law but by professional or administrative guidelines (e.g., Denmark and the UK [PND]). In general, use of PGD is limited to testing for serious health conditions, and many jurisdictions explicitly prohibit testing for nonmedical purposes, including testing to establish the gender unless for diagnostic purposes. In some countries (e.g., Denmark and France), testing is limited to known family genetic disorders. A number of countries also require that there is no appropriate treatment available for the condition, for which the embryos are tested. In addition, some countries have introduced more specific criteria. For example, the Swiss legislation requires that the disease manifest itself before the age of 50 (Article 5a of the Reproductive Medicine Act), and, according to German law, PGD is not allowed for conditions which do not manifest themselves before the age of 18 (Act on Human Genetic Examination, Article 15). Most countries do not specify in their legislation those conditions for which testing is available but may provide regulation in administrative or professional guidelines. As an example, the UK provides a list for which testing can be carried out (currently around 600 disorders). Apart from identifying a genetic or chromosomal disorder, PGD can also be used for HLA-­typing, where the purpose is to identify an embryo with a tissue type matching a severely sick sibling. Whereas most European countries allow for some kind of PGD to

962   Mette Hartlev identify an embryonic disorder, fewer countries permit use of PGD/HLA-­typing due to the ethical concerns affiliated with this technology. The ECHR does not expressly clarify whether the unborn is subject to protection by the ECHR. Based on case law from the ECtHR, the legal situation can best be described as blurred. In general, the ECtHR leaves a wide margin of appreciation to national laws in this area. On the other hand, the ECtHR has also stressed that the foetus is not completely outside the scope of protection of the ECHR. Still, the ECtHR also acknowledges that the interests of the foetus must be balanced against the rights of the pregnant woman, thus justifying abortion in situations where the woman’s life and health is at risk. The case law of the ECtHR also shows that once a Member State has allowed certain reproductive choices, it should be applied in a consistent manner. Indeed, in a case against Italy, the ECtHR concluded that it was not compliant with the right to private and family life, laid down in Article 8 ECHR, to prohibit PGD while accepting PND.12 In the case, a couple who both were healthy carriers of a cystic fibrosis gene were prevented from getting access to PGD, which was only available to infertile couples. The Court found that the restriction on use of PGD in Italian law served the legitimate purpose of protecting morals and the rights and freedoms of others. However, the Court also found a blanket ban on PGD for fertile couples to be disproportionate. It referred to the fact that Italian law allows the applicants to terminate pregnancy on medical grounds if the foetus is affected by a disease. Likewise, in a case against Poland, the ECtHR stressed that once a diagnostic service is allowed by law (in this case PND), it is a violation of the pregnant woman’s right to private and family life if this service is not available for the women concerned due to, for example, ignorant or disrespectful treatment of the treating physician.13 The European Group on Ethics in Society (EGE) has provided an opinion regarding PND where it recommends that PND should rely on free and informed consent of the woman/couple and that nondirective genetic counselling should be available both prior to the testing and afterwards. It also stipulates that selection based on sex and other nonmedical characteristics is ethically unacceptable and should be prohibited and that there should be careful consideration regarding respect for privacy and sharing of information in a family setting.14 The Oviedo Convention touches on PGD and prenatal testing in Article 14, which stipulates that use of techniques for medically assisted procreation should not allow choosing a future child’s sex except in situations where serious hereditary sex-­related diseases could be avoided. This does not specifically cover situations of prenatal testing of foetuses where artificial reproductive technologies has not been involved. The principle of nondiscrimination enshrined in Article 14 is, however, a general principle in both European human rights regulation as well as in EU regulation. Thus it could be argued 12  Costa and Pavan v Italy App no. 54270/10 (ECtHR, 28 August 2012). 13  R.R. v Poland App no. 27617/04 (ECtHR, 26 May 2011). 14  European Group on Ethics in Science and New Technologies, Opinion no. 6, Ethical aspects of prenatal diagnosis, 20/02/1996.

Genetics and the Law: Europe   963 that abortion of foetuses exclusively based on gender is not compliant with European ethical principles and legislation. In general, this is an area of ethical sensitivity where there is a wide room for national discretion. However, this discretion is not unfettered and must be exercised within the framework of CoE and EU regulation, which specifies some limits and requires con­sist­ ency in national law. As prenatal testing is becoming more and more advanced and is moving from targeted testing to applying more comprehensive tests revealing a vast array of information, it will raise even more difficult questions regarding how to balance reproductive autonomy with basic ethical and legal principles at national and European levels.

2.6  Direct-­to-­Consumer Genetic Testing The issues outlined in the previous sections presuppose that genetic testing takes place within a healthcare context. However, genetic testing is increasingly available as a commercial product outside the healthcare setting. Individuals can purchase so-­called direct-­to-­consumer genetic tests (DCT) for various purposes, spanning from curiosity about their ancestry, to suspicion regarding biological links (e.g., paternity), or for health-­related purposes. European regulation in this area consists first and foremost of EU consumer protection laws15 and regulation of in vitro diagnostic (IVD) medical devices. The focus is here first and foremost on safety and analytical and clinical evidence and validity of genetic tests, as well as on labelling and ensuring trustworthy information to the consumer. The upcoming EU Regulation on IVD medical devices (Article 4) also requires Member States to ensure that patients are properly informed before being tested and provided with appropriate access to genetic counselling when offered predictive testing for a condition where no treatment options are available. In addition to EU regulation, the Oviedo Convention and the Additional Protocol may also have some impact. In general, Article 12 of the Oviedo Convention restricts use of predictive testing to health-­related purposes. DCT may be seen as beneficial for the individual as it facilitates access to medical technologies without depending on national healthcare services as a gatekeeper. Individuals may also be better suited to protect the privacy of genetic information when it is not collected and stored within healthcare institutions. On the other hand, use of DCT also gives rise to both ethical and legal concerns, regarding, for example, the lack of proper information and genetic counselling and risks for privacy. As DCT provides for non-­health related genetic testing, this product also seems to be at odds with Article 12 of the Oviedo Convention.

15  See, e.g., Directive 2005/29 EC.

964   Mette Hartlev European countries have taken different approaches to DCT, and the regulatory situation has been described as a ‘fragmented landscape’.16 In practice, DCT seems to be banned in France due to a number of legal requirements and restrictions. In other countries, general or special regulation applies to some aspects of DCT. It could, for example, be rules banning genetic testing for other than healthcare purposes or requirements regarding medical supervision, genetic counselling, or special informed consent obligations (see Sections 2.1–2.2). Such regulation is typically restricted to activities which fall within the definition of healthcare services and may consequently not apply to all kinds of DCTs. And, in general, it may be hard to regulate these tests in national laws because they are mostly sold through the internet and thus lie outside the scope of the national jurisdiction.

2.7  Genetic Research Genetic analyses are increasingly used in research, especially in research projects in the area of precision medicine (PM). Such research can both involve testing of individuals participating in a research project or be based on tissue samples stored in a biobank. In the case of biobank research, the individuals from whom the tissue sample originate may—or may not—have given informed consent to the use of samples for genetic research, and, with an increasing use of combination of big data and genetic research, the individual’s options for influencing the use of tissue and data for research purposes is challenged. Another related and pertinent issue is the management of secondary findings during the research project. May or should researchers return results to research participants, or should they rather keep silent? The Oviedo Convention explicitly states that the interest and welfare of the individual prevails over the interests of society and science (Article 2), thereby stressing the Kantian principle of respect for human dignity and noninstrumentalization of the human being. This means that genetic research must be performed with respect for the research subjects’ autonomy and right to be informed (or not informed) as well as the interests in genetic counselling and in privacy. Article 12 of the Oviedo Convention allows for predictive genetic testing in a research context provided the research is linked to a health purpose. This is the only provision in the CoE legal framework specifically addressing genetic research, but the general rules of the Oviedo Convention regarding informed consent (Article 5), right to information and right not to know (Article 10), and regarding scientific research (Article 15–18) also apply to genetic interventions including predictive testing as part of a research project. In addition, the CoE Additional Protocol for Biomedical Research applies to all sorts of research projects, including projects involving genetic testing. This additional protocol elaborates on and clarifies the Oviedo Convention’s provisions in a research context, and it includes several provi16  Louisa Kalokairinou et al., ‘Legislation on Direct-­To-­Consumer Genetic Testing in Europe: A Fragmented Regulatory Landscape’, J Comm Genetics (2018) 9, 117.

Genetics and the Law: Europe   965 sions of special relevance for genetic research evolving around the right to information and the right to privacy. According to Article 13(2) of the Additional Protocol for Medical Research, arrangements must be in place both to ensure respect for the research participant’s private life and confidentiality of personal data (see also Article 25) and to ensure access to information relevant for the participant’s decision to participate in the research project as well as access to information collected about the participants’ health as part of the research project (Article 26). In addition, Article 27 also prescribes a duty of care towards the participant in situations where the research ‘gives rise to information of relevance to the current or future health or quality of life of research participants’. This is of particular relevance to secondary findings. In this case, the information must be offered to the participant within a framework of healthcare or counselling, and due care must be taken to protect confidentiality and respect the wish of the participant not to receive such information (right not to know). EU regulation regarding clinical trials and medical devices is also relevant for genetic research. Although it does not specifically address genetic research, regulation regarding privacy, informed consent, and return of incidental findings also applies to genetic research. This means that genetic research in Europe takes place within a normative framework, which provides rather uniform guidance for research involving human research participants (especially within the European Union). At the same time, European regulation leaves room for national variation, and, in the area of biobank research, national regulation is rather heterogeneous. All research involving human research participants requires the research participants’ informed consent. Consequently, there is a legal obligation to inform participants about the purpose and methods used in the project, which will also involve information regarding use of genetic testing if this is part of the project. Some countries (e.g., Spain, Denmark, and France17) also require research participants to be specifically informed about the possibility of secondary findings and to have an option beforehand to express their wishes regarding return of results. In this way, research participants are in a position to decide whether to receive information about findings or rather stay uninformed. Some countries, however, still require researchers to return information—despite the participant’s expressed wish not to be informed—if the findings concern a serious disease for which prevention or treatment is available (e.g., Spain).18 When genetic research is exclusively based on tissues from clinical or research biobanks, the wishes of the persons from whom the tissue stems may be uncertain. Patients are not necessarily informed about and asked to give consent to the storage of tissue in biobanks, and, even if they have provided an informed consent, it may not be clear to them that the tissue will be used for genetic research. The same applies to research participants

17  Spanish Act on Biomedical Investigations, Article 15 and Articles 46–49; Denmark’s Executive Order no. 825 of 4 June 2020, Article 11; and France’s Public Health Code, Article L113–1. 18  Spanish Act on Biomedical Investigations, Article 4.

966   Mette Hartlev who have given informed consent to storage of tissue samples for secondary use without having specific knowledge about the character of future research. As in other areas, there is no unified position in European national legislation regarding this point. A number of countries require as a main rule informed consent from the person concerned in biobank research, but the legislation applies various forms of consent, and it may also provide for derogations. As examples, the Swiss Human Research Act accepts derogations from the consent requirement in exceptional situations where it is impossible or disproportionately difficult to obtain consent, or where asking for consent would present an undue burden for the individual concerned (Article 34). It is furthermore a condition that there is no documented refusal and that the interests of research outweigh the interest of the individual concerned to decide herself about the further use of tissue samples or data. A similar model applies in Spain and in Denmark,19 where it is also possible to derogate from the consent requirement provided it would be disproportionately difficult to obtain consent and that this will not impose any risk or burden to the individual concerned. In France, informed consent is as a main rule always required for genetic testing, but the Public Health Code authorizes an opt-­out model in situations where research is performed on tissue samples originally collected for another purpose with explicit consent (Article L1131-­1-­1, Article 1211-­2). In this situation, the individual concerned must be informed about the new purpose and is entitled to object to the tissue samples being used for a new purpose. Denmark also has an opt-­ out model for secondary use of tissue samples collected in a clinical context.20 In contrast to the French model, the person concerned is not specifically informed about the new purpose for which the tissue samples are being used but is entitled at any time to register in a special database if he or she does not want tissue samples to be used for research purposes. In situations where an individual is not informed and has had an opportunity to consider the use of tissue samples for genetic research, it leaves the researcher with difficult decisions regarding return of both primary and secondary findings. The individual concerned may have both an interest in being informed and in staying uninformed. Some countries have issued legal regulation (e.g., Spain, France, and Denmark21) which require researchers to return secondary findings in this situation provided the finding is actionable and concerned with a serious disorder. Other countries haven’t clarified this issue in law, but rely on professional guidance (e.g., the UK22) or leave it to the individual researcher to make a professional assessment.

19  Spain’s Act on Biomedical Investigations, Article 58, and Denmark’s Act on Research Ethics Review of Health Research Projects, Article 10. 20  Consolidated Act on Health no. 903 of 26 August 2019, Article 29. 21  Act on Biomedical Investigations, Article 4 and Article 9 (Spain), Public Health Code, Article L-­1-­2 (France), National Committee on Health Research Ethics, 2020 (Denmark). 22  UK Medical Research Council and Welcome Trust, Framework on the feedback of health-­related findings in research (2014).

Genetics and the Law: Europe   967

3  Genetic Discrimination Europe has a dark history of genetic discrimination. The Nazi social policy was based on eugenics to justify not only involuntary sterilization but also to endorse mass use of euthanasia. However, the eugenic movement was not exclusively a characteristic of the Nazi regime. It was supported throughout Europe, and forced sterilization was widely practiced in many European countries—also many years after World War II. These practices are today condemned in most European countries and constitute a violation of European fundamental and human rights provisions. The EU Charter of Fundamental Rights explicitly prohibits eugenic practices in medicine and biology (Article 3(2)), and it also puts a general ban on discrimination (Article 21(1)), which has been interpreted to include genetic discrimination. Article 11 of the Oviedo Convention explicitly prohibits any form of discrimination based on genetic heritage, and Article 14 of the European Convention on Human Rights prohibits discrimination in regards to the rights laid down in the Convention based on ‘any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property or birth and other status’. ‘Genetic feature’ does not appear on the list, but as it is a nonexhaustive list, and as genetic features share the same characteristics as the listed criteria, genetic discrimination could be covered by the European Convention on Human Rights. Even though former eugenic practices are condemned, there are concerns that the information provided by genetic testing about disorders or a predisposition or susceptibility to a genetic disorder could be used as a means for discrimination in various areas of life. In this context, it is important to recall that unequal treatment does not per se amount to discrimination. The concept of discrimination traditionally refers to unequal treatment without any objective and reasonable justification, and there may be justifications for treating persons differently based on their (different) genetic features. Considerations at both the European and national levels regarding the legal and ethical limits for use of genetic information has so far mostly focused on employment and insurance relations. However, in some European countries, genetic discrimination is addressed in more general anti-­discrimination laws (e.g., in Belgium, Croatia, France, and Serbia). The focus in the following will be on employment and insurance relations. In Europe, the general prohibitions laid down in Article 14 of the European Convention on Human Rights, together with Article 3(2) and Article 21(1) of the Charter for Fundamental Rights of the European Union and Article 11 of the Oviedo Convention provide a general legal platform for assessment of unequal treatment in employment and insurance relations. However, it is necessary to make a more context-­specific assessment to evaluate whether unequal treatment in these relations amounts to discrimination. There is only limited European regulation specifically targeting genetic discrimination in an employment or insurance context. At the national level, a number of countries have adopted laws which in one way or another regulate the use of genetic information for

968   Mette Hartlev nonmedical purposes, such as employment and insurance. Some countries have included provisions on use of genetic testing in employment and insurance relations in general regulation on genetic testing (e.g., Switzerland, Spain, Germany, Norway, and Sweden). Other countries have included special provisions on genetic testing and use of genetic information in general regulation of insurance or employment (e.g., Denmark and Belgium). In Ireland, the issue is addressed in the Disability Act, and in the UK the Equality Act and common law principles applies. In employment relations, employers may have various interests in having access to genetic information regarding their employees. There are important commercial interests in having a well-­functioning, healthy workforce, and this could lead employers to recruit individuals without undesirable genetic traits—and maybe even look for persons with certain welcomed genetic/personality characteristics (e.g., risk-­taking or leadership talent). In addition, employers have responsibilities for the health and safety of their workforce. If some employees have genetic susceptibilities exposing them to a high risk from particular work environments, employers could in certain situations consider offering genetic testing to employees in order to comply with health and safety responsibilities. Job applicants and employees certainly have an interest in not being exposed to discrimination based on an actual or uncertain risk of having a genetically related disorder in the future. At the same time, employees also have an interest in working in a safe and healthy working environment. Apart from the directly work-­related interests, employees also have general interests in protecting their right to privacy and the right not to know about a genetic predisposition. If, for example, employers are allowed to require a genetic test when recruiting new candidates, job applicants can be exposed to information about their genetic makeup that they would prefer not to know about. National legislation in Europe demonstrates a variety of approaches to genetic testing in employment relations. Most countries strive to balance the interests just outlined, allowing some kind of genetic testing if this is of vital interest for the employer and the employee (e.g., if health and safety issues are at stake). The German Act on Human Genetic Examination (Article 19–22) prohibits the employer from demanding an employee to have a genetic test or to ask for results of a previously taken genetic test, but derogations based on federal regulation may apply in the area of occupational health examinations and screening provided certain strict conditions are fulfilled. The German Act also explicitly prohibits genetic discrimination in employment relations (Article 21). The Swiss Act on Human Genetic Testing (Article 21–25) is very similar to the German regulation, but does only apply to predictive genetic testing, and it does not include a specific provision regarding prohibition of genetic discrimination in employment relations. Denmark has a general regulation of use of health examination and health information in employment relations23 which also includes use of genetic testing. The Act makes a distinction between diagnostic testing of existing health conditions, which may be justified in the job requirements (e.g., testing of eyesight for pilots), and predictive 23  Act no. 286 of 14 April 1996, on Use of Health Information on the Labor Market.

Genetics and the Law: Europe   969 tests of future health risks, which can only be justified by occupational health interests and under specified strict conditions. The national legislation is in line with the (nonbinding but interpretative) opinion adopted by the EGE regarding genetic testing in the workplace with recommendations regarding how to balance between the (partly) opposing interests just outlined. The opinion first of all recommends that medical examination must not be a criterion for selection of candidates and may first be offered after the selection phase. As a general rule, examination may only be offered for present health conditions and requires the informed consent of the employee. In exceptional cases, predictive testing may be offered if this is justified in health and safety concerns that cannot be addressed properly by other means. Furthermore, information obtained through such tests must be kept confidential to protect the privacy of the employee. The opinion is aligned with Article 12 of the Oviedo Convention which prohibits non-­health related predictive testing. This does not exclude testing for health and safety purposes. The EU Directive 2000/78/EC on equal treatment in employment relations may potentially also have an impact on unequal treatment based on genetic features. According to Article 1, the purpose of the Directive is to combat discrimination ‘on the grounds of religion or belief, disability, age or sexual orientation’ in employment and occupation. The provision does not refer to genetic features, but as genetic disorders could amount to a disability, this Directive may offer some protection. The European Court of Justice has in recent case law ‘stretched’ the concept of disability to include severe obesity, which shows some interpretational flexibility.24 Turning to insurance relations, there is a long tradition for using health-­related information—including information about heritable diseases—to assess the actuarial risk. Seen from the perspective of insurers, information regarding an existing genetically related disorder, as well as the risk or susceptibility of developing genetically related disorders in the future, is important information to avoid adverse selection. The insurance system runs the risk of being distorted if an individual—who knows about a genetic predisposition—obtains life or disability insurance with a very high coverage. Consequently, there may be legitimate interests for insurance companies in obtaining and making use of genetic information to make decisions regarding insurance coverage. Seen from the perspective of the person who wishes to obtain insurance, the situation looks different. There are crucial issues at stake regarding access to healthcare services, protection of right to self-­determination (including the right not to know), and the right to privacy. In regard to access, individuals living in countries with a mostly private healthcare system may be deprived of or have reduced access to healthcare services if insurance companies are allowed to refuse or reduce coverage based on a risk of having a genetic condition. In addition, if insurers are allowed to require predictive genetic testing as a condition for issuing a policy, this may subject individuals to a test and to obtain

24  See joined cases C-­335/11 and C 337/11, 2013 ECLI:EU:C:2013:222, and C 354/13, 2014 ECLI:EU:C:2014:2463.

970   Mette Hartlev information that they would prefer not to know about, thus depriving them of their right not to know. The most comprehensive instrument adopted at the European level regarding genetic discrimination in insurance relations is the CoE’s recommendation regarding use of health data—including genetic information—for insurance purposes. This is a soft law instrument, which—although not formally legally binding—should be taken into consideration in legal assessments. The recommendation deals with use of health information in general and is not restricted to genetic information, but some provisions are specifically concerned with the use of genetic testing and genetic information. In general, the recommendation requires insurers to justify the use of health (including genetic) information (paragraph 5). In regards to predictive genetic testing, it stipulates that predictive genetic testing must never be required as a condition to obtain insurance (paragraph 15) and that an ‘insurer will not be entitled to subject the conclusion or modification of an insurance policy to the holding of a predictive genetic test. Nor will the insurer be able to refuse the conclusion or modification of such a policy on the ground that the applicant has not submitted to a test’.25 As a main rule, results from predictive testing already performed must not be required or used by the insurer unless provided by national law and subject to appropriate safeguards (paragraph 16). This paragraph opens a space for national legal solutions, where such information may be required in regard to, for instance, life insurance beyond a certain level. This recommendation provides for a balanced and human rights-­compliant approach to the use of genetic information in insurance relations. National laws in Europe present diverse ways of balancing between the competing interests in this area. Some countries put most emphasis on protecting the right to self-­ determination and insurance holders’ right not to know as well as their right to privacy (e.g., France, Belgium, Denmark, Norway, and Sweden). Other countries are more open to acknowledge insurance companies’ interests in controlling actuarial risks and permit access to and use of genetic information for insurance policies exceeding a determined level (e.g., Switzerland, Germany, the Netherlands, and the UK). The Swiss Act on Genetic Testing (Article 26–28) prohibits both predictive and prenatal genetic testing for insurance purposes. However, use of genetic testing and information is allowed where a life insurance contract exceeds 400,000 CHF and disability insurance higher than 40,000 CHF, provided the reliability of the test and its scientific value has been demonstrated. The UK has no statutory regulation, but the government and the Association of British insurers have agreed on a Code for Genetic Testing in Insurance which restricts insurers’ possibility of requiring genetic testing or information as a precondition for obtaining a policy other than in situations where, for example, the life insurance exceeds £500,000 and the insured person has had a predictive test positive for Huntington’s disease. 25  CoE, Committee of Ministers, Explanatory report to Recommendation CM/Rec(2016)8 on the processing of health-­related data for insurances purposes—including data resulting from genetic tests, para 39–41.

Genetics and the Law: Europe   971 EU law does not specifically touch on the use of genetic information and testing in insurance relations. However, it has been argued that the general prohibition against discrimination laid down in Article 21(1) of the EU Fundamental Rights Charter could potentially be interpreted as including a prohibition against use of genetic features in insurance relations. This argument is based on the ECJ ruling in the Test Achats case, which was concerned with the use of gender as a basis for insurance policy calculation, but where the arguments could be transferred to discrimination based on genetic features.26 As the foregoing discussion indicates, there is no comprehensive European legal framework addressing the risk of genetic discrimination in various areas of life. Some areas—especially employment and insurance—have attracted regulatory attention. In these areas, the regulation is concerned with both the possible risk of discrimination as well as with protecting individuals against being ‘forced’ to accept predictive genetic testing and thereby being exposed to information they would prefer not to know about. Consequently, the right to self-­determination and right not to know are also a concern in this regulation. The risk of genetic discrimination also exists in other areas that have not yet attracted similar attention. However, a few European countries have general anti-­discrimination regulation which also covers discrimination based on genetic features. In a context where the use of genetic testing—including WGS and GWAS—is increasingly used, a more comprehensive legal framework seems to be of utmost importance.

4  Privacy Issues Genetic information relates to the health condition of the individual but also to other areas, such as family relations (including paternity). Genetic information can also be used for forensic purposes. In European law, it is generally recognized that such information belongs to the private sphere and is protected by the right to private and family life laid down in Article 8 of the European Convention on Human Rights and Articles 7 and 8 of the Charter of Fundamental Rights of the European Union. Protection of genetic information is evidently important not only to respect the privacy of the tested persons and their families, but also to avoid potential discrimination. With the EU GDPR, genetic information has become more visible as a data protection issue. Genetic data are defined in the GDPR Article 4 (13) as ‘personal data relating to the inherited or acquired genetic characteristics of a natural person which give unique information about the physiology or the health of that natural person and which result, in particular, from an analysis of a biological sample from the natural person in question’.27

26  C-­236/09.

27  See also GDPR, recital 34–35.

972   Mette Hartlev According to Article 9, genetic data are regarded as sensitive information, which means that they can only be processed in accordance with the special criteria provided in Article 9 and subject to the general data protection principles. In general, processing of sensitive data is allowed with the explicit consent of the data subject (Article 9(2)(a)), provided the data will be processed for a legal purpose. In other words, it is not possible to consent to the processing of data when the processing is prohibited according to national or EU law. This could be the case if use of genetic information in employment and insurance relations is prohibited in national or EU law (see Section 3). Article 9 also allows for processing of genetic data without an individual’s consent, for example, if this is necessary for the purposes of preventive health, diagnosis, and provision of health care; for occupational medicine or for the assessment of the working capacity of the employee; or for the provision of social care (Article 9(2)(h)). It is a condition that the processing is done by a professional who is subject to a legal or professional duty of confidentiality. It is also possible to process sensitive data for purposes related to public health (Article 9(2)(i)) and for research purposes (Article 9(2)(j) and Article 89). The right to privacy is also stressed in Article 10 of the Oviedo Convention and in more detail in Articles 16–18 of the Additional Protocol concerning Genetic Testing for Health Purposes, where Article 16 is concerned with privacy of genetic information, Article 17 with security and confidentiality of tissue samples, and Article 18 with relation to family members. The issue of privacy within family relations is of particular interest. As individuals share their genes with family members, information obtained through a genetic test may also reveal information regarding the health of genetically related family members. If, for example, prenatal testing reveals a monogenetic disorder (as, e.g., Huntington’s disease), this implies that a parent—who may not have been tested— is affected. Consequently, a person having a genetic test may obtain private information about other persons and may be in a position to control whether they should have access to this information. This has an impact on family members’ right to information as well as their right not to know. Normally, tested persons are encouraged to contact family members if the test shows a positive result so that they can decide whether to be tested or not. This will protect family members’ right to information and self-­determination but will, on the other hand, have an impact on the privacy of the tested person. Accordingly, it is not possible to uphold complete privacy regarding genetic information in a family context. The Additional Protocol strives to provide recommendations regarding how to navigate in this complex legal and ethical landscape. The explanatory report to the Additional Protocol recommends that national law provide appropriate provisions to ensure that family members can receive information, which may be crucial for their future health, while at the same time bearing in mind general principles of confidentiality and privacy protection. If the tested person does not agree to contact family members, it could be considered to establish a mediating or decision-­making body which could balance the interest of the persons concerned. The act of balancing between respecting patients’ rights of confidentially and self-­ determination and relatives’ interests and rights in obtaining important information regarding genetic susceptibilities has been addressed in various ways in the European

Genetics and the Law: Europe   973 countries. Most countries seem to rely on general medical law principles of confidentiality, professional duty of care, and emergency exemptions. Only a few countries have adopted legislations specifically addressing this issue. As examples, the German Human Genetic Examination Act (Article 8) requires a patient’s specific consent to inform third parties, including relatives, about test results that could be of importance for the relatives. The Norwegian Act on Biotechnology (Article 5–9) leaves it as a main principle to the patient to inform relatives. However, if patients are not able or willing to inform relatives, the healthcare services may inform relatives directly if the results concern a serious genetic disorder for which prevention or treatment is available and provided there is clear and significant risk of relatives being affected. However, if the patient protests against the relatives being informed, it must be respected. The French Public Health Act (Article L1131-­1-­2) puts more emphasis on the interests of relatives, making it obligatory for patients to inform relatives about a serious condition which can be prevented or treated. Still, patients cannot be sanctioned if they don’t inform family members. If patients are not interested in knowing the results themselves or in communicating the results to relatives, they must consent to information being communicated by the healthcare services. However, it seems that the French law does not allow disclosure if the patient refuses consent, making it very similar to the Norwegian law. The Swiss Act on Human Genetic Testing goes a bit further as it stipulates (Article 19.3) that, in cases where patients refuse consent, the physician can apply to the competent cantonal court to be released from his or her duty of professional secrecy. Under this policy, the protection of the overriding interests of the family members require that they receive information. Privacy issues can also occur in other situations, for example, when genetic testing is used to confirm biological links in cases of paternity testing, family reunification (wars, refugee situations, migration), and for forensic purposes. European legislation has especially been concerned with the storage of DNA profiles for forensic purposes28 and with use of DNA for paternity testing.29 In cases of storage of DNA for forensic purposes, the ECtHR has drawn a line based on a proportionality assessment of the storage period seen in relation to the gravity and nature of the crime. In cases regarding paternity testing, it has put significant emphasis on the child’s rights and interests in knowing his or her identity vis-­à-­vis possible competing interests of the potential biological father.

28 See S. and Marper v UK App nos 30562/04 and 30566/04 (ECtHR, 2 December 2008) and Aycaguer v France App no 8806/12 (ECtHR, 22 June 2017), where the Court found a violation of ECHR Article 8, and Peruzzo and Martens v Germany App no 7841/08 and 57900/12 (ECtHR, 4 June 2013) which the Court found inadmissible because domestic law clearly had struck a fair balance on the taking and retention of DNA material from persons convicted of serious criminal offences. 29  See e.g., Mifsud v Malta App no 62257/15 (ECtHR, 29 January 2019), Pascaud v France App no19535/08 (ECtHR, 16 June 2011), Jäggi v. Switzerland App no 58757/00 (ECHR 2006-­X), and Mikulić v Croatia App no 53176/99 (ECHR 2002‑I).

974   Mette Hartlev

5  Legal and Ethical Aspects Related to Treatment Currently, the possibilities of preventing genetically related disorders or providing treatment are limited. However, for some genetic disorders, preventive treatment is possible, as in cases of phenylketonuria (PKU), which is one of the conditions included in newborn screening programs (see Section 2.4) and where complications and impairments can be avoided by following a special low-­protein diet. Gene therapy is a treatment where genetic material is introduced to a person’s cells with the purpose of fighting or preventing a disease. It is still at an experimental stage, being explored in a research context or provided as experimental treatment. According to Article 13 of the Oviedo Convention, somatic cell gene therapy can only be undertaken for preventive, diagnostic, or therapeutic purposes and only provided if it does not aim to introduce modification in the genome of any descendants. Apart from this, gene therapy is first and foremost regulated by EU law within the framework of clinical trials and medicinal products. EU regulation on advanced therapy medicinal products is of particular relevance for both clinical trials involving gene therapy and for marketing of medicinal products and devices involving gene therapy. More recently, the development of precision medicine (PM) has been high on the scientific and political agenda. PM is a new medical phenomenon, which is promoted in national health strategies across the globe. PM is also known as personalized medicine, individualized medicine, tailored medicine, and P4-­medicine. PM represents a new development in the way that healthcare services provide health promotion, disease prevention, diagnostics, and treatment to patients. PM is information-­intensive. It relies on new advances in gene technologies, such as WGS and GWAS, as well as other ‘omics’ technologies, which together with all sorts of other data from various sources can be linked together and create knowledge regarding risks, susceptibilities, and the efficacy of pharmaceuticals and other treatments. Like gene therapy, the development of PM is currently first and foremost a research endeavour. Its application in the clinic is still limited, and it is mostly used in treatment of certain types of cancers. Most of the focus is on supporting the research that can generate the necessary knowledge to stratify patients and support more individualized risk assessments and provision of tailored diagnostics and treatments. Many countries have adopted national strategies and provided funding to support this kind of research. The 100,000 genomes project30 in the United Kingdome is well known, as is the Obama ­initiative for PM in the United States.31 Several other countries have also launched PM

30  See https://www.genomicsengland.co.uk/the-­100000-­genomes-­project/ 31  See https://obamawhitehouse.archives.gov/the-­press-­office/2015/01/30/fact-­sheet-­president-­obama-­s-­ precision-­medicine-­initiative

Genetics and the Law: Europe   975 initiatives, and the EU is supporting cross-­European cooperation to integrate personalized medicine into the EU’s healthcare systems.32 PM is expected to be beneficial for both the individual and society. Individuals may gain from a more tailored treatment that will allow them to profit from medical interventions without being subjected to treatments with limited effectiveness and damaging side effects. It is also thought to be beneficial for society due to cost savings when treatment can be directed precisely to those who can actually benefit from it. However, there are also concerns. As PM relies on intensive data sourcing—including intensive use of genetic data—it raises issues about the right to know (and not know), rights to privacy and confidentiality, duties and rights in regard to relatives, and risk of genetic discrimination. PM’s effect on equitable access to healthcare services—both nationally and globally—has also been discussed. Consequently, most of the issues discussed in this chapter are also at stake in a PM context. However, the challenges in accommodating traditional patients’ rights principles, such as the right to self-­determination and the right to privacy in regard to genetic testing and screening, are accelerated in a PM context. Characteristic of patients’ rights is the focus on the individual as the rights holder. Genetic testing in general and the development of PM in particular are challenging the individualistic approach in current regulation. As PM relies on intensive data sourcing, including use of genetic information, the boundaries between the individual and various collectives are blurring. Use of genetic information may both reveal information about one’s relatives and rely on genomic analyses of relatives in order to provide a diagnosis to the patient. Consequently, in a PM context, it is difficult to understand the position of the patient in isolation from genetically related family members, and this is a challenge for the perception of both individual autonomy and privacy. The boundary between the individual and the general public or society as a whole is also increasingly blurring, again raising the issue of preserving the notion of individual autonomy and privacy as foundational pillars in patients’ rights law. In addition, well-­established patients’ rights, such as the right to information and the equivalent right not to be informed, may not be suited for this kind of data environment where information is complex and highly uncertain and where it is difficult to predict whether patients would like to be informed or not about various incidental findings. The uncertainty also applies to the consequences that genetic information may have for individuals’ access to insurance, for their position in the labour market, and in several other relations having an impact on the conditions under which individuals can live a dignified life. As already addressed (Section 3), it is well-­established that genetic information may lead to discrimination and stigma, and, with increasing information being available for the individual and for others, it may be challenging to uphold principles of nondiscrimination in a PM environment.

32  See, e.g., European Union, Council conclusions on personalised medicine for patients (2015/C 421/03).

976   Mette Hartlev Non-discrimination may also turn out to be an issue in regard to access to healthcare services. The reference genomes and models developed to stratify patients to the right prevention or treatment will be well-­suited for individuals who resemble the population group from whom data has been collected in terms of, for example, age, ethnicity, and gender—but, at the same time, less-­suited for individuals within another age group or with another ethnicity or gender. For example, the development of national reference genomes deliberately excludes data from ‘non-­national genomes’—so-­called outliers— because they will compromise the validity of the reference genome. Ethnicity and age under 40 years has been shown to be a factor resulting in reduced accuracy of the PM models used in breast cancer. Consequently, PM may turn out not to be available for all population groups, and there is a risk that vulnerable and marginalized groups—those who are already left behind—will be deprived from profiting from PM to the same extent as the average population.

Bibliography Henriette DC Roscam Abbing, ‘The Convention on Human Rights and Biomedicine—An Appraisal of the Council of Europe Convention’ (1998) Eur J Health Law 5, 377. Henriette Roscam Abbing, ‘Medical (Genetic) Examinations for Non-health Purposes. The Relevance of European Legal Standard Setting’, in Gerard Quinn, Aisleen de Paor, and Peter Blanck (eds) Genetic Discrimination. Transatlantic Perspectives for the Case of a EuropeanLevel Response ( Routledge 2015) 177. Roberto Adorno, ‘The Oviedo Convention: A European Legal Framework at the Intersection of Human Rights and Health Law’ (2005) Intl J Intl Biotechnology Law 2, 133. Anne Andermann, Ingeborg Blancquaert, and Véronique Déry, ‘Revisiting Wilson and Jungner in the Genomic Age: A Review of Screening Criteria over the Past 40 Years’ (2008) Bull WHO 86, 317. Ida Helene Asmussen and Katharina Eva Ò Cathaoir, ‘Making Access to a Population of Bodies in the Name of Autonomy’ (2018) 25 Eur J Health Law 25, 555. Peter Blanck and Aisling de Paor, ‘US Legislative and Policy Response: Some Historical Context to GINA’ in Gerard Quinn, Aisleen de Paor, and Peter Blanck (eds) Genetic Discrimination. Transatlantic Perspectives for the Case of a European-Level Response (Routledge 2015) 114. Frederik Zuiderveen Borgesius, Discrimination, Artificial Intelligence and Algorithmic Decision-Making (Council of Europe 2018). Helen Busby, Tamara Hervey, and Alison Mohr, ‘Ethical EU Law? The Influence of the European Group on Ethics in Science and Technology’ (2008) Eur Law Rev 33, 803. Isabelle Chopin and Catharina Germaine, ‘A Comparative Analyses of Non-discrimination Laws in Europe 2017’ (European network of experts in gender quality and non-discrimination 2018), https://publications.europa.eu/en/publication-detail/-/publication/36c9bb78-db0111e7-a506-01aa75ed71a1 Martine Cornel et al., ‘Newborn Screening in Europe. Expert Opinion Document, 7 July 2011’, http://www.isns-neoscreening.org/wp-content/uploads/2016/06/Expert-opinion-documenton-NBS-FINAL.pdf

Genetics and the Law: Europe   977 Council of Europe, ‘Additional Protocol to the CoE Convention on Human Rights and Biomedicine on Biomedical Research’, January 25, 2005, CETS No. 195. Council of Europe, ‘Additional Protocol to the CoE Convention on Human Rights and Biomedicine, concerning Genetic Testing for Health Purposes’, November 27, 2008, CETS No. 203. Council of Europe, ‘Explanatory Report to the Additional Protocol on Genetic Testing for Health Purposes’, November 27, 2008, CETS No. 203. Council of Europe, ‘Background Document on Preimplantation and Prenatal Genetic Testing. Clinical Situation. Legal Situation’, DH-BIO/INF(2015) 6. Council of Europe, ‘Convention for the Protection of Human Rights and Fundamental Freedoms as Amended, Opened for Signatures November 4, 1950’, ETS No. 005. Council of Europe, ‘Convention on the Protection of Human Rights and Dignity of the Human Being with Regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine, Opened for Signatures April 4, 1997’, ETS No. 164. Council of Europa, ‘Explanatory Report to Convention on the Protection of Human Rights and Dignity of the Human Being with Regard to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine, April 4, 1997’, ETS No. 164. Council of Europe, ‘Committee of Ministers, Recommendation No. R (84) 16 Concerning Notification of Work Involving Recombinant Deoxyribonucleic Acid (DNA)’, September 25, 1984. Council of Europe, ‘Committee of Ministers, Recommendation No. R (90)13 on Prenatal Genetic Screening, Prenatal Genetic Diagnosis and Associated Genetic Counselling’, June 21, 1990. Council of Europe, ‘Committee of Ministers, Recommendation No. R (92) 3 on Genetic Testing and Screening for Health Care Purposes’, February 10, 1992. Council of Europe, ‘Committee of Ministers, Recommendation No. R (94) 11 on Screening as a Tool of Preventive Medicine’, October 10, 1994. Council of Europe, ‘Committee of Ministers, Recommendation CM/REC (2010)11 on the Impact of Genetics on the Organisation of Health Care Services and Training of Health Professionals’, November 23, 2010. Council of Europe, ‘Committee of Ministers, Recommendation CM/Rec(2016)8 on the Processing of Health-Related Data for Insurances Purposes—Including Data Resulting from Genetic Tests’, October 26, 2016. Council of Europe, ‘Committee of Ministers, Explanatory Report to Recommendation CM/ Rec(2016)8 on the Processing of Health-Related Data for Insurances Purposes—Including Data Resulting from Genetic Tests, Paragraphs’, October 26, 2016. James Davey, ‘Genetic Discrimination in Insurance. Lessons from Test Achats’, in Gerard Quinn, Aisleen de Paor, and Peter Blanck (eds) Genetic Discrimination. Transatlantic Perspectives for the Case of a European-Level Response ( Routledge 2015) 147. Aisling de Paor, ‘Employment Practices in a New Genomic Era’, in Gerard Quinn, Aisleen de Paor, and Peter Blanck (eds) Genetic Discrimination. Transatlantic Perspectives for the Case of a European-Level Response ( Routledge 2015) 161. European Group on Ethics in Science and New Technologies, ‘Opinion No. 6, Ethical Aspects of Prenatal Diagnosis’, 20/02/1996. European Group on Ethics in Science and New Technologies, ‘Opinion No. 11, Ethical Aspects of Human Tissue Banking’, 21/07/1998.

978   Mette Hartlev European Group on Ethics in Science and New Technologies, ‘Opinion No. 18, ‘Genetic Testing in the Workplace’, 28/07/2003. European Group on Ethics in Science and New Technologies, ‘Opinion No. 25, ‘Ethics of Synthetic Biology’, 17/11/2009. European Group on Ethics in Science and New Technologies, ‘Opinion No. 26, Ethics of Information and Communication Technologies’, 22/02/12. European Group on Ethics in Science and New Technologies, ‘Opinion No. 29, The Ethical Implications of New Health Technologies and Citizen Participation’, 13/10/2015. European Union, ‘Charter of Fundamental Rights of the European Union’, 2010 O.J. (C 83) 389. European Union, ‘Consolidated Version of the Treaty on the Functioning of the European Union’, 2012 O.J. (C 326). European Union, ‘Council Conclusions on Personalised Medicine for Patients’ (2015/C 421/03). European Union, ‘Directive 98/79/EC of the European Parliament and of the Council of 27 October 1998 on in Vitro Diagnostics Medical Services’, 1998 O.J. (L 331) 1. European Unions, ‘Directive 2000/78/EC of 27 November 2000 Establishing a General Framework for Equal Treatment in Employment and Occupation’, 2000 O.J. (L 303) 16. European Union, ‘Directive 2001/20/EC of the European Parliament and of the Council of 4 April 2001 on the Approximation of the Laws, Regulations and Administrative Provisions of the Member States Relating to the Implementation of Good Clinical Practice in the Conduct of Clinical Trials on Medicinal Products for Human Use’, 2001 O.J. (L 121) 34. European Union, ‘Directive 2005/29 EC of the European Parliament and of the Council of 11 May 2005 Concerning Unfair Business-to-Consumer Commercial Practices in the Internal Market and Amending Council Directive 84/450/EEC, Directives 97/7/EC, 98/27/EC and 2002/65/EC’, 2005 O.J. (L 149) 22. European Union, ‘Directive 2007/47/EC of the European Parliament and of the Council of 5 September 2007 Amending Council Directive 90/385/EEC on the Approximation of the Laws of the Member States Relating to Active Implantable Medical Devices, Council Directive 93/42/EEC Concerning Medical Devices and Directive 98/8/EC Concerning the Placing of Biocidal Products on the Market’, 2007 O.J. (L 247) 21. European Union, ‘Regulation 536/2014 of the European Parliament and of the Council of 16 April 2014 on Clinical Trials on Medicinal Products for Human Use, and Repealing Directive 2001/20/EC’, 2014 O.J. (L 158) 1. European Union, ‘Regulation 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation)’, 2016 O.J. (L 119) 1. European Union, ‘Regulation 2017/745 of the European Parliament and of the Council of 5 April 2017 on Medical Devices, Amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and Repealing Council Directives 90/385/EEC and 93/42/EEC’, 2017 O.J. (L 117) 1. European Union, ‘Regulation 2017/746 of the European Parliament and of the Council of 5 April 2017 on in Vitro Diagnostic Medical Devices Repealing Council Directive 98/79/EC and Commission Decision 2010/227/EU’, 2017 O.J. (L 117) 176. European Union, ‘Regulation (EC) 1394/2007 of the European Parliament and of the Council of 13 November 2007 on Advanced Therapy Medicinal Products and Amending Directive 2001/83/EC and Regulation (EC) No 726/2004’, 2007 O.J. (L 324) 121.

Genetics and the Law: Europe   979 Joel Feinberg, ‘The Child’s Right to an Open Future’ in William Aiken and Hugh Lafollette (eds) Whose Child? Children’s Rights, Parental Authority and State Power (Rowman and Littlefield 1980). Delia Ferri, ‘Discrimination: Is It Time for the EU to Take on a New Challenge?’ in Gerard Quinn, Aisleen de Paor, and Peter Blanck (eds) Genetic Discrimination. Transatlantic Perspectives for the Case of a European-Level Response (Routledge 2015) 245. Mark L Flear, Anne-Marie Farrell, Tamara K Hervey, and Thérèse Murphy (eds), European Law and New Health Technologies. Studies in European Law (Oxford University Press 2013). JKM Gevers, EH Hondius, and JH Hubben (eds), Health Law, Human Rights and the Biomedicine Convention: Essays in Honour of Henriette Roscam Abbing (Martinus Nijhoff Publishers 2005). Mette Hartlev, ‘Diversity and Harmonisation. Trends and Challenges in European Health Law’ (2010) Eur J Health Law 17, 37. Janne Rothmar Herrmann, ‘Reproductive Health’ in Brigit Toebes, Mette Hartlev, Aart Hendrix, and Janne Rothmar Herrmann (eds) Health and Human Rights in Europe (Intersentia 2012). Josephine Johnston et al., ‘Sequencing Newborns: A Call for Nuanced Use of Genomic Technologies, The Ethics of Sequencing Newborns: Recommendations and Reflections, Special Report’ (2018) Hastings Center Rep 48(4). Leslie London, ‘What Is a Human Rights-Based Approach to Health—and Does It Matter?’ (2008) Health Human Rights 10, 65. KA McClellan, D Avard, J Simmard, and BM Knoppers, ‘Personalized Medicine and Access to Health Care: Potential for Inequitable Access’ (2013) Eur J Human Genetics, 21, 143. Nuffield Council of Bioethics, ‘Genome Sequencing of Babies. Bioethics Briefing Note’, March 2018, http://nuffieldbioethics.org/wp-content/uploads/Nuffield-Council-on-Bioethics-briefingnote-whole-genome-sequencing-of-babies.pdf Z Obermeyer, P Powers, C Vogeli, and S Mullainathan, ‘Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations’ (2019) Science 366. M Otlowski, S Taylor, and Y Bombard, ‘Genetic Discrimination: International Perspectives’, Ann Rev Genomics Human Genetics (2012) 13, 433. Barbara Prainsack, and Alisa Buyx, Solidarity in Biomedicine and Beyond (Cambridge University Press 2017). Gerard Quinn, Aisling de Paor, and Peter Blanck (eds), Genetic Discrimination. Transatlantic Perspectives for the Case of a European-Level Response (Routledge 2015). Sonia Soni, ‘Genetic Testing Legislation in Western Europe: A Fluctuating Regulatory Target’ (2012) J Comm Genetics 3, 143. A Thorogood, D Dalpé, and BM Knoppers, ‘Return of Individual Genomic Research Results: Are Laws and Policies Keeping in Step?’ (2018) Eur J Human Genetics, 27, 535. United Nations, ‘Convention on the Rights of the Child’, November 20, 1989, Treaty Series, 3 vol. 1577. E Vayena, A Blasimme, and IG Cohen, ‘Machine Learning in Medicine: Addressing Ethical Challenges’ (2018 Nov 6) PLoS Med 15(11), 1–4. Amanda Warren-Jones, ‘Mapping Science and New Health Technologies: In Search for a Definition’ in Mark L Flear, Anne-Maree Farrell, Tamara K Hervey and Thérèse Murphy (eds) European Law and New Health Technologies. Studies in European Law (Oxford University Press 2013) 70.

chapter 46

Gen etics a n d the L aw Conclusion Maxwell Mehlman, Mette Hartlev, and Sonia Suter

As discussed in the preceding chapters, there are a number of similarities in the legal approaches in the United States and Europe to genetic issues, but there are also some areas where the issues are treated quite differently. Both the United States and Europe value informed consent and autonomy in this context. As a result, individuals must give informed consent to genetic testing and can decide to opt out of being informed by their health professionals about genetic sequencing results, including results that are clinically relevant. Even though such results may be of potential importance to family members, both US and European legislation and courts are reluctant to impose a duty on health professionals to inform a patient’s family members about their genetic risks without the patient’s consent. Such disclosures can take place only in limited circumstances, if at all. Parents in the United States have the right to test their children’s DNA, although professionals recommend delaying predictive testing for adult-­onset conditions. Even though European regulation respects parents’ right to make healthcare decisions on behalf of their children, legislation in some European countries explicitly restricts genetic testing during childhood except to allow for preventive and therapeutic measures to be taken during childhood or to spare the child from highly invasive regular examinations. This is aligned with recommendations in European soft law regulation. Consequently, European regulation seems to be more aligned with the principles enshrined in the UN Convention on the Rights of the Child than does US regulation. All states in the United States conduct newborn screening using DNA samples, and most do so without parental consent. In Europe, most countries have a newborn screening program, and most of these require parental consent. In both the United States and Europe, professionals are debating whether newborn screening should include whole-­ genome sequencing. In addition, questions have arisen about the propriety of states

982   Maxwell Mehlman, Mette Hartlev, and Sonia Suter conducting research on newborn screening samples. Some states in the United States and some European countries allow parents to block research by ordering samples to be destroyed or by declining to give consent. Others allow research without consent from parents or the grown child. In the United States, physicians have a legal obligation to offer prenatal testing if it is medically indicated. They can be sued in most states for “wrongful birth” if they do not do so. An additional question is whether parents should have the right to terminate a pregnancy based on the results of genetic testing or screening, with some states seeking to limit abortions on the basis of genetic variatnts. In Europe, there is no common position to the ethical acceptability or medical obligation to offer prenatal testing and preimplantation diagnostics of embryos. Consequently, this is an area of ethical sensitivity where European countries’ national laws differ significantly. However, national discretion is not unfettered and must be exercised within the framework of the Council of Europe and EU regulation, which specifies some limits for and requires consistency in national law. The US Food and Drug Administration (FDA) regulates genetic tests as medical devices, albeit somewhat haphazardly depending on the type of test. The FDA also regulates the growing “direct-­to-­consumer” (DTC) online genetic testing industry, but legal questions arise about DTC companies’ use of genetic information for research or sale. The federal government also regulates clinical laboratories that conduct genetic testing. In Europe, genetic tests are considered in vitro medical devices and are subject to both EU and national regulation. The legal landscape regarding DTC online genetic testing is quite fragmented in Europe, with significant differences among the European countries. Genetic discrimination has been the subject of a number of laws in the United States. These laws protect against employment and insurance discrimination on the basis of genetic information, but certain types of insurance, such as long-­term care, are not covered in many states. Federal and state genetic privacy laws also provide some protection, but again with limits, such as a lack of protection for information held by DTC genetic testing companies. Additional concerns arise from potential government use of military members’ genetic information. Compared to the United States, there is no comprehensive regulation of genetic discrimination at the European level. However, in the area of employment and insurance, both the Council of Europe and EU regulation provide a legal framework for national legislation. At the national level, a few European countries have included genetic discrimination as part of general discrimination laws, whereas others have special regulation in specific areas such as employment and insurance. Human gene therapy and gene therapy research in the United States is regulated by the FDA but only when conducted in corporate or university settings. Currently, the FDA is not permitted by federal law to approve research on germline modifications, and the federal government is not allowed to fund it. The FDA has begun requiring drug manufacturers to include information about pharmacogenetic drug–gene interactions in drug product labeling and to recommend genetic testing before prescribing in certain instances. In Europe, gene therapy is first and foremost regulated by EU law within the

Genetics and the Law: Conclusion   983 framework of clinical trials, approval and marketing of medicinal products, and the ­special regulation on advanced therapy medicinal products. As in the United States, germline modification is prohibited in European regulation. As genetic technologies continue to evolve, the United States and Europe will likely continue to face similar ethical and legal issues regarding the regulation of those technologies. As we have seen, however, their approaches will likely overlap in some ways and diverge in others, reflecting differences in the legal structures and cultures of each jurisdiction.

Pa rt V

H E A LT HC A R E AT T H E BE GI N N I NG A N D   E N D OF L I F E

REPRODUCTIVE RIGHTS

chapter 47

I n troduction to the R ight to Procr e ate a n d Assisted R eproducti v e Tech nol ogies I. Glenn Cohen and Emily Jackson

The right to procreate is one of the most important personal rights, and it has received widespread recognition under the law. Yet its protections can be tenuous for many people, particularly those who are poor and those who need assisted reproductive technologies to have children. In our two chapters (Cohen; Jackson, this volume), we will examine the meaning and extent of rights to procreate, with particular attention to the use of assisted reproductive technologies (ARTs) since that is where most of the legal controversies lie. The right to procreate is at its strongest when invoked to safeguard an individual’s natural ability to reproduce. Thus, for example, involuntary sterilization is forbidden by both the European Convention of Human Rights (ECHR) and the US Constitution. Under the ECHR, this follows from Article 8, which safeguards a right to private and family law, and Article 12, which guarantees the right to marry and found a family. When the US Supreme Court invoked the US Constitution to reject involuntary sterilization in a 1942 decision, Skinner v. Oklahoma, the Court observed that the right to procreate is “one of the basic civil rights of man.” But the promises of the ECHR and the US Constitution tend to protect “negative” rather than “positive” rights. In the United States, this means that while the state cannot interfere with procreation, it need not offer assistance to make it feasible. Accordingly, the government does not have to provide financial support for the indigent so they can afford the costs of parenting, though the state does so to some extent. Poverty, then, ­represents a significant barrier to procreation.

990   I. Glenn Cohen and Emily Jackson In Europe, the right to procreate has more of a positive element. Although the ECHR mainly protects negative liberty, there have been cases involving prisoners who would otherwise be prevented from reproducing where the state has been under a positive obligation to provide access to services. The European Social Charter prohibits dismissal on the grounds of pregnancy and offers certain protections to pregnant women in the workplace.1 Even so, poverty can prevent poor persons from exercising their rights to procreate. Individuals seeking assisted reproductive services also can face serious barriers to having children. The absence of a positive right to assistance together with the high costs of most reproductive technologies can make procreation too expensive for many persons. In addition, courts view assisted reproduction as sufficiently different from unassisted reproduction such that governments have significant authority to prohibit access to reproductive technologies. For example, commercial surrogacy raises concerns about exploitation of women and commodification of children that are not present with unassisted reproduction. Thus, even though prohibiting paid surrogacy entails an interference with procreation rather than a denial of assistance, it is not clear that a surrogacy ban interferes with people’s rights to have children. The right to procreate simply might not include a right to become a parent through commercial surrogacy. Our consideration of rights to procreate via technological assistance begins in 1978. The birth of Louise Brown in Oldham, England, that year represented one of the most extraordinary scientific achievements of the 20th century, for which Bob Edwards won the Nobel Prize for Medicine in 2010. Now millions of children have been born worldwide as a result of in vitro fertilisation (IVF), and it is a safe and routine clinical procedure. Although much of the early skepticism about the wisdom or appropriateness of “interfering” in the process of conception has now disappeared, assisted conception continues to raise some complex and contentious questions for law: Should gamete donors be anonymous, for example, and should they be paid? Should commercial surrogacy be allowed, and who are the parents of a child conceived through a surrogacy arrangement? If one simply reads the mass of case law, statutes, and regulations in this area one will find . . . a mess. Of all the areas of health law and bioethics, this may be the one with the largest disagreements across countries. On whether (and under what conditions) to permit surrogacy, anonymous gamete donation, pre-­embryo disposition agreements, and posthumous reproduction, to name but a few examples, there are sharp divergences among, and even within, countries (in the United States, much of this is committed to state not federal law). Perhaps this is not surprising given the centrality of ideas of mother- and fatherhood, pregnancy, and genetic relatedness to religious and cultural self-­definitions across the world. As we discuss, this divergence has led many to travel abroad to circumvent domestic rules and achieve pregnancy, so-­called medical travel or medical tourism (though many object to that latter term), which itself has generated tangled law. 1  Similar protections are provided in the United States by statute.

Introduction to the Right to Procreate   991 But indeed it gets still more complicated. Beyond the level of disagreements as to the substance of what the rules should be in this area across and within jurisdictions, we also find sharp disagreements as to the level/mechanism of regulation. Are there constitutional or treaty-­based rights to procreate that limit the ability of various governments to implement their preferred vision of reproductive technology? To what extent should top-­down regulation be used as opposed to regulating through tort law or even professional self-­regulation? When should a government be allowed to regulate the reproductive activities of its citizens beyond territorial boundaries—that is, can a government act to prevent its policies being circumvented through travel to a more permissive jurisdiction? And many more. Out of this chaos, our chapters attempt to bring some order while recognizing that, as Aristotle put it in the Nichomachean Ethics, “[o]ur discussion will be adequate if it has [only] as much clarity as the subject-­matter allows.”2 The task is even more challenging when looking at so many jurisdictions, but our chapters do highlight the following areas of commonality: Notwithstanding some treaty or constitutional limits, there is a large amount of discretion given to individual governments in setting their policies. In the European Union and wider Council of Europe this has been the result of giving Member States a wide “margin of appreciation” as to their policies in this area. In the United States, this has largely been the result of significant uncertainty as to what the federal Constitution means for reproductive technologies. Some countries have instituted rigorous regulatory regimes, whereas others have few legal restrictions and have left it to professional bodies to issue their own good practice guidance. In some places, an absence of regulation which in practice resulted in an “anything goes” approach to assisted conception has been replaced, overnight, by restrictive laws. Legal rules which restrict or confine access to treatment are also commonly subject to judicial review, often on the grounds that they interfere with people’s human rights. The regulation of assisted conception is not static, therefore, but may reflect much broader political, cultural, and social changes. Relatedly, many of the existing judicial encounters with reproductive technologies have focused on negative not positive liberty rights. Without meaning to treat the distinction as sacrosanct, the vast majority of reported cases concern a right to be free from governmental interference with one’s right to be a gestational, legal, or genetic parent or some combination of them. Given the fact that there can be no guarantees that undergoing IVF will lead to a successful pregnancy and birth, it would be difficult to assert that anyone had a right to a child. More plausible is the idea of a right to access assisted conception services. For fertile, heterosexual couples, interference with their ability to conceive a child would represent a wholly extraordinary invasion of their privacy and bodily autonomy. If someone needs medical assistance in order to conceive, placing restrictions on their reproductive choices is easier, but that does not necessarily mean that state interference is warranted 2  Aristotle, Ethics 5 (John Warrington trans. & ed., J. M. Dent & Sons Limited 1963).

992   I. Glenn Cohen and Emily Jackson or justifiable. In some European countries, lesbian women, gay men, and single people are not able to access assisted conception services on the same terms as heterosexual couples, while, in others, this is regarded as unlawful discrimination. Governments, by contrast, have largely been given free rein in deciding whether to fund reproductive technology usage—both directly in terms of national health systems and/or public health insurance and indirectly in terms of whether to force coverage by private insurers. The result has been, generally speaking, fairly little public funding for reproductive technology needs and, in the United States, very few private insurers covering the more expensive procedures. Even in countries where there is some state funding or insurance coverage for treatment, access tends to be restricted and exceptional. Most patients are therefore “self-­pay,” and many of the rules are set by sellers in the market, either directly (e.g., choosing which types of patients to serve) or through industry self-­regulation, with some role for tort law to set outer boundaries. Both of our chapters highlight whether the importance of soft law in this area, rather than more direct regulation, is desirable and how, in this respect, reproductive technology services look very different from other healthcare services. In the UK, for example, almost all medical treatment is free at the point of use, whereas the majority of IVF cycles take place in the private sector and are unaffordable for most people. Finally, as our chapters detail, both the United States and the European Union have struggled with how to deal with cross-­border travel for reproductive technology services. This has taken the form of not only patients traveling for price reasons—surrogacy, in particular, has been prohibitively expensive for many patients, causing them to travel to less developed countries—but also circumventing domestic prohibitions on particular technologies or access for particular groups. Should we view these patients as reproductive outlaws or refugees, and how should (and can) the law respond?

chapter 48

The R ight to Procr eate a n d Assisted R eproducti v e Tech nol ogies i n Eu rope Emily Jackson

1  Why Regulate? Having children is of central importance to many people’s lives, and the inability to do so naturally can cause huge distress.1 If fertility treatments are available and patients want them, what is the appropriate role for regulation? Is it to try to expand reproductive choices or just to make sure that treatments reach a threshold level of safety and efficacy? Should regulation aim to protect the interests of children born as a result or determine what sort of people should be allowed to become parents through artificial means? There is undoubtedly an important role for law in regulating the provision of assisted conception services by, for example, ensuring that patients receive safe and effective treatment from appropriately qualified providers and that consent to treatment is properly informed. Where donated gametes are used in treatment, there is a need for clarity about the identity of the child’s legal parents. Using the law to make moral choices about who deserves to become a parent is another matter, however. There is now considerable evidence that children born as a

1  Rebecca CH Brown et al, ‘Reframing the Debate Around State Responses to Infertility: Considering the Harms of Subfertility and Involuntary Childlessness’ (2016) 9 Pub. Health Ethics 290–300.

994   Emily Jackson result of assisted conception are doing at least as well as children conceived naturally.2 This should not be surprising given that people who have children as a result of fertility treatment have thought about becoming parents and will often also be comparatively wealthy, whereas the pool of people who reproduce naturally includes those who may find parenting challenging, such as children and drug addicts. More importantly, history teaches us that when the state makes judgments about eligibility for parenting, it often reflects invidious biases against disfavoured groups, as with the eugenics movement and its forced sterilizations a century ago and distinctions on the basis of sexual orientation more recently.3 Even restrictions which might be justified on child welfare grounds—such as the removal of donor anonymity or an upper age limit for female patients—are problematic. They typically embody normative prescriptions about the right and wrong way to bring up children that are rooted more in tradition than science.4 While in many policy areas the government can prefer one political philosophy over another, there is much less room for that kind of authority when it comes to something as integral to a satisfying life as reproductive freedom.5

2  Regulation Within Europe There is no consensus within Europe on how assisted reproductive technologies should be regulated. While there has been a trend within the European Union (EU) towards the harmonisation of laws, for example in relation to trade, consumer protection, and workers’ rights, states have always had a wide ‘margin of appreciation’ when it comes to sensitive questions like the moral status of the human embryo or who should have access to reproductive assistance. Some countries have prohibited specific procedures, such as embryo donation (Sweden), egg donation (Germany and Turkey), and preimplantation genetic diagnosis (PGD) (until recently, Germany), while others, like Spain and the Czech Republic have become major destinations for cross-­border reproductive care. This chapter will not attempt to provide a comprehensive overview of laws within Europe. Rather it will explore some of the themes identified in I. Glenn Cohen’s chapter through a European and perhaps a particularly UK-­orientated lens. It will begin by con2  See further Susan Golombok, Modern Families: Parents and Children in New Family Forms (Cambridge University Press 2015). 3  S Golombok, A Brewaeys, MT Giavazzi, D Guerra, F MacCallum and J Rust, ‘The European Study of Assisted Reproduction Families: The Transition to Adolescence’ (2002) 17 Human Reproduction 830–840; C Murray and S Golombok, ‘Solo Mothers and Their Donor Insemination Infants: Follow-­up at Age 2 Years’ (2005) 20 Human Reproduction 1655–1660; E Ilioi, L Blake, G Roman and V Jadva, ‘A Longitudinal Study of Families Formed Through Reproductive Donation: Parent-­Adolescent Relationships and Adolescent Adjustment at Age 14’ (2017) 53 Developmental Psychology 1966–1977. 4  G Pennings, ‘Disclosure of Donor Conception, Age of Disclosure and the Well-­Being of Donor Offspring’ (2017) 32 Human Reproduction 969–973. 5  Ronald Dworkin, Life’s Dominion: An Argument About Abortion and Euthanasia (Harper Collins) 200–202.

The Right to Procreate in Europe   995 sidering some of the reasons for the different approaches to the regulation of reproduction within Europe. Second, it will look at the role of rights and rights-­based legal claims in relation to assisted reproductive technologies, concluding that they have played a relatively minor role. There are limits on access to assisted conception services, and, in most countries, there are restrictions on funding, but these have seldom been challenged ­successfully on human rights grounds. Third, compared with the United States, professional self-­regulation and tort law are background considerations in many European countries, with detailed regulations more commonly found in statute and in the ‘soft law’ guidance and codes issued by regulators. Fourth, increasing numbers of people are engaging in cross-­border reproductive care, not only within Europe but much further afield as well. For ‘low-­tech’ procedures, such as sperm donation, it is increasingly common for would-­be parents to seek informal providers online and thus bypass whatever legal restrictions exist in regulated clinics. Finally, although there is state funding for assisted conception services in many European countries, access tends to be limited, and hence there is a market in fertility treatment to a much greater extent than there is for most other medical procedures. In the UK, for example, which has the most comprehensive single-­payer healthcare system in the world, most fertility treatment is provided in the private sector.

3  Variation within Europe The progressive harmonisation of laws within the EU has had relatively little impact on the provision of fertility treatment. The EU Tissues and Cells Directive (EUTCD) does set some EU-­wide quality and safety standards and traceability requirements,6 and it also attempts to control payments to donors through Article 12(1). Member States shall endeavour to ensure voluntary and unpaid donations of tissues and cells. Donors may receive compensation, which is strictly limited to making good the expenses and inconveniences related to the donation. In that case, Member States define the conditions under which compensation may be granted.

Within Europe, this has been interpreted in different ways. Some countries, like France, compensate donors only for their actual expenses, while others, like the UK, Portugal, Spain, and Denmark, also compensate donors for the ‘inconvenience’ of donation. These ‘inconvenience’ payments tend to be fairly modest (less than £50 per clinic visit for sperm donors and in the region of £600–900 per cycle of donation for egg donors).

6 2004/23/EC.

996   Emily Jackson On more fundamental questions, such as the lawfulness of gamete donation or PGD or restrictions on access to treatment, states have a wide margin of appreciation, and the rules therefore differ markedly. Explaining these differences is complicated, however, because they often reflect variations in the political, historical, religious, and cultural backdrops to regulation. For example, the history of eugenics in Europe has cast a long shadow, and PGD was prohibited in Germany until 2011, when access was permitted in strictly limited circumstances. The Catholic Church, which opposes in vitro fertilisation (IVF) on principle because of the inevitability of the destruction of embryos, influenced the extraordinary volte face on regulation in Italy in 2004. Before the law of 2004, difficulties in reaching agreement on how IVF should be regulated meant that there was no regulation and Italy had become, by default, the preferred destination for ‘rogue’ providers, such as Severino Antinori, who became infamous for treating elderly women and claiming to have produced cloned babies.7 The Italian law of 2004 reflected Catholicism’s concern about the destruction of embryos and provided that a maximum of three embryos could be created in one cycle and—contrary to good medical practice—that all of them would have to be transferred to the woman’s uterus in a single treatment cycle. PGD and donor insemination were also banned. A series of legal challenges have dismantled much of this law, which was found to be constitutionally illegitimate and to be in breach of couples’ human rights, but some restrictions remain. In particular, access to assisted conception is confined to couples of opposite sex, leading many Italian lesbians and single women to seek treatment in other European countries. France has also restricted access to opposite sex couples, although, following a ruling of the Conseil d’État in 2018, this is likely to change in the near future. In Spain, Catholicism was strongly associated with the Franco dictatorship. Its influence on social policy is now minimal, and Spain has liberal laws on abortion and assisted conception. Indeed, Spain is a thriving destination for cross-­border reproductive treatment. In contrast, in Poland, the Catholic Church was associated with the Solidarity movement’s resistance to the Soviet regime, and it continues to have considerable political influence. IVF is available in Poland, but only to heterosexual couples.8 Even countries with permissive regulatory regimes have enacted some restrictions which could be said to be grounded in morality. In the UK, for example, it is a criminal offence to carry out PGD for the purposes of nonmedical sex selection as opposed to testing for genetic diseases. Until the rules for PGD appeared on the face of the statute (following amending legislation in 2008), the ban on sex-­selection for ‘social’ reasons was a policy decision of the UK regulator, the Human Fertilisation and Embryology Authority (HFEA). In deciding to prohibit sex selection, the HFEA was influenced by

7  ‘Top Italian fertility Doctor Severino Antinori Arrested’, BBC News 14 May 2016. 8  Elżbieta Korolczuk, ‘“The Purest Citizens” and “IVF Children”: Reproductive Citizenship in Contemporary Poland’ (2016) 3 Reproductive Biomedicine & Society Online 126–133.

The Right to Procreate in Europe   997 the strength of public opposition to sex selection.9 This ban was translated into law in 2008, along with a largely symbolic ban on the positive selection of embryos known to be affected by a genetic condition.10 Within Europe, there is no consensus on whether gamete donors should be anonymous or identifiable. In some countries, donor anonymity has been mandated by law (e.g., France, Belgium, Spain, Portugal, Greece, Denmark, Bulgaria, Czech Republic), while others have prohibited anonymous donation (e.g., Sweden, UK, Netherlands). Although there are judgments of the European Court of Human Rights, most involving adoption, which suggest that the right to know your identity is part of the right to private and family life protected by the European Convention on Human Rights, this has not eliminated anonymous gamete donation within Europe. Other common restrictions on access to IVF are intended to promote patient safety rather than public morality. For example, Sweden was the world leader in the promotion of elective single embryo transfer (eSET) in order to minimise the risks to women and babies from multiple births. Now many other European countries, including Belgium and the UK, restrict the number of embryos that can be replaced in a single treatment cycle, usually by requiring clinics to have an elective single-­embryo transfer policy, which provides that only one embryo should be transferred in good-­prognosis patients (e.g., women under the age of 35 or 37). Most countries in Europe have some public funding of healthcare, but in no country is IVF available free of charge to anyone who wants or even needs it. All publicly funded healthcare systems operate with limited resources, and ‘tragic choices’ are inevitable.11 If some cancer drugs cannot be funded because they are insufficiently cost-­effective,12 it is unsurprising that unlimited, universal access to fertility services would be unaffordable. Even in countries where public funding for assisted conception is available, it is rationed, generally using ‘clinical effectiveness’ criteria such as age limits for female patients, supplemented by what might be described as ‘fairness-­based’ considerations (e.g., whether a couple seeking publicly funded PGD already has an unaffected child). It could, however, be argued that the common assumption that fertility treatment should be a lower priority for public funding than most other healthcare services underestimates the serious negative impact unwanted childlessness can have on a person’s

9  Sex Selection: Options for Regulation (HFEA, 2003) para 147. See further Sara Fovargue and Rebecca Bennett, ‘What Role Should Public Opinion Play in Ethico-­Legal Decision Making? The Example of Selecting Sex for Non-­medical Reasons’ (2016) 24 Med L Rev 34–58. 10  This provision prevents parents from ‘preferring’ an embryo known to be affected by a genetic condition. It caused considerable disquiet among the deaf community who argued that it ‘sent a message’ that deaf children should be regarded as less valuable than hearing children. See further Gerard Porter and Malcolm K. Smith, ‘Preventing the Selection of “Deaf Embryos” Under the Human Fertilisation and Embryology Act 2008: Problematizing Disability?’ (2013) 32 New Genetics and Society 171–189. 11  Guido Calabresi and Philip Bobbitt, Tragic Choices (WHW Norton 1978). 12  Sarah Boseley, ‘Breast Cancer Drug Rejected for NHS Use on Cost-­Benefit Grounds’, The Guardian 29 December 2016.

998   Emily Jackson wellbeing.13 If a lack of funding prompts people to seek treatment in less well-­regulated jurisdictions, it may also be a false economy since the costs to the domestic healthcare system from the side effects of unsafe treatment, including multiple births, can be enormous.

4  Rights to Procreate The European Convention on Human Rights (ECHR) contains two free-­standing rights that might be invoked to claim that one’s right to reproduce had been violated. First, under Article 8, the right to private and family law is protected unless interference is ‘necessary in a democratic society’, to protect important competing interests, such as ‘health or morals’, or ‘the rights and freedoms of others’. Second, Article 12 protects the right to marry and found a family, and Article 14 provides a derivative right not to be unlawfully discriminated against in the exercise of any of these Convention rights. It is well established that Articles 8 and 12 protect a negative rather than a positive right to reproduce. That is, someone who wishes to reproduce must not have their freedom to do so interfered with by others. Sterilising someone against their wishes, for example, would be a clear violation of Articles 8 and 12. Whether caps on government benefits for larger families might violate Article 8, or might be justifiable under Article 8(2) as being ‘necessary in a democratic society’ in the interests of ‘the economic well-­ being of the country’ is open to question.14 It is, however, clear that Articles 8 and 12 do not give people the positive right to whatever medical assistance they may need in order to have a baby. There are two further reasons why the ECHR has not proved to be overwhelmingly useful to individuals or couples seeking access to assisted conception services. First, states have a wide ‘margin of appreciation’ in relation to human reproduction, and there is no consensus within Europe on how assisted conception should be regulated. Second, and more practically, there are good reasons why litigation is seldom the best option for couples or individuals who are denied access to assisted conception services. If they can afford to mount a legal challenge to a restriction upon access, they will be able to afford to travel to another clinic or another jurisdiction where they can avoid the restriction in question immediately. This is important because fertility patients are almost always in a hurry to start treatment. Women’s age-­related fertility decline means that delaying treatment is likely to 13  Rebecca CH Brown et al, ‘Reframing the Debate Around State Responses to Infertility: Considering the Harms of Subfertility and Involuntary Childlessness’ (2016) 9 Public Health Ethics 290–300. 14  In the case of (R SC and others) v Secretary of State for Work and Pensions [2019] EWCA Civ 615 the Court of Appeal confirmed that a welfare cap for families with more than two children was not a breach of Article 8.

The Right to Procreate in Europe   999 reduce its chance of success. For heterosexual couples, infertility usually comes as a ­surprise. After years of trying to prevent conception, people tend to assume that conception will happen soon after making the decision to start a family. As a result, by the time a couple knows that they will need medical assistance in order to conceive, their plans will already have been delayed by at least a year, often by several years. For obvious reasons, waiting another couple of years to pursue a legal challenge is a much less attractive option than starting treatment somewhere else without delay. Cases in which people have sought to challenge restrictions on access to assisted conception have tended to involve would-­be parents who do not have a simple choice between litigation and travelling to a more accommodating clinic or jurisdiction. For example, there have been cases in which prisoners have sought to challenge a refusal to grant them access to artificial insemination facilities. Prisoners retain their human rights on incarceration, so although the prison authorities are entitled to have a policy that does not facilitate artificial insemination, they are also not entitled to prevent people from having any chance to reproduce. Where incarceration will merely delay a couple’s ability to start a family, a restriction on access may be compatible with their human rights;15 where it will prevent it altogether, this has been treated as an unjustifiable interference.16 In the UK, it is a criminal offence to use gametes or embryos created from a person’s gametes without the gamete provider’s written consent. This has led to two other types of legal claim in which a person has argued that they should be able to use gametes or embryos in the absence of the gamete provider’s written consent. The first sort of claim has arisen after the gamete provider’s death, when he or she died without having given written consent to the posthumous use of his or her gametes. In the absence of written consent, it would be unlawful to use the gametes in the UK and impossible to export them without the consent of the HFEA. In three cases, the courts have found that the HFEA has a broad discretion to permit export where there is compelling evidence that this is what the deceased person would have wanted.17 In a second type of claim, a woman sought to challenge the right of veto created by the need for both parties’ consent to the use of an embryo. In Evans v United Kingdom,18 Natallie Evans and her ex-­partner had six frozen embryos in storage, created before she underwent treatment for ovarian cancer. He subsequently withdrew his consent to the embryos’ continued storage and use, as a result of which they would have to be destroyed. Ms Evans sought a declaration that the consent provisions of the Human Fertilisation and Embryology Act 1990, which gave him this right to prevent her from using their embryos, were incompatible with her Article 8 rights. Her claim failed 15  R v Secretary of State for the Home Department, ex parte Mellor [2001] EWCA Civ 472. 16  Dickson v UK (App no 44362/04) (2007). 17  R v Human Fertilisation and Embryology Authority, ex parte Blood [1997] 2 WLR 806 (CA); L v Human Fertilisation and Embryology Authority [2008] EWHC 2149 (Fam) and R (on the application of M) v Human Fertilisation and Embryology Authority [2016] EWCA Civ 611. 18  App no 6339/05 (2007).

1000   Emily Jackson because, although her Article 8 rights were undoubtedly engaged, so were those of her ex-­partner. In those circumstances, and given ‘the lack of any European consensus on this point’, the Grand Chamber of the European Court of Human Rights (ECtHR) did ‘not consider that the applicant’s right to respect for the decision to become a parent in the genetic sense should be accorded greater weight than J’s right to respect for his decision not to have a genetically-­related child with her’.19 In practice, such a claim is now much less likely because a relatively new egg-­freezing technique, known as vitrification, means that a woman facing premature infertility would be likely to freeze eggs, rather than embryos, and these would be available for her future use without the need for anyone else’s consent. A different sort of human rights claim takes place after a child has been born, perhaps as a result of a surrogacy arrangement, and the intended parents wish to challenge a state’s failure to recognise them as their child’s legal parents. In Mennesson v France,20 for example, the ECtHR held that France’s refusal to recognize the relationship between a father and his biological child, born as a result of a surrogacy arrangement, ‘overstepped the permissible limits of its margin of appreciation’. In contrast, in Paradiso and Campanelli v Italy,21 there was no biological relationship between the intended parents and the child born as a result of a surrogacy arrangement in Russia, leading to concerns about the provision of fraudulent information on a birth certificate. The Italian authorities had decided to remove the child from the intended parents, and the ECtHR held that ‘the Italian courts, having assessed that the child would not suffer grave or irreparable harm from the separation, struck a fair balance between the different interests at stake, while remaining within the wide margin of appreciation available to them in the present case’.22 In practice, then, there are relatively few cases in which would-­be parents have asserted that they have a right to fertility treatment. That does not mean that human rights considerations play no part in regulation, however. For example, in the UK, the HFEA’s Code of Practice instructs clinics that they must not discriminate against patients on the basis of their protected characteristics.23 As a result, there would be no need for a lesbian couple in the UK to bring a human rights claim in order to challenge a clinic’s refusal to treat them. Such a restriction would be a breach of equality legislation and of the HFEA’s Code of Practice. Any clinic which refused to treat would-­be patients on the grounds of their sexual orientation or marital status would be likely to be subject to swift regulatory action by the HFEA, which has the power to suspend or revoke the clinic’s licence.

19 Id. 20  App no 65192/11 (2014). 21  App no 25358/12 (2017). 22 Id. 23  HFEA, 9th Code of Practice, para 29.4.

The Right to Procreate in Europe   1001

5  Forms of Regulation In most European countries, there is some state regulation of fertility treatment, which coexists with good practice guidance from professional organisations. The most comprehensive regime within Europe is also one of the oldest. In the UK, the Human Fertilisation and Embryology Act 1990 set up the HFEA, and, without a licence from the HFEA, it is a criminal offence to create embryos and to store and use embryos and gametes. In addition to its licensing regime, the HFEA also has a statutory duty to collect and publish information, to maintain a Code of Practice, and to advise ministers on new developments. Some of the rules that the HFEA administers are ‘on the face of the legislation’. There is, for example, a statutory duty upon clinics to ensure that, before any woman is provided with treatment services, account is taken of the welfare of any child who may be born.24 As mentioned earlier, the statute also contains restrictions on the circumstances in which it is lawful to carry out PGD; namely, that there must be a significant risk that the child would suffer from a serious genetic condition. Other rules, such as the storage time limit for gametes and embryos, are contained in primary legislation but supplemented by regulations, which can, in theory, be amended more quickly than the statute itself. For example, the Act provides for a 10-­year storage time limit for embryos and gametes,25 but Regulations give patients who are, or who are likely to become prematurely infertile, the option to extend this for up to 55 years.26 While the storage time limit and the option for extension is largely unproblematic for sperm and embryos, it creates difficulties for what is known as ‘social egg freezing’. If a woman freezes her eggs at the age of 25, she is unlikely to be prematurely infertile at the age of 35, and hence her eggs would have to be disposed of before she is likely to want to use them in treatment. The poor fit between the storage time limit and social egg freezing is not deliberate: social egg freezers were simply not within the contemplation of the drafters of the 2009 Regulations. It is therefore to be hoped that it will be remedied by an amendment to the Regulations to allow for extensions to the 10-­year limit for women who freeze their eggs in order to preserve their fertility.27 Less formal still are the HFEA’s Code of Practice and the Directions it issues to clinics, which can be changed unilaterally by the HFEA. These have been used, for example, to bring down the multiple pregnancy rate by requiring all clinics to have a ‘multiple birth reduction strategy’ and to insist that clinics should not exceed the HFEA’s maximum

24  Section 13(5). 25  Section 14. 26  The Human Fertilisation and Embryology (Statutory Storage Period for Embryos and Gametes) Regulations 2009. 27  Emily Jackson, ‘“Social” Egg Freezing and the UK’s Statutory Storage Time Limits’ (2016) 42 J Med Ethics 738–741.

1002   Emily Jackson multiple birth rate (which has come down from 24 to 10 per cent). The HFEA also sets and can change the compensation rates for gamete donors.28

6  Cross-­B order and Internet-­A ssisted Reproductive Care Throughout the world, it is increasingly common for people to travel in order to avoid domestic restrictions on access to reproductive technologies. Within and outside Europe, this has led to several well-­established routes through which restrictions are bypassed: French single women and lesbian couples receive treatment with donated sperm in Belgium; women or couples living in countries where anonymous donation is banned can travel to Denmark for treatment with anonymous donor sperm. Legal restrictions are not the only reason why people seek treatment overseas. People also travel to avoid waiting lists for gamete donors or to access gamete donors from particular ethnic backgrounds. Cost is another important factor, with residents of countries with permissive regimes, such as the UK, travelling to receive cheaper care in Eastern Europe or in South East Asia. Until the rules changed to prohibit foreigners from entering into surrogacy arrangements, India and Thailand were major providers of relatively cheap commercial surrogacy services. Now alternative destinations have emerged, such as Laos and Cambodia, as well as increasingly complex cross-­border surrogacy arrangements in which women from unregulated jurisdictions, like Kenya, are flown to India for embryo transfer and the first 24 weeks of pregnancy before returning to Kenya in time for the births.29 Even people travelling from the same country for the same treatment may do so for different reasons: for example, French single women travel to Belgium for gamete donation because they are ineligible for treatment at home. In contrast, French heterosexual couples are entitled to receive fully reimbursed gamete donation at home, but they commonly travel to Greece or Spain to avoid long waiting lists.30 It is also important to acknowledge that cross-­border reproductive care is not necessarily the desperate last resort of people who have been excluded from services in their home country through statutory or funding restrictions or, de facto, as a result of long waiting times. Rather, it is often a positive choice, with patients believing that success

28  Currently, for sperm donors, a fixed sum of £35 per clinic visit to cover all of their expenses, and for egg donors, a fixed sum of £750 per cycle of donation, again to include all expenses. 29  Sharmila Rudrappa, ‘India Outlawed Commercial Surrogacy – Clinics Are Finding Loopholes’ (2017) The Conversation 24. 30  Virginie Rozée Gomez and Elise de La Rochebrochard, ‘Cross-­Border Reproductive Care Among French Patients: Experiences in Greece, Spain and Belgium’ (2013) 28 Human Reproduction 3103–3110.

The Right to Procreate in Europe   1003 rates will be better or that care will be more personalised. If treatment at home has failed repeatedly, patients may go abroad to ‘try something different’.31 For some patients, seeking treatment overseas may be empowering. In Van Hoof et al’s study of French lesbian and single women seeking treatment with donor sperm in Belgium, most interviewees reported how much they appreciated being treated ‘as a standard case’: as one woman put it, ‘we’re treated like anybody else, not like we’re sick’.32 One of Jenny Gunnarsson Payne’s respondents, Helena, who had travelled from Sweden to Latvia, described ‘the position of being simultaneously a patient and a health care consumer as “empowering”, viewing both the service mindedness and the efficacy of the treatment as a direct result of her ability to pay for the treatment’.33 Helena believed that the Riga clinic in which she had received IVF with donated eggs was ‘more professional’ than the Swedish publicly funded system because it was dealing with ‘paying customers’. Where the restrictions on access to treatment at home are grounded in discrimination, travelling overseas to avoid them can be experienced as a positive and assertive act of civil disobedience. Giulia Zanini interviewed Italian lesbian women who had avoided the ban on their treatment in Italy by seeking treatment overseas.34 Having felt abandoned and discriminated against at home, travelling to receive sperm donation, and subsequently giving birth to Italian babies, was their private act of dissent and re­sist­ ance. Women reported that purchasing treatment overseas had liberated them from the Catholic Church’s influence over the Italian state. Clinics often market their services to foreign clients by stressing the advantages of combining potentially stressful IVF treatment with a holiday in an attractive location. For some, being away from home while undergoing treatment offers greater privacy, perhaps enabling patients to ‘disguis[e] an absence from work for treatment abroad as a “holiday”’.35 In practice, Culley et al found that most people who had travelled abroad had multiple reasons for doing so: as well as avoiding long waiting lists for donor gametes, for example, UK patients were also motivated by the high costs of treatment in the UK, by the prospect of higher success rates abroad, by the hope of receiving treatment in a less stressful environment, and by dissatisfaction with the treatment that they had received in the UK.36

31  L Culley et al, ‘Crossing Borders for Fertility Treatment: Motivations, Destinations and Outcomes of UK Fertility Travellers’ (2011) 26 Human Reproduction 2373–2381. 32  Wannes Van Hoof, Guido Pennings and Petra De Sutter, ‘Cross-­Border Reproductive Care for Law Evasion: A Qualitative Study into the Experiences and Moral Perspectives of French Women Who Go to Belgium for Treatment with Donor Sperm’ (2015) 124 Social Science & Medicine 391–397. 33  Jenny Gunnarsson Payne, ‘Reproduction in Transition: Cross-­Border Egg Donation, Biodesirability and New Reproductive Subjectivities on the European Fertility Market’ (2015) 22 Gender, Place & Culture 107–122. 34  Giulia Zanini, ‘Abandoned by the State, Betrayed by the Church: Italian Experiences of Cross-­ Border Reproductive Care’ (2011) 23 Reproductive BioMedicine Online 565–572. 35  See L Culley et al, n 25. 36 Id.

1004   Emily Jackson One thing patients tend to have in common, however, is that they resist the label ‘reproductive tourism’,37 with many considering its implications positively offensive. Travelling to receive fertility treatment is not a ‘fun’ or ‘optional’ leisure activity but something people feel compelled to try in their quest for a baby. Nor do many patients recognise the description of cross-­border reproductive care as something which is available only to the rich.38 While travelling for fertility treatment will not be an option for those on very low incomes, the evidence does not support the claim that all reproductive travellers are wealthy. Instead, many patients have borrowed money to fund treatment overseas.39 Couples who experience infertility in some parts of the world, such as from the Middle East, South Asia, and Africa, may be ‘under incredible social and marital pressure’40 and, as a result, may place themselves under considerable financial strain to avoid the stigma of childlessness. As well as the costs cross-­reproductive care imposes on patients, there can also be costs to their domestic health services if they return from receiving treatment overseas with complications as a result of receiving poorly regulated services. Many countries outside the EU do not place restrictions on the number of embryos that can be transferred in one cycle. For example, British women have returned from having IVF treatment in India pregnant with triplets and quads.41 As well as the health risks for women and children, higher order multiple births impose significant costs on a country’s neonatal services and on social care and educational special needs services. If it becomes increasingly common for a country’s citizens to bypass its legal restrictions by travelling overseas, might this exert pressure on their home government to change the law? Or, might it have the opposite effect by working as a ‘safety valve’ to decrease the pressure for domestic law reform? In SH and Others v Austria,42 for example, while defending Austria’s restrictive laws on assisted conception, the Grand Chamber of the ECtHR observed ‘that there is no prohibition under Austrian law on going abroad to seek treatment of infertility that uses artificial procreation techniques not allowed in Austria and that in the event of a successful treatment the Civil Code contains clear rules on paternity and maternity that respect the wishes of the parents’. It therefore appeared to be easier to justify Austria’s wide ‘margin of appreciation’ because it was also simple for (some) Austrian citizens to avoid its effects by travelling to another country with a more relaxed regulatory regime.

37  MC Inhorn and P Patrizio, ‘Rethinking Reproductive “Tourism” as Reproductive “Exile”’ (2009) 92 Fertility and Sterility 904–906. 38  Debora Spar, ‘Reproductive Tourism and the Regulatory Map’ (2005) 352 N Engl J Med 531–533. 39  Marcia Inhorn, Pankaj Shrivastav and Pasquale Patrizio, ‘Assisted Reproductive Technologies and Fertility “Tourism”: Examples from Global Dubai and the Ivy League’ (2012) 31 Med Anthropology 249–265. 40 Id. 41  A McKelvey et al, ‘The Impact of Cross-­Border Reproductive Care or “Fertility Tourism” on NHS Maternity Services’ (2009) 116 BJOG 1520–1523. 42  [2011] ECHR 1879.

The Right to Procreate in Europe   1005 In addition, Pennings et al highlight a further reason why the civil disobedience of reproductive travellers is unlikely to be an effective driver of law reform.43 Few people wish to publicise the fact that they have had fertility treatment, and it is hard to build a protest movement around highly personal and strictly confidential information. A parallel might be drawn with abortion in Northern Ireland. Northern Irish women routinely avoid the prohibition on abortion at home by travelling to England, Scotland, or Wales for terminations, but until Sarah Ewart’s decision to speak publicly about her English termination,44 which was followed by a human rights challenge to Northern Irish abortion law,45 comparatively few women were willing to speak publicly about having done so. Cross-­border reproductive care is not the only way in which people bypass national regulation of assisted conception services. Informal sperm donation arrangements between strangers are becoming more common as a result of introduction websites through which potential donors and recipients make contact with each other. These work in the same way as dating websites, and it could be argued that this sort of internet-­ assisted conception is not, in fact, all that different from the now common practice of looking for sexual and romantic partners online. Cost may be one reason why people look for sperm donors online, but it is also common for would-­be recipients and donors to be looking for something that they cannot obtain from treatment in a regulated clinic, such as contact between the donor and child during childhood or even a co-­parenting arrangement.46 Recipients and donors may also value the opportunity to meet one another before conception. There are risks associated with informal sperm donation arrangements, however. Unlike treatment with donated sperm in a clinic, would-­be donors are not screened or vetted. There is no limit on how many families can be produced from one donor, and there is no register of information for donor-­conceived individuals. Some ‘donors’ advertise themselves as offering ‘NI [natural intercourse] only’. Often as a result of differing expectations of the relationship, these arrangements can break down after the child’s birth, and the family courts may then have to resolve disputes between the parents about paternity and contact arrangements.47 43  G Pennings et al, ‘ESHRE Task Force on Ethics and Law 15: Cross-­Border Reproductive Care’ (2008) 23 Human Reproduction 2182–2184. 44  ‘Abortion Laws NI: Sarah Ewart Says Travelling “Made Abortion More Traumatic”’ BBC News, 28 May 2018. 45  In the matter of an application by the Northern Ireland Human Rights Commission for Judicial Review (Northern Ireland) (2018) UKSC 27. See also Claire Simpson, ‘Fresh High Court Challenge to North’s Abortion Laws to Be Launched’ The Irish Times 29 January 2019. 46  Vasanti Jadva et al, ‘“Friendly Allies in Raising a Child”: A Survey of Men and Women Seeking Elective Co-­Parenting Arrangements via an Online Connection Website’ (2015) 30 Human Reproduction 1896–1906; Tabitha Freeman et al, ‘Online Sperm Donation: A Survey of the Demographic Characteristics, Motivations, Preferences and Experiences of Sperm Donors on a Connection Website’ (2016) 31 Human Reproduction 2082–2089. 47  See, for example, M v F (Legal Paternity)? [2013] EWHC 1901 (Fam); JB v KS (Contact: Parental Responsibility)? [2015] EWHC 180 (Fam).

1006   Emily Jackson

7  Paying for Fertility Services There is a thriving market in assisted conception services throughout Europe, and, in many European countries, this co-­exists with state-­funded fertility services. In no country is there unlimited access to state funding: rather, strict limits on funding are the norm. First, there are usually age limits for female patients. While preventing elderly motherhood may be a subsidiary concern, age limits are principally directed towards the clinical effectiveness of treatment, which declines during a woman’s 30s, and drops rapidly after the age of 40. In the UK, for example, the National Institute for Health and Care Excellence (NICE) has recommended that the National Health Service (NHS) should fund three full cycles of IVF (ie a fresh cycle followed by further cycles using the frozen embryos) for women younger than 40 years and one full cycle for women aged 40–42, who must additionally not have received IVF treatment before and not have low ovarian reserve. Second, there are usually restrictions on the number of cycles that people can have at public expense. Indeed, the NICE guideline is not mandatory—it is guidance—and few local health authorities (known as Clinical Commissioning Groups or CCGs) fund the full three cycles for eligible women. In 2018, it was reported that only 13 per cent of CCGs provide three full cycles of IVF to eligible women; 60 per cent offer one NHS-­funded cycle (most of which fund only one fresh cycle), and 4 per cent provide no cycles at all.48 Third, state funding is sometimes available only to those who have a clinical need for it, thus excluding people who need assistance because they do not have an opposite-­sex partner. The NHS will fund single and lesbian women treatment, but this must be on the grounds of infertility. That is, just as heterosexual couples must have failed to conceive through sexual intercourse, for single and lesbian women, they may be able to access NHS-­funded treatment only after they have failed to conceive after a year of regular self-­ funded intrauterine insemination.49 As a result of these restrictions, a significant proportion of people who need access to fertility services in order to conceive will have to pay for it out of pocket. In addition, there can be no guarantee that a state-­funded cycle of fertility treatment will succeed— for almost all categories of patient, fertility treatment fails more frequently than it succeeds. As a result, couples whose NHS-­funded treatment has failed will have to pay for any further cycles of treatment. The existence of a flourishing free market in fertility services, even in countries with comprehensive national health services, has led to the emergence of two new practices in the fertility clinic. First, egg-­sharing arrangements enable women under the age of 36 to access free or cheap treatment in return for donating some of their eggs for use in the 48  See further www.fertilityfairness.co.uk. 49  See further Atina Krajewska, ‘Access of Single Women to Fertility Treatment: A Case of Incidental Discrimination’ (2015) 23 Med L Rev 620–645.

The Right to Procreate in Europe   1007 treatment of others. On the one hand, egg-­sharing schemes might look like a ‘win-­win’ option: it enables some women to have IVF that they could not otherwise afford while simultaneously increasing the number of eggs available to women who need donor eggs. On the other hand, egg-­sharing agreements represent substantial indirect payment for eggs (a free cycle of IVF is commonly worth about £5,000). As we saw earlier, within the EU, payments to donors are supposed to be limited to ‘making good’ the inconvenience of donation. Not only is £5,000 significantly greater than the ‘going rate’ for in­con­ven­ ience payments, but also an egg sharer is not, in fact, being inconvenienced at all because she is undergoing egg retrieval as part of her own treatment. It has also been suggested that egg sharing might cause psychological difficulties if the egg sharer’s own treatment fails while another woman is treated successfully with her eggs. Interestingly, this latter concern is not borne out by empirical research, which instead indicates that egg sharers feel considerable empathy for other infertile women.50 Second, there is a growing market in the fertility sector in ‘add-­ons’; that is, additional services which clinics sell to patients in order to increase their chance of success.51 Fertility patients who are paying a large sum of money for one cycle of treatment have a significant financial as well as emotional interest in that cycle’s success. They are then particularly susceptible to being sold ‘extras’ that are said to be capable of increasing the chance that an embryo will implant in the woman’s uterus. Many add-­on treatments lack any robust evidence base. Some (like acupuncture) are simply a waste of money, while others (like reproductive immunology) can be positively harmful. But it is easy to see why patients might be willing to ‘try anything’ to increase their chance of success. The potential for overselling and sharp practices is therefore considerable. Even if a clinic is honest with patients that there are no robust clinical trials supporting the use of a particular add-­on treatment, they might also tell patients that they have had some success with it in their clinic. This is anecdote, rather than evidence, but it may be persuasive to a patient who is desperate for her cycle of IVF to succeed.52

8 Conclusion When IVF was in its infancy, it was relatively straightforward for countries to control their citizens’ access to fertility treatments. As a result of the internet, it is now exceptionally easy for citizens who can afford it to travel elsewhere to receive services that are not available to them at home. If the consequence of legal restrictions is to propel 50  ‘Emotional and Relational Aspects of Egg-­Sharing: Egg-­Share Donors’ and Recipients’ Feelings About Each Other, Each Others’ Treatment Outcome, and Any Resulting Children’ (2012) 27 Human Reproduction 1690–1701. 51  Joyce Harper et al, ‘Adjuncts in the IVF Laboratory: Where Is the Evidence for “Add-­On” Interventions?’ (2017) 32 Human Reproduction 485–491. 52  J Wilkinson et al, ‘Reproductive Medicine: Still More ART than Science?’ (2019) 126 BJOG 138–141.

1008   Emily Jackson patients across borders, there may be sound practical reasons for liberal regulation. Encouraging patients to receive treatment in local, regulated clinics has many advantages over pushing them towards the range of services, both safe and unsafe, that they might be likely to find via a Google search. We may all be familiar with families created through fertility treatment and the transformational impact IVF has had on many families’ lives. It is, however, important to acknowledge that IVF has certainly not ended involuntary childlessness. Treatment is stressful, invasive, and often fails; it is also expensive. While statutory restrictions on access to fertility treatment are important, for many would-­be parents, it is a lack of funding which represents the most significant barrier to treatment. Academic commentary on reproductive technologies often focuses on the implications of dazzling new treatment options, such as mitochondrial transfer and genome editing. More mundane, but no less important, is the invisibility of the pain of infertility among the poor and underprivileged.

chapter 49

The R ight(s) to Procr eate a n d Assisted R eproducti v e Tech nologies i n the U n ited States I. Glenn Cohen

Roughly 12% of US women of reproductive age (15–44) have sought fertility care at some point in their lives, and an estimated 1.2 million women seek treatment per year.1 According to data from 2015, the most recent year available, 231,936 in vitro fertilization (IVF) cycles were undertaken nationwide, resulting in 72,913 births (48,820 singleton births and 24,093 multiple infant births).2 These IVF births represented about 2 percent of the nearly 4 million births that year.3 Much less is known about artificial insemination, preimplantation genetic diagnosis, anonymous sperm donation, etc. Demographically, women of color suffer higher infertility rates in America, while in terms of assisted reproductive technology (ART) outcomes Caucasians fare best, followed by Asians, Hispanics, and then African Americans.4 This chapter focuses on the right to procreate—or, as I will suggest, rights (plural) to  procreate—in the United States, with a focus on reproductive technology use. The  United States has been—sometimes derisively, sometimes enviously—too often 1   Judith Daar, Assisted Reproductive Technologies and Abortion, in The Oxford Handbook of U.S. Health Law 330, 335 (I. Glenn Cohen, Allison K. Hoffman & William M. Sage eds., 2015). 2   National Data, Assisted Reproductive Technology (ART) Data, Ctrs. for Disease Control and Prevention, https://nccd.cdc.gov/drh_art/rdPage.aspx?rdReport=DRH_ART.ClinicInfo&rdRequestF orward=True&ClinicId=9999&ShowNational=1 (last visited June 11, 2018). 3   Joyce A. Martin, et al., Births: Final Data for 2015, 66(1) Nat’l Vital Stat. Rep. 1 (2017), https:// www.cdc.gov/nchs/data/nvsr/nvsr66/nvsr66_01.pdf 4  Daar, supra note 1, at 336.

1010   I. Glenn Cohen described as the “wild west” of reproductive technology use. When measured against many of its comparators—Canada, Australia, the UK, Germany, etc.—it is undoubtedly true that more forms of reproductive technology use are permitted in the United States than elsewhere. It is for this reason that the United States has been a frequent destination for “circumvention tourism” or “fertility tourism,” a topic I touch on in this chapter. At the same time, it would be wrong to think that reproductive medicine is unregulated in the United States. It is just that the regulation is more fragmented, both in terms of the locus of control (federal vs. state authority, governmental vs. professional self-­regulation, etc.) and also of the legal sources involved (more of a focus on tort law and family law than direct regulation at the statutory or constitutional level). This chapter does not aim to give a comprehensive account of reproductive technology law in the United States.5 Instead, the goal is to telescope a considerable amount of law and theory to give one an overall comparative impression of directions and texture. Again, largely in the interest of space, precious little attention will be given to the details of reproductive medicine, and instead the focus will be on its legal and ethical dimensions. At a more thematic level, if one were to ask “What is distinctive about the United States when it comes to the law’s interface with reproductive technologies?” I believe there would be a few key salient points. • There is surprisingly little US constitutional case law pertaining to reproductive technology. The result has been significant uncertainty as to which forms of regulation of reproductive technology might violate the US Constitution. • Particularly concerning family law aspects of reproductive technology, the locus of authority has been left much more to the states than the federal government. The result has been significant divergences between states and also intra-­US fertility tourism. • Professional self-­regulation and tort law play a major role in determining what reproductive services are offered in the United States, likely much more so than in peer countries. • The regulatory status of reproductive technologies is inexorably linked with continued public contestation and political positioning on the topic of abortion. • As compared to peer countries, market forces play a much larger role in determining what reproductive services are offered in the United States. The result has been that access considerations are much more heavily mediated by ability to pay rather than direct regulation. In the United States, even when the government takes steps to ensure access, it typically does so through an overlay on top of the market. These five themes are explored in the remainder of this chapter.

  For that there are many good sources such as Judith Daar, Reproductive Technologies and the Law (2d ed. 2013). 5

The Right(s) to Procreate in the United States   1011

1  The Constitution and the Right(s) to Procreate Some scholars have discussed a “right to procreate” alongside a “right not to procreate.” As I have argued elsewhere, that term masks a significant complexity when applied to reproductive technology use. A woman can be a parent in (at least) three possible senses: gestational parent, genetic parent, and legal parent (i.e., a parent whose rights and obligations are recognized by the state), while a man can be a parent in the latter two senses (who knows whether technologies like uterine transplants might one day give men gestational parenthood possibilities as well). In the case of coital reproduction, for women, these three kinds of parentage and the right to enjoy them are bundled: a woman pregnant through coital intercourse is the gestational and genetic parent of the child, and the law will (at least initially—adoption may come later) assign her the role of legal parent. For reproductive technologies both the rights to be and not to be a parent can be unbundled, generating: A right not to be a gestational parent A right not to be a genetic parent A right not to be a legal parent

A right to be a gestational parent A right to be a genetic parent A right to be a legal parent.6

Such rights can be understood in negative or positive liberty terms, but here for the most part I focus on the negative liberty conception—a right to be free from state interference with one’s right to be (for example) a gestational parent. This unbundling helps us see the way in which various forms of state regulation of reproductive technology use might be thought to involve one of these rights. For example, in a state such as Michigan, where commercial surrogacy is criminally prohibited,7 a would-­be gestational surrogate (i.e., one who gestates a fetus made from the genetic material of others) might claim that her right to be a gestational parent is being contravened but not her right to be a genetic parent. This unbundling framework reveals significant complexity: How do these rights intersect? Is there a lexical priority, or are their conflicts subject to case-­by-­case balancing? What is the moral basis for each of these unbundled rights? Can they be waived in advance? I have discussed these issues elsewhere.8 Here, though, I want to focus on the question of whether any of the rights to be a parent are constitutionally recognized when it comes to reproductive technology usage. Surprisingly, the answer is far from clear. We have three sets of Supreme Court cases that speak to the issue, but none speaks directly to reproductive technology. First, there are the cases on a right to contraception, focused on the state’s attempt to limit the right to access contraception—so the rights not to procreate not the rights to procreate. But in Eisenstadt v. Baird, the Court’s language (but not holding) goes significantly further:   I. Glenn Cohen, The Constitution and the Rights Not to Procreate, 60 Stan. L. Rev. 1135, 1140 (2008).   Mich. Comp. Laws Ann. § 722.859 (West 2011). 8   E.g., Cohen, supra note 6; I. Glenn Cohen, The Right Not to Be a Genetic Parent, 81 S. Cal. L. Rev. 1115 (2008). 6 7

1012   I. Glenn Cohen If the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child.9

The Supreme Court has cited this language to prohibit policies that impose a “heavy burden” on the right to procreate in the absence of a substantial justification for the ­policies.10 For example, the Court invalidated a maternity leave policy that required teachers to go on leave at least four months prior to their due date for the birth of their child.11 At the same time, the Court permits burdens on the right to procreate that are “marginal and indirect.”12 Thus, the State of Maryland could cap welfare benefits for poor families with children once the family size reached six people.13 In upholding the cap on benefits, the Court’s decision presaged its abortion funding decisions, in which it has held that while the government cannot prevent a woman from seeking an abortion, it does not have to fund her abortion. All of this suggests that while the Court would be comfortable with policies that deny funding for assisted reproduction services, it would be concerned with policies that prevent people from obtaining the services with their own funding. And in drawing the line between acceptable and unacceptable impediments to obtaining services, the strength of the state’s justifications for its regulations would be very important. For example, the Court would likely be more sympathetic to restrictions on commercial surrogacy than on IVF for an infertile couple. The second set of cases pertain to the abortion right. These cases speak clearly to a right for a woman to serve as a gestational parent even against the will of the child’s genetic father—that is, they reject a parental veto over a gestational mother’s abortion rights; some of the language (if not the holding) of these cases could be construed to extend farther. For example, in Planned Parenthood, the Plurality opinion contains a famous passage pontificating that [M]atters involving the most intimate and personal choices a person may make in a lifetime, choices central to personal dignity and autonomy, are central to the liberty protected by the Fourteenth Amendment. At the heart of liberty is the right to define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life. Beliefs about these matters could not define the attributes of personhood were they formed under compulsion of the State.14

Elsewhere I have argued that there are problems with reading this language too broadly.15 In any event, as I and others have argued, the applicability of this language 9   405 U.S. 438, 453 (1972); see also Carey v. Population Servs. Int’l, 431 U.S. 678, 685 (1977) (“[I]n a field that by definition concerns the most intimate of human activities and relationships, decisions whether to accomplish or to prevent conception are among the most private and sensitive.”). 10   Cleveland Board of Education v. LaFleur, 414 U.S. 632, 640 (1974).    11  Id. at 648. 12   Dandridge v. Williams, 397 U.S. 471, 520 n.14 (1970) (Marshall, J., and Brennan, J., dissenting) 13   Id. at 487.    14  See Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 851 (1992). 15  See Cohen, supra note 5, at 1159–1167.

The Right(s) to Procreate in the United States   1013 and indeed its underlying political philosophy has limited application to surrogacy, or indeed any reproductive technology usage where the state is not seeking to prevent a woman’s gestation. Moreover, the state’s interest in the surrogacy context might be considered quite differently than in attempts to restrict abortion; among other things, child welfare-­type arguments might play a quite different role. In sum, I think one could easily write a plausible Supreme Court opinion distinguishing or relying on these cases and come out any which way on most reproductive technology disputes. The third case is probably the most helpful in arguing for a constitutional right pertaining to reproductive technology: Skinner v. Oklahoma.16 That case involved a law that permitted the sterilization of thieves (but not embezzlers) after three convictions. In striking down the law as unconstitutional, Justice Douglas famously inveighed that We are dealing here with legislation which involves one of the basic civil rights of man. Marriage and procreation are fundamental to the very existence and survival of the race. The power to sterilize, if exercised, may have subtle, far-­reaching and devastating effects. In evil or reckless hands it can cause races or types which are inimical to the dominant group to wither and disappear. There is no redemption for the individual whom the law touches. Any experiment which the State conducts is to his irreparable injury. He is forever deprived of a basic liberty.17

However, it is far from certain that one can go from this language to rights to procreate relevant in the reproductive technology context.18 To begin with, Skinner rests on an Equal Protection claim that the disparate treatment of the two convicted criminals was problematic, with the claim augmented by but not squarely decided on a claim of substantive due process.19 The most narrow reading of the case (but one that is very implausible) is that there is no substantive due process claim even for coital reproduction.20 Given subsequent case law among the lower courts and the Supreme Court’s citing Skinner as providing a right to have children or procreate, that reading seems highly unlikely. Somewhat more plausibly, one might argue that Skinner is applicable only to coital reproduction and inapplicable in the realm of assisted reproduction—“for one thing, sterilization has a component of physical invasion, and thus a bodily integrity violation, not present in these cases”;21 moreover, it was taking away something (reproductive capacity) the individual already had, not denying them access to a treatment that would enable them to reproduce genetically.

  316 U.S. 535 (1942).    17  Id. at 541.   But see, e.g., John A. Robertson, Children of Choice: Freedom and the New Reproductive Technologies 36–38 (1994) (relying on Skinner for such an argument). 19   See Radhika Rao, Equal Liberty: Assisted Reproductive Technology and Reproductive Equality, 76 Geo. Wash. L. Rev. 1457, 1466 (2008). 20   Cf. Victoria F. Nourse, In Reckless Hands: Skinner v. Oklahoma and the Near Triumph of American Eugenics 165 (2008). 21   See Cohen, supra note 6, at 1195 n.244; David B. Cruz, “The Sexual Freedom Cases”? Contraception, Abortion, Abstinence, and the Constitution, 35 Harv. C.R.-C.L. L.Rev. 299, 361 (2000). 16 18

1014   I. Glenn Cohen There are other readings that have been suggested that would make room for some reproductive technology uses as constitutionally protected but not others. Skinner protects as a fundamental right any use of reproductive technologies that simulates that which would be achievable by coital reproduction in the fertile individual22— this would exclude gene editing for example, but potentially also mitochondrial replacement therapy depending on how we define the parameters; it would protect IVF and artificial insemination, among others. Alternatively, Skinner protects as fundamental rights only noncommercial forms of reproduction but not forms that require payment, in analogy to the way that Lawrence v. Texas seems to subject the criminalization of sexually intimate activities to heightened constitutional scrutiny, but that does not mean the state is subject to the same scrutiny if it criminalizes paying for those activities.23 This view would permit a right to noncommercial egg donation, for example, but not commercial egg provision. There are many other possible readings each of which would have its own implications for state attempts to curtail ART use.24 If none of these readings is doctrinally preordained, which to prefer might depend either on high-­order constitutional law commitments (e.g., minimalism, originalism) or the values one sees bred in the bones of the relevant constitutional provisions (e.g., the right to traditional family structures vs. the right to reproductive liberty and the importance of limiting state control over the body, etc.). In other words, the answer will depend very much on the theorist. Surprisingly, notwithstanding the fact that ARTs such as commercial surrogacy are actually prohibited in some states, there have been precious few constitutional challenges.25 In one of the very few, J. R. v. Utah, a Utah federal court declared unconstitutional a  Utah statute declaring that the gestational mother would always be granted legal ­parentage in a surrogacy arrangement.26 It found that it was not breaking new ground by recognizing a constitutional right in this case, instead being asked to apply the “clearly established” and “fundamental right to bear and raise children within the context of a marriage.”27 Finding such a right is only the first step in an American constitutional   I associate this view with the late John Robertson, supra note 12, although I do not do justice to the nuance of his view in this short space. 23   Or at least so suggests the Lawrence majority. 539 U.S. 558, 577–579 (2003). But see id. at 592, 603 (Scalia, J., dissenting) (attacking this distinction). 24   See, e.g., Carl Wellman, Medical Law and Moral Rights 145–146 (2005) (reading Skinner as limited to marriage); Carter J. Dillard, Rethinking the Procreative Right, 10 Yale Hum. Rts. Dev. L.J. 1, 44–45 (2007) (reading Skinner as protecting only a right to “self-replace” and thus a fundamental right to only one or two children per couple). 25   The constitutional rights of a gestational surrogate were raised and mentioned in the famous surrogacy case of Baby M, but the court found it did not need to reach the issue to decide the case. In re Baby M, 537 A.2d 1227, 1253–1255 (N.J. 1988) (discussing but not deciding the issue). The only Supreme Court case to actually involve reproductive technologies directly, Astrue v. Capato ex rel. B.N.C., 566 U.S. 541 (2012), involving eligibility for social security benefits for posthumous children, did not dwell on the Constitutional issue. 26   261 F. Supp. 2d 1268, 1296–1297 (D. Utah 2002).    27  Id. at 1277–1278. 22

The Right(s) to Procreate in the United States   1015 analysis: the court went on to decide that the Utah statute could not survive the appropriate level of constitutional scrutiny. In an even earlier case from 1990, an Illinois district court struck down as unconstitutionally vague an Illinois law that prohibited “experimentation” on a fetus, which the Court worried could reach embryo transfer or chorionic villi sampling.28 It connected this holding to the contraception and abortion cases, writing that “[i]t takes no great leap of logic to see that within the cluster of constitutionally protected choices that includes the right to have access to contraceptives, there must be included within that cluster the right to submit to a medical procedure that may bring about, rather than prevent, pregnancy” and that as to prenatal testing the “cluster of constitutional choices that includes the right to abort a fetus within the first trimester must also include the right to submit to a procedure designed to give information about that fetus which can then lead to a decision to abort.”29 But the case is almost 30 years old and has not been much discussed. Beyond these cases we find little case law. The Supreme Court has never weighed in on a surrogacy case. It is generally assumed that surrogates retain a right to have or refrain from having an abortion as well, though there is only scattered authority for the proposition and it is most often discussed in the context of whether that right can be waived through a surrogacy agreement—to which case and statutory law in some states has said no.30 If (at least absent agreements) gestational surrogates can have or refrain from having an abortion, as appears to be the case, this suggests that, for reproductive technologies, the right to be a gestational parent attaches even if there is no right to be a genetic or legal parent that would flow from compelled continuation of pregnancy. But that is a pregnancy already begun. For attempts to regulate whether surrogacy or other ART happens at all there is a range of open questions. Is there a right to retain a surrogate? What about a right to become a surrogate? What about a right to retain a sperm or egg donor? A right to become a sperm or egg donor?31 What about a right to be a uterine transplant donor,32 wherein one’s uterus will be used to permit another woman to gestate?   Lifchez v. Hartigan, 735 F. Supp. 1361 (N.D. Ill. 1990).    29  Id. at 1377.   See, e.g., Cohen, supra note 5, at 1191–1192 (summarizing law and commentary on the point). 31   While it might seem more intuitive that the right to retain a sperm or egg donor or surrogate is on more solid footing than the right to become one, in fact it depends on one’s theory of what the Constitution protects regarding the rights to procreate. A party seeking to become a surrogate or sperm or egg donor is making claims relating to control of his or her own body and its materials to produce a child for which he or she will not have legal parentage; a party seeking to hire a surrogate or sperm or egg donor is making a claim as to his or her desire to fulfill a reproductive goal of producing a child he or she intends to parent. In fact, the configurations can get more complicated, depending on the source of sperm, egg, and gestation, in that one part of a couple might be making one kind of claim versus another. For example, consider a case involving (1) an egg donor, (2) gestational surrogate, (3) fertile intended father, and (4) infertile intended mother. (1), (2), and (3) are making claims about the use of his or her body and its products, (3) and (4) are making a claim about fulfilling reproductive desires to produce a child he or she intends to parent. Indeed, (3) has available both kinds of claims. 32   For more on this technology, see, e.g., John A. Robertson, Other Women’s Wombs: Uterus Transplants and Gestational Surrogacy, 3 J.L. & Biosciences 68 (2016). 28

30

1016   I. Glenn Cohen In a country where everything from “gays to guns” makes it to the Supreme Court and is resolved in relation to the Constitution, the rights to procreate through reproductive technologies remain an area of marked constitutional uncertainty.

2  Federal Versus State Regulation and the Prominence of Intra-­U S Fertility Tourism A US person is governed by huge numbers of federal statutes and regulations—indeed, no one knows quite how many exist, though one estimate suggested 4,312 federal laws and 88,899 federal rules and regulations went into effect from 1995 to 2016.33 Given this mass of federal law, it is curious how little directly pertains to reproductive technologies. It has been claimed that there is only a “single federal statute that deals with reproductive technology,”34 with the author pointing to a law requiring fertility clinics to report their success rates to the Centers for Disease Control on a standardized form.35 This is probably a bit of an overstatement: for example, there is a Congressional appropriations rider that President Obama signed into law in December 2015 that prohibits the US Food and Drug Administration (FDA) from considering “research in which a human embryo is intentionally created or modified to include heritable genetic modification” and thus essentially blocks US approval of mitochondrial replacement therapy.36 Still, it is remarkable how much of the regulation of reproductive technologies is left to individual US states.37

  E.g., Clyde Wayne Crews Jr., How Many Rules and Regulations Do Federal Agencies Issue?, Forbes (Aug. 15, 2017, 12:48 PM), https://www.forbes.com/sites/waynecrews/2017/08/15/ how-many-rules-and-regulations-do-federal-agencies-issue/#59afd6071e64. 34   Dov Fox, Reproductive Negligence, 117 Colum. L. Rev. 149, 162 (2017). 35   Fertility Clinic Success Rate and Certification Act of 1992, Pub. L. No. 102-493, 106 Stat. 3146 (codified as amended in scattered sections of 42 U.S.C. (2012)). 36   Eli Y. Adashi & I. Glenn Cohen, Preventing Mitochondrial Disease: A Path Forward, 131 Obstetrics & Gynecology 553 (2018). There are also federal regulations setting “health and safety parameters affecting donated gametes and embryos.” Daar, supra note 1, at 345 (citing Human Cells, Tissues, and Cellular and Tissue-Based Products, 21 C.F.R. § 1271). 37   At least as compared to our peer countries. Within the United States, how much of an outlier reproductive technologies are depends on the comparison set. While drugs and devices are primarily 33

The Right(s) to Procreate in the United States   1017 And boy, do the states disagree. Let us take the example of surrogacy. There are at least six different legal regimes pertaining to gestational surrogacy across the United States. 1. Criminalize. The act of entering into a surrogacy agreement is, under some circumstances, a criminal offense punished by the law. 2. Void. The surrogacy agreements are not valid or enforceable by either party. 3. Voidable. The surrogacy agreements are not valid or enforceable if one side of the bargain, almost always the surrogate, objects at the time of performance. If, however, the surrogate does not object, the agreement is enforceable. Importantly, in a case where the intended parents want to back out, the surrogate may enforce the agreement against them. 4. Enforceable. The court will give legal effect to the agreement and allow either party to seek enforcement of the agreement if the other party tries to breach. 5. Judicially preapproved. The surrogacy agreement will be treated as enforceable only if, before the agreement is entered into, a court reviews the agreement and determines that it comports with various court or legislatively mandated requirements pertaining not only to its terms but also to the process by which it was negotiated (e.g., each party had a lawyer review it). 6. Legally valid to determine parentage. The surrogacy agreement may or may not be enforceable by either party, but when the parties together seek to require the state to legally recognize the parental rights of the intended parents, the surrogacy agreement carries the legal force to determine parentage.38 When it comes to parentage achieved through known and unknown sperm providers, there is also significant disagreement about what is required to terminate the parentage of a sperm donor. States divide on questions such as whether insemination must take place in a doctor’s office, whether the recipient of sperm must be married, whether agreements with donors are enforceable, whether post-­birth contact between donor and child can in some cases nullify any termination of parentage, and how to treat same-­sex couples.39

regulated at the federal level via the FDA, surgical techniques and the like are often regulated by state law, especially tort law, and insurance law relating to reimbursement. 38   I. Glenn Cohen & Katherine L. Kraschel, Gestational Surrogacy Agreements: Enforcement and Breach, in Handbook of Gestational Surrogacy: International Clinical Practice and Policy Issues 85, 86–87 (E. Scott Sills ed., 2016). For more specific state-by-state analysis, see, e.g., Radhika Rao, Surrogacy Law in the United States: The Outcome of Ambivalence, in Surrogate Motherhood: International Perspectives 23, 23–31 (Rachel Cook et al. eds., 2003); Darra L. Hofman, “Mama’s Baby, Daddy’s Maybe”: A State-by-State Survey of Surrogacy Laws and Their Disparate Gender Impact, 35 Wm. Mitchell L. Rev. 449 (2009); Joseph F. Morrissey, Surrogacy: The Process, the Law, and the Contracts, 51 Willamette L. Rev. 459, 485–503 (2015). 39   For a good summary of these divergences see Mark A. Hall, David Orentlicher, Mary Anne Bobinski, Nicholas Bagley & I. Glenn Cohen, Health Care Law and Ethics 809–816 (9th ed. 2018).

1018   I. Glenn Cohen States also differ on how they treat embryo disposition disputes—disputes between reproductive partners after the dissolution of a relationship on who may use or prevent use of embryos frozen as part of IVF. When faced with this clash between rights to and not to procreate, most states have forbidden the woman from using frozen embryos against the man’s objection, but some have carved out exceptions for cases where the woman would otherwise have no further opportunities for genetic reproduction, and, in some states, the existence of an agreement between the parties is important for the disposition.40 To give another example, states have come to different conclusions on issues of posthumous reproduction—in particular, related to whether the posthumously conceived may receive social security benefits and whether they count as “issue” for a trust.41 There are many more such disagreements. Speaking at a very high level, the marked differences in approaches in the United States may be driven by a myriad of kinds of arguments: arguments about protecting child welfare, arguments about protecting the intended parents, arguments about protecting reproductive strangers (i.e., the surrogate, the sperm or egg donor), legal moralistic arguments about traditional family formation, anti-­commodificationist concerns, and preferences for contractual enforcement over beliefs in the potential for changed selves, among others. How should one feel about this level of disagreement? On the one hand, one might invoke Justice Brandeis’s famous phrase: “It is one of the happy incidents of the federal system that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”42 And in thinking more historically and more globally, one might recall the influential jurist Friedrich Carl von Savigny’s teaching that, unlike contract law (which he thought universal), “family law gave voice to the spirit of the people, which was inevitably local.”43 On the other hand, there are two significant implications (which some may view as unfortunate) to the fact that these issues are left to states that disagree. The first has to do with the fact that citizens do, in fact, move, and complex questions arise when families who are formed under one legal regime encounter a change in status—such as divorce— in another. This has been particularly complicated when it involves same-­sex parents, one of whom does not have a biological relationship to the child.44 Second, the diversity of legal regimes gives rise to intra-­US fertility tourism, with many residents of other 40   For a recent summary of the cases by state, see I. Glenn Cohen & Eli Y. Adashi, Embryo Disposition Disputes: Controversies and Case Law, 46 Hastings Ctr. Rep. 13, 14 (2016). 41   See Finley v. Astrue, 270 S.W.3d 849, 850 (Ark. 2008); Woodward v. Comm’r of Soc. Sec., 760 N.E.2d 257, 272 (Mass. 2002); Khabbaz ex rel. Eng v. Comm’r, Soc. Sec. Admin., 930 A.2d 1180, 1182 (N.H. 2007); In re Martin B., 841 N.Y.S.2d 207, 208 (N.Y. Sur. Ct. 2007). 42   New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting). 43   Janet Halley, What Is Family Law?: A Genealogy Part II, 23 Yale J.L. & Human. 189, 190 (2011). 44   See Jessica Feinberg, Consideration of Genetic Connections in Child Custody Disputes Between Same-Sex Parents: Fair or Foul?, 81 Mo. L. Rev. 331, 351–54 (2016); Joanna L. Grossman, Parentage Without Gender, 17 Cardozo J. Conflict Resol. 717, 731–736 (2016); Courtney G. Joslin, Interstate Recognition of Parentage in a Time of Disharmony: Same-Sex Parent Families and Beyond, 70 Ohio

The Right(s) to Procreate in the United States   1019 states flocking to California, in particular, to take advantage of its well-­established case law giving genetic parents enforceable rights to parentage over the possible objection of a gestational surrogate.45 Should we view those traveling to permissive states as reproductive refugees or outlaws? As I have argued in depth elsewhere as to the international context, much will depend on the reason why a particular state is concerned with surrogacy within its borders—whether it is concerns about the welfare of children born through surrogacy, anti-­commodification concerns, concerns about the exploitation of the women who act as surrogates, etc.—because different reasons will lead to different views about whether this form of “circumvention tourism” is problematic or not.46

3  Professional Self-­R egulation, Tort, and the Reluctance to Legislate In most countries, governmental regulation guides much of reproductive practice. In the United States, in determining how ART services are delivered and what ARTs become available, professional self-­regulation plays a much more important role. The most important such regulator is the American Society for Reproductive Medicine (ASRM), a membership organization founded in 1944 and made up of those who work in reproductive medicine and ancillary professions.47 As Judith Daar has put it, ASRM aspires to be the “leader of ‘standards in the field’” through “its two main policy-­setting bodies: the Practice Committee that issues reports and guidelines on clinical practice and the Ethics Committee that produces periodic statements and guiding principles for physicians and others in the field.”48 A recent visit to the Practice Committee website reveals publications on, inter alia, the use of preimplantation genetic testing, guidance on the limits to the number of embryos to transfer, ovarian tissue cryopreservation, and many more.49 ASRM’s Ethics Committee website also lists Ethics Committee opinions on wide-­ranging topics including access to fertility services for

St. L.J. 563, 563–569 (2009); see also Morrissey, supra note 38, at 506–509 (discussing complexities relating to states enforcing choice-of-law clauses in surrogacy agreements versus finding them contrary to public policy). 45   See, e.g., Lisa C. Ikemoto, Reproductive Tourism: Equality Concerns in the Global Market for Fertility Services, 27 Law & Ineq. 277, 298 (2009); Seema Mohapatra, Stateless Babies & Adoption Scams: A Bioethical Analysis of International Commercial Surrogacy, 30 Berkeley J. Int’l L. 412, 423 (2012). 46   See I. Glenn Cohen, Patients with Passports: Medical Tourism, Law, and Ethics 371–420 (2015); I. Glenn Cohen, Circumvention Tourism, 97 Cornell L. Rev. 1309, 1373–1386 (2012). 47   Judith Daar, Federalizing Embryo Transfers: Taming the Wild West of Reproductive Medicine?, 23 Colum. J. Gender & L. 257, 272 (2012). 48   Id. at 273. 49   Practice Committee Documents, asrm, http://www.asrm.org/news-and-publications/practicecommittee-documents/ (last visited April 16, 2018).

1020   I. Glenn Cohen transgender individuals, embryo donation, futility in reproductive medicine, and informing offspring that they are donor conceived.50 We lack strong empirical evidence as to how effective professional self-­regulation by institutions like ASRM really is in ensuring ethical and quality medical care. To be fair, this is not because legal and medical scholars have simply decided not to look. Instead, designing high-­quality empirical investigations of the topic is very difficult—in some ways, more difficult than the (already difficult) questions about the effects of tort liability discussed later because ASRM’s Practice Guidelines and Ethics Committee opinions are national, and thus empirical scholars cannot make use of statewide variation or differences in the timing of introduction. In the absence of deep empirical knowledge, there is disagreement among legal scholars on how effective ASRM has been. The issue reached its highest prominence in the wake of the well-­publicized birth of octuplets achieved through IVF to Nadya Suleman, the so-­called Octomom, in January 2009.51 For example, Theresa Glennon complained that In the absence of specific governmental regulation regarding multiple embryo implantations, the ASRM, together with the Practice Committee of SART, has developed practice guidelines that many clinics may follow but that do not carry the force of law. The ASRM Practice Committee develops its guidelines behind closed doors, and the public is not able to see the evidence the Committee weighs in order to reach its recommendations . . . . Rather than providing strong leadership to make single-­embryo transfer the anchor for patient and physician decision-­making, ASRM guidelines appear designed to give fertility specialists wide discretion in their practices. . . . These guidelines have not had any noticeable impact on the twinning birth rate. Nor has ASRM or SART taken a strong public stance to reduce twinning rates . . . . ASRM Practice Guidelines on embryo transfer also lag behind the current recommendations of European professional associations. Unlike the professional societies in the several European countries discussed below, SART and ASRM have failed to provide effective leadership regarding the reduction of multiple gestations, including twins. . . . [I]t is impossible to know the degree to which this failure stems from concerns that less aggressive forms of fertility treatment might reduce clinic profits or that fertility clinics will refuse to follow tougher guidelines.52

While accepting some of this critique, Judith Daar has a more charitable view of the effect of ASRM actions. ASRM “self-­regulators” appear to be somewhat sensitive to the critique that they are not doing enough to address the persistent twin rate in the U.S., revising the embryo   Ethics Committee Opinions and Webinars, asrm, http://www.asrm.org/news-and-publications/ ethics-committee-documents/ (last visited April 16, 2018). 51  Daar, supra note 47, at 258. 52   Theresa Glennon, Choosing One: Resolving the Epidemic of Multiples in Assisted Reproduction, 55 Vill. L. Rev. 147, 176–177 (2010) (footnotes omitted). 50

The Right(s) to Procreate in the United States   1021 transfer guidelines five times since their first issue in 1998, with each version recommending fewer transfers per cycle on average. These downward revisions may have contributed to the decrease in the higher-­order multiple birth rate (triplets or greater) which fell from 5 percent in 1999 to 2 percent in 2008. Defenders of voluntary guidelines cite this data subset as a reason to resist more formal regulation . . . . One interesting study found that embryo transfer rates show the greatest decline immediately after the publication of revised (downward) ASRM practice guidelines. Knowing that self-­regulation can have an impact on clinical practice, even if just a temporal impact, may increase its relative weight in the trilogy of regulatory sources that govern the practice of reproductive medicine.53

Outside of direct regulation and professional self-­regulation, tort liability also plays some role in guiding ART practice, but it is hard to know how much. More generally, there is significant contestation about whether medical malpractice leads to reductions in medical error as to medical practice generally.54 But to ARTs specifically, there is even more reason to be skeptical because many states limit wrongful life and (to a lesser extent) wrongful birth liability as well as create difficulties for other forms of tort recovery.55 Given significant doubts about the effectiveness of self-­regulation and regulation by tort law in the ART space, what explains the reluctance to press for more direct regulation by law in the United States? There are, I think, a few overlapping reasons. First, there is some doubt about the constitutionality of some restrictions for the reasons discussed earlier. Second, as discussed more later, one of the main tools for regulating medical practice in the United States—the regulation of insurance and reimbursement—is much less potent for reproductive technologies where (unlike for most medical procedures) most insurers do not cover it, so that patients are self-­pay. Second, there is a deep fear among progressives, in particular, that attempts to regulate the ART market will turn into regulation aimed at restricting access for the poor, LGBTQ persons, etc. Third, and relatedly, proposals for the regulation of ART in the United States are frequently intermingled with proposals to regulate abortion, again in ways that progressives find troubling.

 Daar, supra note 40, at 275–276 (footnotes omitted); see id. at 274.   There is a voluminous empirical literature on this subject. For some good starting places, see, for example, Michael Frakes, The Impact of Medical Liability Standards on Regional Variations in Physician Behavior: Evidence from the Adoption of National-Standard Rules, 103 Am. Econ. Rev. 257, 275 (2013); Michelle M. Mello & Troyen A. Brennan, Deterrence of Medical Errors: Theory and Evidence for Malpractice Reform, 80 Tex. L. Rev. 1595, 1606–1607 (2002); Myungho Paik, Bernard Black & David Hyman, The Receding Tide of Medical Malpractice Litigation: Part 2—Effect of Damage Caps, 10 J. Empirical Legal Stud. 639 (2013); Joanna C. Schwartz, A Dose of Reality for Medical Malpractice Reform, 88 N.Y.U. L. Rev. 1224, 1232 (2013); David M. Studdert et al., Claims, Errors, and Compensation Payments in Medical Malpractice Litigation, 354 New Eng. J. Med. 2024, 2029–2032 (2006). 55   See, e.g., Fox, supra note 34, at 164–173; Wendy F. Hensel, The Disabling Impact of Wrongful Birth and Wrongful Life Actions, 40 Harv. C.R.-C.L. L. Rev. 141, 143–145 (2005). 53

54

1022   I. Glenn Cohen Let me say a bit more about this last reason. The recent rise of initiatives of the personhood movement is a good example. In Roe v. Wade, Justice Blackmun wrote that “[i]f this suggestion of [fetal] personhood is established, the appellant’s case [arguing in favor of women’s choice], of course, collapses, for the fetus’ right to life would then be guaranteed specifically by the [Fourteenth] Amendment.”56 The Supreme Court did not, and never has, accepted the proposition that fetuses (let alone embryos) are persons within the meaning of the Constitution. But the “personhood movement” has attempted to get the issue revisited with a series of legislative and court moves. While the language and form of these proposals vary from state to state (legislative bills in some states versus ballot initiatives voted on directly by the public in others), each essentially attempts to secure legal rights for pre-­born human beings starting from the moment of fertilization or conception.57

While the movement has thus far been largely unsuccessful in its legislative attempts, more recently it has tried to use the judiciary to force recognition of embryonic personhood. In several embryo disposition disputes, personhood proponents have entered as amici curiae on the side of the woman seeking to use the embryos for reproduction and argued that the personhood of the embryos ought to tilt the decision in her favor.58 Thus, any serious attempt at regulation of reproductive technologies, which may be laudable on its own terms (depending on the proposed regulation), threatens to become a tool for restricting abortion as well. It is for this reason that US progressives may be leerier of opening up this “can of worms” than their peers in other countries. In the United States, abortion remains a deeply morally contested question, one for which both sides of the debate are constantly fighting for advantage using the courts and legislatures. In countries where the abortion question is better settled in one direction or the other, attempts at regulating reproductive technologies do not pose the same fraught issues.

4  The Market In part because of these prior factors (murkiness about when the state can restrict reproductive technology use, fragmentation of locus of regulation with a focus on state regulation and substantial divergence therein, heavy reliance on self-­regulation by the fertility industry, political reluctance to intercede), much of what is available in the United States by way of reproductive technologies is dictated by market forces.   Roe v. Wade, 410 U.S. 113, 156–157 (1973).   Lee Rubin Collins & Susan L. Crockin, Fighting “Personhood” Initiatives in the United States, 24 Reprod. BioMed. Online 689, 689 (2012). For more on the movement, see Daar, supra note 1, at 349–352; Jonathan F. Will, Beyond Abortion: Why the Personhood Movement Implicates Reproductive Choice, 39 Am. J.L. & Med. 573, 580 (2013); Jonathan F. Will, I. Glenn Cohen & Eli Y. Adashi, Personhood Seeking New Life with Republican Control, 93 Ind. L.J. 499 (2018). 58   See I. Glenn Cohen & Eli Y. Adashi, Personhood and the Three Branches of Government (under review). 56 57

The Right(s) to Procreate in the United States   1023 Surrogacy remains a lucrative business not just for US prospective parents but for those from abroad. Older mothers may have a chance of accessing reproductive services in the United States that would not be made available in many countries, assuming they can pay for those services. And to use one of the most extreme examples, as I noted in a 2013 article, one Los Angeles based clinic sought to “sharply cut costs by creating a single batch of embryos from one oocyte donor and one sperm donor, then divvying it up among several patients” with “the clinic, not the customer, control[ling] the embryos, typically making babies for three or four patients while paying just once for the donors and the laboratory work.”59 Buttressing these market norms is the fact that the vast majority of reproductive technology treatment in the United States is self-­pay.60 Because much of insurance regulation, including decisions of what to cover, is left to state law, we see significant divergences among US states. Only 15 states have insurance mandates requiring some coverage of reproductive services, and many require fairly little.61 “In the states that do not mandate coverage, insurance companies typically do not cover such treatments.”62 Even where insurance mandates are in place, for private employers who self-­insure, the mandates do not apply.63 Further, “[p]ublic programs such as Medicaid similarly treat fertility issues   I. Glenn Cohen & Eli Y. Adashi, Made-to-Order Embryos for Sale—A Brave New World?, 368 New Eng. J. Med. 2517, 2517 (2013) (quoting Alan Zambero, An Ethics Debate over Embryos on the Cheap, L.A. Times [November 19, 2012], http://articles.latimes.com/2012/nov/19/local/ la-me-embryo-20121120). 60   See, e.g., June Carbone & Jody Lyneé Madeira, Buyers in the Baby Market: Toward a Transparent Consumerism, 91 Wash. L. Rev. 71, 77 (2016); Jim Hawkins, Selling ART: An Empirical Assessment of Advertising on Fertility Clinics’ Websites, 88 Ind. L.J. 1147, 1161 (2013); John A. Robertson, Commerce and Regulation in the Assisted Reproduction Industry, 85 Tex. L. Rev. 665, 674 (2007) (reviewing Debora L. Spar, The Baby Business: How Money, Science, and Politics Drive the Commerce of Conception (2006)). 61   As Daniel Chen and I categorized these mandates in 2010: [F]ifteen states have introduced state-level insurance mandates pertaining to IVF and other reproductive technologies. But not all mandates are created equal; we . . . divide these mandates into four types: complete, partial, offer, and non-IVF. In a complete coverage state the mandate requires HMOs and other private insurers to cover the cost of infertility, including IVF. In Illinois, for example, all health insurance policies must cover at least four attempts at oocyte retrieval. A partial coverage state also covers IVF, but does so less generously. The mandate in Hawaii, for example, requires all insurance policies to cover IVF only when the patient has a greater than 5-year history of infertility, and then it requires covering the cost of only one IVF cycle. Mandate to offer states require insurance companies to offer one insurance policy that covers IVF to employers but does not require employers to adopt that policy. Texas is an example of a state that has such a mandate. A few states, such as Louisiana, have enacted reproductive technology mandates that specifically exclude IVF but cover other reproductive technology services. We call these non-IVF mandates. I. Glenn Cohen & Daniel L. Chen, Trading-Off Reproductive Technology and Adoption: Does Subsidizing IVF Decrease Adoption Rates and Should It Matter?, 95 Minn. L. Rev. 485, 537–538 (2010) (footnotes omitted). 62   Carbone & Madeira, supra note 60, at 77. 63   To offer a more technical but more accurate description, the state insurance mandates are preempted by the federal Employee Retirement Income Security Act of 1974 (ERISA). See 29 U.S.C.  § 1144(b)(2)(B) (2012); Cohen & Chen, supra note 61, at 501 n.79; Allison Overbay & Mark Hall, Insurance Regulation of Providers that Bear Risk, 22 Am. J.L. & Med. 361, 380 (1996). 59

1024   I. Glenn Cohen as elective and uncovered, and private charities do not place much emphasis on access to services such as IVF.”64 Orentlicher has argued that the failure to include reproductive technologies as an insured service in most US states constitutes a form of “discrimination out of dismissiveness” and suggests it is a mistake not to have infertility encompassed by existing legal protections for discrimination against persons with disabilities.65 In a similar vein, Daniel Chen and I have tried to offer an overlapping consensus of reasons for covering reproductive technologies by insurance—on Capabilities, Rawlsian, and Consequentialist theories—and argue that infertility is more properly thought of as a full healthcare need.66 There is an inherent opportunity for inequitable treatment in systems where reimbursement rather than prohibition becomes the key policy lever: those with means can decide to self-­pay to receive services unavailable to those who cannot afford to do so. The resulting largely self-­pay and market-­driven “baby business,” to use Debora Spar’s evocative phrase,67 has several implications: first, discourse about reproductive technology services in the United States much more openly embraces the language of commodification and the role of a patient as a “consumer” or “client” than elsewhere.68 Second, the industry has drawn the attention of antitrust scholars (and advocates) for its alleged attempts to engage in price-­setting, just as one might expect in other markets.69 Third, when patients pay out of pocket their behaviors may diverge from that which would be optimal from a public health perspective. The most discussed example has to do with multiple births.70 This shows that the United States lacks one important tool for regulating reproductive technology enjoyed by many of its peer countries: using the power of the purse to control

  Carbone & Madeira, supra note 60, at 77.   David Orentlicher, Discrimination Out of Dismissiveness: The Example of Infertility, 85 Ind. L.J. 143 (2010). 66   Cohen & Chen, supra note 61, at 500–528. 67   Debora L. Spar, The Baby Business: How Money, Science, and Politics Drive the Commerce of Conception (2006). 68   See generally id.; Carbone & Madeira, supra note 60, at 74. To be sure, there are definitely strategic uses of this discourse that interface in important ways with gender—for example, the way in which sperm providers are recruited, paid, etc., as opposed to egg providers. See generally Rene Almeling, Sex Cells: The Medical Market for Eggs and Sperm (2011). 69   See Kimberly D. Krawiec, Markets, Morals, and Limits in the Exchange of Human Eggs, 13 Geo. J.L. & Pub. Pol’y 349, 358 (2015); Kelly Knaub, Egg Donors Get Pay Limits Axed with Antitrust Settlement, law360 (Feb. 1, 2016, 7:01 PM est), https://www.law360.com/competition/ articles/753389?nl_pk=ad0ed999-c219-40d1-a23f-addcaed2be7c&utm_source=newsletter&utm_ medium=email&utm_campaign=competition. 70   Though to be sure the issue is not uncontested. For some of the key papers on the topic see Melinda B. Henne & M. Kate Bundorf, Insurance Mandates and Trends in Infertility Treatments, 89 Fertility & Sterility 66, 73 (2008); Tarun Jain et al., Insurance Coverage and Outcomes of in Vitro Fertilization, 347 New Eng. J. Med. 661 (2002); Meredith A. Reynolds et al., Does Insurance Coverage Decrease the Risk for Multiple Births Associated with Assisted Reproductive Technology?, 80 Fertility & Sterility 16, 22 (2003). 64 65

The Right(s) to Procreate in the United States   1025 the industry. In countries where a larger share of the reproductive technology usage is reimbursed by government, the state can discourage certain practices by failing to fund them without having to prohibit them outright. Moreover, the threat of terminating reimbursement for certain practices or to certain providers can be a powerful cudgel with which to persuade providers to acquiesce to the government’s preferred approach. That said, even when outright prohibition is selected, those with means can decide to travel abroad and engage in circumvention tourism to access services. Short of attempts to dissuade such activity through denying citizenship to the children who result or parental recognition,71 a better way of thinking of this is as imposing various levels of transaction costs on to those who would circumvent.

5 Conclusion The simple picture of the US position on reproductive technology regulation and its peer countries is quite Manichean: the United States is the wild west, where reproductive freedom reigns, be it as a port for fertility “refugees” or a Las Vegas of the worst kinds of excess; our peer countries are carefully regulated. I have tried to paint a more complex picture of the US situation. Instead of a lack of regulation we get multiple loci of control: divergent states rather than a single federal approach, more self-­regulatory than governmental. Moreover, while market dynamics play some role in resisting the regulation of the reproductive technology industry, I have suggested other forces are at play: reluctance to touch the third rail of abortion politics and uncertainty as to how wide a constitutional berth states have in regulating reproduction.

  See Cohen, supra note 46, at 371–420.

71

chapter 50

I n troduction to the R ight to Avoid Procr eation a n d the R egu l ation of Pr egna ncy Janne Rothmar Herrmann and Elizabeth Sepper

Fifty years have passed since international human rights law first formulated the right not to procreate as a cornerstone of reproductive freedom.1 Forty-­five have gone by since Denmark through legislation and the United States through constitutional interpretation recognized a right to abortion. However, the boundaries of the right to not procreate are greatly contested and unsettled on either side of the pond. For example, in May 2018, Ireland voted overwhelmingly to repeal the Eighth Amendment of the Irish Constitution which had recognized the equal right to life of the unborn and the pregnant woman. Approved by referendum in 1983, the constitutional provision had ensured that the Irish Supreme Court would not follow the lead of the US Supreme Court in Roe v. Wade and recognize a right to abortion. With the Eighth Amendment repealed, the Irish parliament has liberalized abortion law. In October of that same year, the United States prepared to move in the opposite direction with the appointment of Brett Kavanaugh as justice of the US Supreme Court. While it remains unclear whether the Court will overturn Roe, thereby allowing abortion bans, or will leave abortion doctrine in place while upholding virtually all regulation, it is generally agreed that the right to abortion stands on the precipice and that a number of US states may soon look like Ireland once did (and Malta still does) even as others retain liberal abortion laws. 1  The right to not to procreate, phrased as the right to decide on the number and spacing of children, was expressed for the first time in the Proclamation of Teheran in 1968, as a final act of the UN Conference on Human Rights.

1028   Janne Rothmar Herrmann and Elizabeth Sepper Despite the wide disagreement and divergence in approaches among European nations as well as among US states, the same principles guide legal, political, and moral debates over the right not to procreate and of the regulation of pregnant women. Interests in autonomy, equality, family formation, family life, privacy, medical ethics, health, and fetal life cut across jurisdictions, although their impact varies to greater and lesser degrees. These interests are reflected in the contours of the right not to procreate under the international human rights regime, which emerged out of two very different UN conferences focusing on population control (the United Nations International Conference on Population and Development in Cairo in 1994) and women’s rights (the Fourth World Conference on Women in Beijing in 1995), respectively.

1 Autonomy Autonomy, a basic liberty right fundamental in health law, is a, if not the, primary interest at stake in decision-­making about reproduction and pregnancy. The principle’s function is first and foremost to ensure that human beings are not subjected to unwanted interventions. In health law generally, autonomy ensures that the individual can freely choose her own preferences in relation to her person and body without undue interference, coercion, or control. To achieve this situation of free choice, she must be presented with all relevant information, including alternative options, in a way which is tangible to her. Legal requirements to deliver information and obtain consent in relation to medical treatment and biomedical research seek to minimize the risk of patients being subjected to harmful or unwanted interventions. In relation to procreative autonomy specifically, the freedom to make choices concerning reproduction can be seen as a vital aspect of human dignity, a fundamental feature of all democratic societies.2 In no country, however, does a woman have full autonomy over procreation, and, in most countries, her choices for childbirth or contraception are circumscribed by legal or financing regimes. Likewise, states impose limits on autonomy out of concern for public health, women’s health, and the interests of third parties, for example, regulating the terms of access to home birth. They also sometimes override individual autonomy over reproduction in order to uphold societal moral boundaries—for example, barring sterilization below a certain age or prohibiting certain forms of contraception.

2  Health and Medicine On both sides of the Atlantic, the regulation of contraception, abortion, and pregnant women emphasizes physician’s professional assessments. As with most medical care, 2  Ronald Dworkin, Life’s Dominion, 166–167 (1993).

Introduction to the Right to Avoid Procreation   1029 physicians serve as gate-­keepers to access reproductive care. Concerns about medical ethics shape legal decisions about when and whether to protect the right not to procreate. With regard to abortion in particular, the legality of abortion may turn on a physician’s assessment. While some legal models allow physicians to perform abortions for any reason up to a certain point in fetal development, others limit abortion to a specified list of reasons. Under the latter model, physicians often have the authority to formally decide whether an abortion is legally permissible even where nonmedical reasons are required, such as the pregnancy being the result of a criminal offence.

3 Equality Laws regulating abortion, pregnancy, and much of contraceptive use inherently have sex-­specific effects. While male and female people seek to avoid pregnancy, the risk of pregnancy falls only on those who are biologically female. Because of the role of women and the female body in procreation, the right not to procreate becomes especially significant for women (and transgendered men) in a way that does not apply to cisgendered men. If, for example, the woman’s partner or husband prefers a pregnancy to be either terminated or continued contrary to her wishes, his autonomy may not be exercised without it posing an intervention into her bodily integrity. As such, the rights of the woman carry more weight in the legal balance than any competing rights of the man. Unlike a liberty doctrine, a sex equality doctrine also requires examining the way in which restrictions on the right not to reproduce reinforce stereotypical gender roles and subordinate women. The UN Convention on the Elimination of All Forms of Discrimination of Women (CEDAW), for example, takes the perspective that a woman’s biological and social role in procreation should not affect her options and possibilities to take part in society. State regulation of access to reproductive choices thus becomes a question of nondiscrimination. Focusing on gender role stereotypes also helps surface the “referendum on the place and meaning of motherhood.”3

4  Family Life Versus Private Life Whereas Roe saw pregnancy in the first trimester as pertaining to a woman’s private life and therefore off-­limits for government interference, the European human rights system in its first ever decision on abortion4 found the opposite to be true: “the claim to respect for private life is automatically reduced to the extent that the individual himself brings his private life into contact with public life or into close connection with other 3  Kristin Luker, Abortion and the Politics of Motherhood 193, 118, 159 (1984). 4  Rose Marie Brüggemann & Adelheid Scheuten v. the Federal Republic of Germany, App. no. 6959/75, Report adopted by the Commission on 12 July 1977.

1030   Janne Rothmar Herrmann and Elizabeth Sepper protected interests . . . pregnancy cannot be said to pertain uniquely to the sphere of private life. Whenever a woman is pregnant her private life becomes closely connected with the developing foetus.” Consequently, regulation of pregnancy and abortion did not necessarily infringe on a woman’s privacy interests. This corresponds with the observation that European jurisprudence is more focused on family formation and family life (seen, for example, in the Irish constitutional protection of the family, which became the legal foundation for legalizing contraception), compared to the US focus on privacy. Consequently, reason-­giving is a focus in Europe, where a number of countries require a woman to explain her reasons for seeking abortion. Previously, women in the United States were under no such compulsion to give reasons for seeking an abortion before fetal viability, but recently some states have barred, for example, sex-­selective abortions, thus limiting appropriate reasons for abortion. Likewise, financing regimes limited to abortions for pregnancies resulting from rape or incest could be thought of as demanding reasons.

5  Negative Versus Positive Rights Some view the right to reproductive autonomy as merely a negative right to be free from State interference in the reproductive sphere while others regard it as a positive right to access medically assisted reproductive care. In international human rights law, Article 12 of CEDAW presents the most clearly worded protection of reproductive autonomy and choice. States party to CEDAW, which include all Member States of the European Union but not the United States, are divided by national perception, religious practice, and culture over the extent to which reproductive choice is protected under international law: some parties to the convention are authoritarian and do not accept equality between the sexes, others may formally recognize the equal worth and equal rights of women and men but accept de facto discrimination because of tradition, religion, and culture. Whereas regulation of pregnant women is increasingly framed in terms of positive rights (to workplace protections, antenatal healthcare, and autonomy in childbirth) on both sides of the Atlantic, there is no clear picture of the right not to procreate. Denmark frames abortion as a positive right, entitling women to receive the procedure free of charge and imposing a corresponding duty on the healthcare-­providing authorities to staff hospitals accordingly. The United States, by contrast, frames abortion and contraception as a right to be free from state interference. US legal doctrine does not give women a right to effective access but rather depends on insurance coverage and physician availability. In a similar vein, under European human rights law, individuals have no right to receive a particular medical service. Thus Italy, for example, protects a legal right to abortion without a corresponding duty of healthcare providers to perform the procedure. And a number of countries have liberalized access to emergency contraception but do not cover its costs through national insurance.

Introduction to the Right to Avoid Procreation   1031

6  Fetal Life The weight assigned to fetal life varies widely across legal and moral viewpoints about abortion. Some assign the fetus legal rights or interests equal to those of pregnant woman. While they often purport to create equality between woman and fetus, as the case of Ireland shows, codification of fetal rights tends to vanquish the rights of the pregnant woman. Other legal regimes permit abortion to protect the woman’s life and sometimes health under the rationale of self-­defense. Judith Thomson famously argued that rather than violate the fetus’s right to life, abortion deprives the fetus of the nonconsensual use of the pregnant woman’s body. Like parents of young children, the pregnant woman has no duty to rescue. Many weigh fetal life more heavily as a pregnancy advances, while others insist that a woman always has the right to protect her bodily integrity. In some countries, religious views about the moral status of the fetus guide legal decision-­making about reproductive healthcare more broadly. In many European countries and the United States, healthcare providers and institutions religiously opposed to contraception, ster­i­li­ za­tion, and abortion sometimes impede women’s access. Outside of abortion, the legal concern for fetal life can be fraught where the regulation of pregnant women is involved. With regard to workplace protections, the legal environment fosters both autonomy and equality, and fetal protection is afforded through protection of the pregnant woman. In this case, regarding the pregnant woman and fetus as equal is uncontroversial as the interests of the woman and the fetus do not collide. In other contexts, pregnant women who have decided to carry their pregnancies to term have found their liberty limited in the interest of the fetus. In both Norway and the United States, for example, the perceived conflict between mother and fetus has justified criminalizing pregnant women’s drug use (US) and detaining pregnant women who use drugs to protect fetal health and life (Norway). While the following chapters by Herrmann and Sepper (this volume) demonstrate some expected patterns in how countries weigh these varying and competing interests, our comparative approach also unveils some surprises. For example, some European countries require substantially longer waiting periods for abortion than US states. With the adoption of the Affordable Care Act, emergency contraception is also better covered by insurance in the United States than in many European states. Our chapters also complicate the superficially appealing argument advanced by some US conservatives that European states’ law is more restrictive than the US framework and provides a model for the United States. Many European countries at first appear more restrictive because, while they have liberal abortion laws for the first trimester, they otherwise restrict access except in cases of risk to health or life. Our comparison, however, highlights that, by contrast to the United States, those European countries offer insurance coverage and healthcare provider availability that makes the abortion right meaningful in the relatively short period of time between when a woman discovers she is pregnant and the first trimester ends.

chapter 51

The R ight to Avoid Procr eation a n d the R egu l ation of Pr egna ncy A European Perspective Janne Rothmar Herrmann

1  The Right to Avoid Procreation As already touched on in the common introduction to these chapters, the right to avoid procreation balances a number of legal principles, including the right to freedom/autonomy, the right to health, and the right to private life with rights or interests that the foetus might have, together with the State’s interests in reproductive decisions. The legal basis for a right to avoid procreation can be said to fall within the scope of several provisions of the European Convention on Human Rights (ECHR), an instrument that is binding for all European countries. Article 12 of the ECHR gives men and women of marriageable age the right to marry and found a family in accordance with the national laws governing this right. It follows from the wording that the right is relatively weak since its conditions are subject to national law (as long as national law does not restrict the essence of the right). The reason for adding the right to found a family in the European human rights catalogue relates to the era in which the human rights were first formulated and adopted; the 1948 UN Universal Declaration of Human Rights had similarly included a right to marry and found a family without any limitation due to race, nationality, or religion. The right was

1034   Janne Rothmar Herrmann very much a response1 to the prejudiced reproductive policies that characterized the Nazi regime but which had also been practiced elsewhere to various degrees in the decade leading up to World War II.2 These had prevented procreation by the disfavoured and installed a sense of obligation of the favoured to reproduce. Securing individual freedom to decide on whether or not to have children without undue influence from the State was thus a paramount concern. Consequently, the right to marry and found a family enshrined in both the Universal Declaration of Human Rights and the ECHR was first and foremost concerned with securing an individual right to decide whether or not to have children. There is no test case to demonstrate the exact scope of the right not to procreate inherent in Article 12. There are, however, a number of cases that demonstrate that the interpretation of the Article is not subject to the same dynamic interpretation often applied by the Court on other subject matters in order to make the Convention a “living instrument,” rather than being focused on the original meaning/will of the Convention signatories and on whether or not there is common European consensus on the matter in question. While there have been small developments on the right to procreate in the case of Christine Goodwin v UK,3 the liberal tone of this case stands relatively alone in the overall case law as conservative doctrines continue to dominate the interpretation of Article 12.4 Whereas the European Court of Human Rights (ECtHR) has expressly rejected the notion that the right to marry and found a family includes a positive right to reproduce,5 it must follow from the wording of the Article as a typical negative right that the State cannot force couples to have children (e.g., by prohibiting contraceptives). As such it is a typical negative right protecting individuals against government interference. In conclusion, Article 12 protects some elements of the right not to procreate but for couples only. The lack of common European consensus in this area is indicative of how matters relating to the right to decide on the number and spacing of children touch on aspects that differ from country to country even in what could appear to be a homogenous region. In fact, the cultural, moral, and historical milieus that surround these rights differ considerably with diverse national perceptions of the role of the family, gender equality, religious and moral obligations, and so on.

1  Collected Edition of the ‘Travaux Preparatoires’ of the European Convention on Human Rights: Volume I: Preparatory Commission of the Council of Europe, Committee of Ministers, Consultative Assembly (11 May–13 July 1949) (Brill Nijhoff 1975) 40–46, 56. 2  Lene Koch, “Eugenic Sterilisation in Scandinavia” (2006) 11 Eur. Legacy 299; Michelle Mouton and German Historical Institute, From Nurturing the Nation to Purifying the Volk, Weimar and Nazi Family Policy, 1918–1945 (Cambridge University Press 2007) ch 3. 3  Christine Goodwin v United Kingdom App no 28957/95 [ECHR, 11 July 2002] §97–104. 4  Bart Van der Sloot, Between Fact and Fiction: An Analysis of the Case-­Law on Article 12 of the European Convention on Human Rights (2014) 26 Child Family L Q 397. 5  S.H. & Others v Austria App no 57813/00 [ECHR, 3 November 2011] and Sijakova and others v the former Yugoslav Republic Macedonia App no 67914/01 [ECHR 6 March 2003].

The Right to Avoid Procreation: A European Perspective   1035

1.1 Abortion Article 8 of the ECHR provides a second basis for a right not to procreate. Article 8 ­recognizes a right to respect for an individual’s private and family life, his home and his correspondence subject to restrictions that are in accordance with law and necessary in a democratic society that has other important interests at stake. These include national security, public safety, the economic well-­being of the country, prevention of disorder or crime, protection of health or morals, and protection of the rights and freedoms of others. According to the ECtHR case law, the notion of “private life” within the meaning of Article 8 of the Convention is a broad concept which encompasses, inter alia, the right to personal autonomy and personal development.6 It concerns such aspects of private life as gender identification, sexual orientation and sexual life,7 and a person’s physical and psychological integrity,8 as well as decisions both to have and not to have a child or to become genetic parents.9 Over the years a substantial number of cases dealing with abortion have been heard by the ECtHR (and the previous European Commission of Human Rights, which heard some of the first cases on abortion as the first instance body at that time). However, the Court has avoided deciding if the ECHR gives European women a right to abortion.10 In the 1980 case X v UK (Paton v UK),11 the European Commission on Human Rights identified three possible views of the foetus’s right to life and laid down some basic principles in the balancing of interests between woman and foetus. In this case, a husband had unsuccessfully tried to get an injunction against his wife undergoing a termination. The case had been dismissed by the national court on the grounds that no rights stood to be violated by the proposed termination. Neither the foetus nor the husband held any rights under English law, and, consequently, the woman terminated the pregnancy a few hours after the court hearing. Mr. Paton subsequently complained to the European Commission on Human Rights claiming that the English Abortion Act violated the ECHR both by allowing abortion, by refusing the foetus any rights, and by denying the father his rights over the foetus. In terms of whether or not the foetus had any rights, the Commission outlined three possible approaches, relying on the right to life in Article 2 of the ECHR. First, the Article could be interpreted in a way which excluded the foetus completely. Another 6  Pretty v the United Kingdom App no 2346/02 [ECHR, 29 April 2002] § 61. 7  Dudgeon v the United Kingdom App no 7525/76 [ECHR, 22 October 1981], § 41, and Laskey, Jaggard and Brown v the United Kingdom App no 21627/93; 21628/93; 21974/93 [ECHR, 19 February 1997] § 36. 8  Tysiac v Poland App no 5410/03 [ECHR, 20 March 2007], § 107. 9  Evans v the United Kingdom App no 6339/05 [ECHR, 10 April 2007], § 71. 10  Open Doors and Dublin Well Women v Ireland App no 14235/88 [ECHR, 29 October 1992] established that there is a right to receive and impart information about abortion. 11  X v the United Kingdom App no 8416/79 [ECmHR, 13 May 1980]. The underlying national case was Paton v British Pregnancy Advisory Service Trustees [1979] QB 276; [1978] 3 WLR 687; [1978] 2 All ER 987 and, as such, the case is often referred to as Paton v UK.

1036   Janne Rothmar Herrmann possible interpretation was to understand Article 2 as including the foetus in its scope but with “certain implied limitations.” Last, the protection laid down in Article 2 could be understood as awarding the foetus an absolute right to life. The last option was dismissed by the Commission as this would render abortion impossible even in emergency cases, where a termination of the pregnancy is necessary in order to save the woman’s life, stating that to place a higher value on the potential life of the unborn than that of the woman would be contrary to the ECHR’s purpose. In reaching this conclusion the Commission relied on the argument that all of the contracting States, except only one, allowed for abortion under such circumstances at the time of drafting the Convention. However, the foetus’s life was intimately connected to and could not be viewed separately from the woman’s life. Thus, the Commission dismissed only one of three identified possible approaches but did not take an explicit stand on the remaining two perspectives. However, the dismissal of the view that the foetus holds an absolute right to life implicates that national regulations must include a right to abortion when the woman’s life is at stake in order to respect the right to life as guaranteed in Article 2 of the Convention.12 In the case Tysiac v Poland (2007),13 a pregnant woman risked losing her eyesight if she carried the pregnancy to term. Her general practitioner advised an abortion and referred her to a gynecologist. The gynecologist, however, did not find that she met the criteria for a termination. Consequently, she gave birth to a child and subsequently became blind. The woman claimed that the state’s failure to provide her with an abortion in this situation, where such a procedure was actually legal under national law, amounted to a violation of her Article 8 right to respect for private life as she was exposed to a considerable risk to her health. The European Court of Human Rights agreed and found that the state had a positive obligation to protect her right effectively. Thus, it was not sufficient that an abortion was in concreto legal: the state was also obliged to provide her with an effective judicial remedy which could have vindicated her right, especially so when she was referred to a specialist by her general practitioner who found that she met the criteria for having the pregnancy terminated but the specialist declined. In conclusion, while Paton v UK stipulated that national regulations must provide women with abortion when it is necessary to save the woman’s life in order to comply with the protection in ECHR Article 2, Tysiac v Poland afforded equal protection of the woman’s health as part of her right to respect for private life under Article 8. Whereas the general issue of abortion had been a question of balancing the right to privacy vis-­à-­vis the state’s legitimate interest in safeguarding public morality, the issue of therapeutic abortion also had to address the state’s positive obligation to safeguard women’s physical integrity. Although the Court emphasised that it was not for the Court in the present case to assess whether or not the ECHR guaranteed a right to abortion because Tysiac already had such a right under national law in case of serious risk to her life or her health, the 12  If follows from this analysis that Malta’s absolute prohibition on abortion is highly likely to be considered a violation of the ECHR if a case were to make it to the Court. 13  Tysiac v Poland App no 5410/03 [ECHR, 20 March 2007].

The Right to Avoid Procreation: A European Perspective   1037 judgment does nonetheless appear to be a first step towards a more dynamic interpretation of the Convention in relation to the sensitive issue of abortion. Notwithstanding that the Convention “does not guarantee as such a right to any specific level of medical care,”14 the Court stressed that the concept of private life had been interpreted in the case law to encompass physical and mental integrity and that States had a positive obligation to ensure that individuals were given an effective respect of their integrity.15 The view that a real and effective protection of women’s integrity implicitly implicates access to therapeutic abortion could be the first step towards the creation of greater legal certainty in relation to abortion in European human rights law.16 Every member state of the European Union (EU) is a party to the ECHR; however, as established according to the case law, there is a wide margin of appreciation in many of the matters relevant to this chapter. Consequently, the legal regulation of abortion in Europe differs considerably. The member states of the EU must, however, respect the effect of the EU treaties even on matters relating to the national domain in terms of their cross-­border elements. As a result, European women may travel freely, seeking out reproductive care and services that are legal in the country that provides them,17 and, consequently, there is considerable reproductive mobility within Europe.18 At the national level, regulation of abortion can largely be structured around the following models: On-­demand model: The Scandinavian countries (except the Pharoe Islands) and Italy are examples of legal frameworks that include on-­demand abortion. Women are entitled to have a termination on request, and they are not required to give particular reasons or meet certain criteria. This also entails that, in its most liberal form, there is no mandatory waiting period and no mandatory counselling. An abortion is covered by the National Health Plan and thus provided free of charge. The cutoff limit for on-­demand abortion varies from country to country; Denmark, Italy, and Norway have set the limit at 12 weeks and Sweden at 16 weeks. After the cutoff limit, the nature of the legal model changes: abortion is no longer available on demand; however women can apply for an abortion and must meet certain criteria in the law

14  Tysiac v Poland App no 5410/03 [ECHR, 20 March 2007] para 107. 15  See further Jill Marshall, Personal Freedom Through Human Rights Law?: Autonomy, Identity and Integrity Under the European Convention on Human Rights (Brill Nijhoff 2008) ch 10. 16  See also Brigit Toebes and others (eds.), Health and Human Rights in Europe (1st edn, Intersentia 2012) ch 5. However, opinions critical of abortion have also been given in various cases (Judge Borrego Borrego’s dissent in Tysiac v Poland App no 5410/03 [ECHR, 20 March 2007] and Judge Ress’s concurring opinion joined by Judge Kuris in Odievre v France App no 42326/98 [ECHR, 13 February 2003]). 17  Grogan v the Society for the Protection of Unborn Children C-­159/90 [ECJ, 4 October 1991]. 18  See, for example, Guido Pennings, “Legal Harmonization and Reproductive Tourism in Europe” (2014) 19 Human Reproduction 2689; and Charlotte Kroløkke, “Destination Spain: Negotiating Nationality and Fertility when Traveling for Eggs” in M Lie and N Lykke (eds), Assisted Reproduction Across Borders: Feminist Perspectives on Normalizations, Disruptions and Transmissions (Routledge 2016) 61–71.

1038   Janne Rothmar Herrmann through, for example, medical, social, eugenic, or ethical grounds. After the cutoff limit, this legal model is effectively similar to the “illegal-­unless model” discussed later. In the following, the Danish and Italian legal models are used to illustrate potential barriers to the effective fulfillment of a right not to procreate in a relatively liberal model of regulation. The Italian Act 194/1978 legalized first-­trimester abortions, but it does not necessarily provide women with an effective right since it is not a right with a corresponding duty on the healthcare system to provide the abortion. Furthermore, if the termination is not considered urgent by the physician, there is a mandatory 7-­day waiting period, after which the woman is entitled to have the termination. In practice, abortions are very hard to obtain in Italy with approximately 70% of gynecologists being conscientious objectors. The Danish Health Act’s chapter on abortion also includes the right of healthcare staff to invoke a conscientious objection; however, since abortion is a legal right, the Danish regions that are the providers of healthcare are obliged to staff hospitals accordingly. The travaux préparatoires explicitly state that a conscientious objector cannot be the sole physician in remote or rural areas as this would violate women’s legal right to access abortion. Both the Italian and Danish Acts equally aim at protecting women from regarding abortion as the only solution. The Italian Act directs family counseling centers to assist a pregnant woman by informing her of her rights under State legislation and of the social, health, and welfare services available to her; how to take advantage of the health, safety, and labour legislation in place to protect her; and in general help overcome the factors which might lead the woman to have her pregnancy terminated. The Danish Act requires the physician to inform the woman of the possibility of getting information from the Regional Council on available support to go through with the pregnancy and raise the child. But, from a public health perspective, the Acts are also concerned that abortions only take place in medical facilities/are performed by physicians. Both Acts centre the protection of reproductive autonomy as one of the most basic principles at play in policy-­ making. However, acknowledging that abortions take place irrespective of their legality, protection of women’s health is also a major concern in the liberalization of abortion, ensuring proper medical care to replace the clandestine back-­alley procedures that, prior to liberalization, had been known to put women’s lives at risk. After the cutoff limit, regulation is usually in place to allow abortion in certain instances. Under the Italian Act, a woman may have an abortion during the first 90 days after the 12-­week limit if a physician at a hospital finds that she meets the requirements (her life or physical or mental health is at risk, included those associated with the malformation or serious abnormalities of the foetus). The Danish Act sets up a regional Abortion Council, which decides on application if the woman meets the Act’s indications for receiving authorization for a late abortion (medical, eugenic, ethical, and social indications). Viability in the foetus is set as the upper limit, and abortion can only be authorized after the point of viability if the foetus is seriously ill and likely to survive only for a very short time after birth. In case of risk to maternal life, the physician is always entitled to perform an abortion, without authorization from the Council and even in case of foetal viability.

The Right to Avoid Procreation: A European Perspective   1039 The on-­demand model frames abortion as legal under the framework in the law, and criminal sanctions only apply (for physicians) in cases where abortions are performed outside the provisions in the Act. The model reflects an underlying ethical gradualist view of foetal status where the foetus acquires increasing status concurrent with gestational age. Illegal-­unless model: Countries such as England and the Netherlands have a model of regulation where abortion is conceptualized as illegal unless certain criteria are met. In the Netherlands women have a mandatory waiting period of 5 days between the initial consultation and the procedure. After the 13th week and until the 24th week, the woman has to be in a state of distress to obtain an abortion. Abortion is allowed in England, Wales, and Scotland to save a woman’s life, for health, or for economic or social reasons. Two registered medical practitioners must certify that the required medical grounds have been met. In practice, these criteria are interpreted broadly, but nevertheless a woman is not entitled to an abortion on request, since 1–2 medical professionals must certify that she meets the (broad) criteria. As such, some explanation that fits the criteria must be provided as to why she wishes a termination. The legal model typically categorizes abortion as a criminal offence for the medical professional, and the legality of the termination for the physician is dependent on the doctor assessing that the woman meets the access criteria with a second doctor agreeing on that assessment. The model usually comes with an extended period (24 weeks) in which abortions can be performed compared to the on-­demand model. The legality of the abortion can also depend on other conditions that must be met in order for an abortion to be legal. Germany allows abortion in the first trimester, but not directly on demand as the legality is conditional on mandatory counselling (even if the woman does not wish to receive counselling) and a mandatory waiting period of 3 days. After the first trimester, abortion can be obtained on certain indications, such as medical grounds. The model reflects less emphasis on reproductive autonomy, placing abortion as a medical procedure that requires professional assessment, even though the grounds that can lead to the legality of an abortion are just as much social and economic in nature as they are medical. The fact that the model often requires a second physician to agree on the medical assessment further underlines that it is just as much a societal balancing between opposing values and interests that the legislator has left in the hands of the medical profession as it is unusual for the law to require a second opinion to make the provision of a medical service legal. Strictly illegal model: Poland and Malta have very strict regulation of abortion. Ireland also belonged to this model until 2019 when the law was liberalized following a referendum which repealed the 8th amendement to the Irish Constitution, thereby removing the constitutional protection of the foetus. After judicial decisions in 197319 by the Irish Supreme Court that married couples enjoyed a right to private life which included the use of contraceptives for family planning despite regulation prohibiting contraceptives and by the US Supreme Court on 19  McGee v The Attorney General [1974] IR 284.

1040   Janne Rothmar Herrmann abortion, it seemed only a matter of time before the Irish Court would recognize a right to abortion. Accordingly, pro-­life activists secured protection for the foetus via the 8th Amendment. Before the referendum repealing the 8th Amendment, a model similar to that of Denmark was outlined in general terms as the framework to be put forward as a Bill in the Irish Parliament following repeal. Typically, the strictly illegal model grants equal rights to protection of the foetus and woman, making abortion legal only in cases where the mother’s life (or in the case of Poland life and health) is at severe risk. It reflects the often religiously motivated moral view that the foetus has full moral status from the moment of conception, effectively making abortion a question of the foetus’s right to life vis-­à-­vis the woman’s right to control her own body. Malta is the only country in the EU that has an absolute prohibition of abortion, with no legal exceptions. According to chapter 9 of the Criminal Code of 1854, abortion whether self-­administered or medically provided is a criminal offence. Placing the right to life of the unborn absolutely before the right to life of the woman, as Malta does, is a historical construct of the Criminal Code of 1854, and there has been very little political appetite for putting abortion on the agenda. Perhaps motivated by the country’s predominant Catholic faith and the presence of a pro-­life lobby (with no pro-­choice lobby), all political parties have agreed on the abortion issue. The stance has, however, not been uncontroversial from an international point of view. In 2004, the United Nations Committee on Economic, Social and Cultural Rights, set up to monitor the International Covenant on Economic, Social and Cultural Rights (CESCR), urged Malta to review its legislation on abortion and consider exceptions to the general prohibition of abortion for cases of therapeutic abortion and when the pregnancy is the result of rape or incest.20 The Council of Europe in 2008 declared that abortion should be legal and accessible for all women in Europe and invited member states to decriminalise abortion if they have not already done so and lift restrictions which ­hinder, de jure or de facto, access to safe abortion, observing that the ultimate decision on whether or not to have an abortion should be a matter for the woman concerned, who should have the means of exercising this right in an effective way.21 European women resident in countries like Malta (or who struggle to get access to abortion because of widespread use of conscientious objections from healthcare providers) are entitled to travel cross-­border within the EU.22 However, availability abroad is subject to self-­payment.

20  UN Committee on Economic, Social and Cultural Rights (CESCR), UN Committee on Economic, Social and Cultural Rights: Concluding Observations, Malta, 14 December 2004, E/C.12/1/ Add.101. http://www.refworld.org/docid/42d26f994.html; accessed 16 July 2018. 21  Council of Europe: Parliamentary Assembly. “Resolution 1607 (2008) on Access to Safe and Legal Abortion in Europe” 16 April 2008. 22  Society for the Protection of Unborn Children v Stephen Grogan, C-­159/90 [1991], 4 October 1991.

The Right to Avoid Procreation: A European Perspective   1041

1.2 Contraceptives In Europe, birth control pills that prevent ovulation are legal in all countries; as ­mentioned previously, in Ireland, the contraceptive pill was first legalized for married couples following a Supreme Court judgment, and Ireland has since allowed the general sale of contraceptives since 1985. In Malta, a single brand of contraceptive pill is available to purchase that prevents ovulation. In the EU, the emergency pill is generally available over the counter with no restriction on purchase, Malta being the only country where the government refuses official import. A few countries require women to be over a certain age to not require parental consent (young women under the age of 14 years in Germany and 16 years in Romania must have parental consent to purchase the emergency pill over the counter) or to not require a prescription (young women under the age of 16 years in the UK and under the age of 18 years in Italy must have a prescription). Some countries provide emergency pills free of charge when prescribed or to very young women or if dispensed in certain clinics (UK, Sweden), some require payment out of pocket (Norway, Denmark, Italy), and some have partial reimbursement (France, Spain).

1.3 Sterilization Article 8 of the ECHR’s respect for private and family life also applies to sterilization procedures. Sterilization constitutes a major interference with a person’s reproductive health status. As it concerns one of the essential bodily functions of human beings, it bears on manifold aspects of the individual’s personal integrity, including his or her physical and mental well-­being and emotional, spiritual, and family life.23 It may legitimately be performed at the request of the person concerned, for example as a method of contraception, or for therapeutic purposes where the medical necessity has been convincingly established. Sterilization of women without their free and informed consent is a violation of Article 8.24 There is no case law to demonstrate whether Article 8 entails a positive right to access sterilization as a means of contraception. There is, however, case law establishing that the decision not to procreate falls under the ambit of Article 8, and, in the case Evans v UK, the right not to procreate prevailed over the right to procreate.25 Most European countries allow sterilization for contraceptive purposes; however, some countries (such as Sweden and Norway) have in place a qualifying minimum age limit of 25 years, which is higher than the general age of legal competency; others such as 23  V.C. v Slovakia App no 18968/07 [ECHR, 8 November 2011] § 106. 24  IG and others v Slovakia App no 15966 [ECHR, 13 November 2012] and G.B. and R.B. v the Republic of Moldova App no 16761/09 [ECHR, 18 December 2012]. Sterilization was considered a violation of Articles 3 and 8 in N.B. v Slovakia App no 29518/10 [ECHR, 12 June 2012]. 25  Evans v the United Kingdom App no 6339/05 [ECHR, 10 April 2007]. After a woman was diagnosed with ovarian cancer, she underwent IVF with her fiancé to create embryos for future use. Subsequently, they split up, and he objected to her use of the embryos to have children. The Court gave priority to the man’s right not to procreate.

1042   Janne Rothmar Herrmann Finland require the person to be at least 30 years old or already a parent to three children. France requires a waiting period of 4 months, whereas Romania makes counselling mandatory and requires notification to both spouses concerning the permanence of sterilization.26 Some countries have unclear regulation on sterilization for contraceptive purposes; in Malta sterilization for contraceptive purposes used to be illegal, but the changed regulation is unclear on the present legality.27 Among those countries that allow sterilization for contraceptive purposes the legal stance is relatively homogenous; reproductive autonomy weighs heavily, while also acknowledging that reproductive desires are fundamental in human nature and can therefore change with age and that the procedure is irreversible, thus requiring that the wish for sterilization be an authentic and mature decision.

1.4  Healthcare Provider Conscience Article 9 of the ECHR is relevant to touch upon in terms of exploring to what extent conscientious objections in reproductive healthcare can limit women’s right not to procreate, in terms of her access to contraceptives or abortion. In Pichon v France28 two pharmacists claimed that their conviction by the French authorities for refusing to dispense oral contraceptives amounted to a violation of their freedom to manifest their religion. The ECtHR declared the case inadmissible, concluding that the pharmacists could not give priority to their personal religious beliefs over their professional obligations “as long as the sale of contraceptive is legal and occurs on medical prescription nowhere other than in a pharmacy.”29 In RR v Poland30 a prenatal genetic examination was denied partly due to a conscientious objection. The ECtHR found that the “shabby treatment” and “acute anguish” it caused the applicant violated her right to be free from inhumane and degrading treatment (Article 3).31 The case marked the turning of a corner for the Court as it found a violation of Article 3 for the very first time in a reproductive rights case.32 The Court also made clear that access to diagnostic services was decisive for the “possibility of exercising her right to take an informed decision as to whether to seek an abortion or not” and noted the crucial importance of timely access to information on one’s health condition by stating that “in the context of pregnancy, the effective access to relevant information 26  EU Member States’ Positions on Sexual and Reproductive Rights Issues, The Family Federation of Finland—Global Development Unit 2.4.2012. http://vaestoliitto-­fi-­bin.directo.fi/@Bin/392059ac6841d17 8e36d1b46f8e793f0/1516718136/application/pdf/1740298/EU%20Country%20Positions%20on%20 Sexual%20and%20Reproductive%20Rights%20Issues.pdf; accessed 16 July 2018. 27  RG Milne and RE Wright, “The Decline of Fertility in Malta: The Role of Family Planning” (1997) 13 Eur J Population [Revue Europeenne De Demographie] 147. 28  Pichon v France App no 49853/99 [ECHR, 2 October 2001]. 29  Para 4. 30  R.R. v Poland App no 27617/04 [ECHR, 26 May 2011]. 31  Paras 15 and 159–162. 32  See more in depth, Ximena Andión-­Ibañez and Christina Zampas, “Conscientious Objection to Sexual and Reproductive Health Services: International Human Rights Standards and European Law and Practice” (2012) 19 Eur J Health L 231.

The Right to Avoid Procreation: A European Perspective   1043 on the mother’s and foetus’ health, where legislation allows for abortion in certain situations, is directly relevant for the exercise of personal autonomy.”33 There is no European consensus at the national level in the area of conscientious objection in relation to the dispensing of contraceptives and the provision of abortion. Many countries do not allow conscientious objections in the dispensing of contraceptives (Czech Republic, Finland, Denmark) and many do (Romania, Poland, UK, Ireland). Most European countries allow health staff to invoke a conscientious ­objection, whereas Sweden, Finland, Bulgaria, Iceland, and the Czech Republic do not.34 Some countries have been criticized for insufficient regulation on conscientious objection. In 2008, a Committee of the Convention on the Elimination of all Forms of Discrimination Against Women (CEDAW) considered conditions in Slovakia. In its Concluding Observations to Slovakia, the Committee noted that it “is deeply concerned about the insufficient regulation of the exercise of conscientious objection by health professionals with regard to sexual and reproductive health” and recommended that Slovakia “adequately regulate the invocation of conscientious objection by health professionals so as to ensure that women’s access to health and reproductive health is not limited.” The Committee recalled its “general recommendation No. 24, which states that it is discriminatory for a State party to refuse to provide legally for the performance of certain reproductive health services for women” and recommended “that, if health serv­ice providers refuse to perform such services based on conscientious objection, measures should be introduced to ensure that women are referred to alternative health providers.35 Healthcare provider conscience remains one of the most prevalent barrier to access in Europe.36

2  The Regulation of Pregnant Women The legal regulation of pregnant women at the national level touches on a number of fundamental rights. Differences in opinion are held throughout Europe as to the moral and legal status of the foetus and the embryo. The Council of Europe’s Convention on Biomedicine and Human Rights (the 1997 Oviedo Convention) as a result leaves the term “everyone” undefined. Whether or not the foetus has a legal status which limits the rights of the pregnant woman is consequently a question left to national discretion. As 33  Paras 197 and 208. 34  Anna Heino and others, “Conscientious Objection and Induced Abortion in Europe” (2013) 18 Eur J Contraception & Reproductive Health Care 231. 35  CEDAW Committee on the Elimination of Discrimination Against Women, Concluding Observations: Slovakia, paras 42–43, U.N. Doc. A/63/38 (2008), paras 42–43. 36  Unlike in the US, where poverty, minority, and social stigma are highlighted as barriers to access, the European concerns in relation to access are articulated as gestational age limits, mandatory waiting period, mandatory counselling, and conscientious objections. See, for example, the ERC research project Europe Abortion Access (Principal Investigator Silvia de Zordo). https:// europeabortionaccessproject.org/.

1044   Janne Rothmar Herrmann previously mentioned, the Paton case outlined a number of ways the relationship between woman and foetus could be understood but only decided on the relatively ­noncontroversial issue of how the relationship should be balanced in case of risk of maternal death. Even in a case where the ECtHR could have considered foetal rights without collision with maternal rights, the Court did not resolve the issue of foetal legal status. In Vo v France37 due to a mix-­up with another patient with the same surname, the applicant’s amniotic sack was punctured, making a therapeutic abortion necessary. She maintained that the unintentional killing of her child should have been classified as manslaughter. The Court held that there had been no violation of Article 2’s right to life. It found that it was not currently desirable or possible to rule on whether an unborn child was a person under Article 2 of the Convention. There was no need for a criminal law remedy since remedies already existed allowing the applicant to prove medical negligence and to seek compensation through her own legal standing. Other aspects of pregnancy have fared more successfully in the ECtHR, however still stressing considerable discretion for the State. The question of home birth has been tried several times. In Duská & Kreizová v the Czech Republic38 the Grand Chamber held that a ban on health professionals attending home births did not violate Article 8. It found in particular that the national authorities had considerable room for maneuver when regulating the question of home births—a matter for which there was no European consensus and which involved complex issues of healthcare policy as well as allocation of State resources. In the applicants’ case, the Grand Chamber considered that the Czech Republic’s policy struck a fair balance between, on the one hand, mothers’ right to respect for their private life and, on the other, the State’s interest in protecting the health and safety of the child and mother during and after delivery. Moreover, since 2014, the Czech government had taken some initiatives with a view to improving the situation in local maternity hospitals, notably by setting up a new governmental expert committee on obstetrics, midwifery, and related women’s rights. Last, the Grand Chamber invited the Czech authorities to make further progress by continuing their constant review of the relevant legal provisions on home births, making sure that they reflect medical and scientific developments whilst fully respecting women’s rights in the field of reproductive rights. However, if a State has legally provided the opportunity to choose home birth, access to home birth must be effective. In Ternovsky v Hungary39 the Court found that the applicant was in effect not free to choose to give birth at home because of the permanent threat of prosecution faced by health professionals which effectively dissuaded them from assisting her during a home birth and the absence of specific and comprehensive legislation on the subject, in violation of Article 8. 37  Vo v France App no 53924/00 [ECHR, 8 July 2004]. 38  Dubská & Krejzová v the Czech Republic App no 28859/11 and 28473/12 [ECHR, 15 November 2016]. 39  Ternovszky v Hungary App no 67545/09 [ECHR, 14 December 2010].

The Right to Avoid Procreation: A European Perspective   1045

2.1  Workplace Protection and Discrimination Whereas medical issues are not governed by EU law (except certain public health issues and free movement issues), the opposite is true for workplace health and safety as well as discrimination. A wide range of EU measures in the field of health and safety at work have been adopted on the basis of Article 153 of the Treaty on the Functioning of the EU. The measures have been adopted by means of Directives, which are legally binding and must be transposed into national legislation. The Directives in the health and safety at work area are all minimum harmonization directives, meaning that Member States are free to set stricter regulation in place. As a result, health and safety at work legislation can vary across the different European countries. Additionally, the European Court of Justice (ECJ), which has often been viewed in literature as a very activist court,40 has indeed pioneered a new approach where pregnancy is not just compared to illness but rather treated as a separate condition. In Dekker v Stichting Vormingscentrum voor Jonge Volwassenen Plus,41 an employer refused to employ a pregnant woman, claiming it could not afford to pay for maternity coverage. The ECJ found that because only women could be refused employment on the grounds of pregnancy, therefore the employer was directly discriminating on the grounds of sex, in violation of the Equal Treatment Directive.42 The 1992 EU Pregnancy Directive (92/85) sets out a number of health and safety rights for employees who are pregnant, breastfeeding, or have recently given birth. These health and safety rights of women pregnant, breastfeeding, or recently back from childbirth relate mainly to the employer’s mandatory assessment of chemical, physical, and biological agents; movements and postures; mental and physical fatigue; other types of physical and mental stress; and industrial processes considered hazardous for their safety or health.43 If a woman covered by the Directive is exposed to a health or safety risk or effect on the pregnancy/breastfeeding, the employer must take steps to ensure that the exposure of the woman to the risks is avoided. These steps include temporarily adjusting the working conditions and/or the working hours of the worker concerned and/or moving the employee to another job. Under the Directive, EU member states are also obliged to provide at least 14 weeks of maternity leave and protect against dismissal during it. They also must ensure that pregnant women are entitled to time off in accordance with national legislation and/or practice, without loss of pay, in order to attend antenatal examinations if such examinations have to take place during working hours. 40  Bruno de Witte, Elise Muir and Mark Dawson, Judicial Activism at the European Court of Justice, Causes, Responses and Solutions (Edward Elgar Publishing 2013). http://ep.fjernadgang.kb.dk/ login?url=http://kbdk.eblib.com/patron/FullRecord.aspx?p=1164208; accessed 16 July 2018. 41  Dekker v Stichting Vormingscentrum voor Jonge Volwassenen C-­177/88 [ECJ, 8 November 1990]. 42  See further Judith Bourne and Caroline Derry, Gender and the Law (Routledge 2018) 87–93. 43  The European Commission has issued guidelines to aid the assessment in COM/2000/0466. https://eur-­lex.europa.eu/legal-­content/EN/TXT/HTML/?uri=CELEX:52000DC0466&from=EN; accessed 10 July 2018.

1046   Janne Rothmar Herrmann Various national Acts give effect to the Directive, which may afford further health and safety rights to pregnant women. In countries such as Denmark, where employers and trade unions negotiate a number of employment rights and duties in collective agreements, further maternity leave, including salaried maternity leave, can also follow from such collective agreements rather than formal regulation.

2.2  Regulation of Pregnant Women’s Lifestyle Choices Foetal health is predominantly addressed through the provision of a preventive antenatal healthcare regime. Most European countries offer regular visits during pregnancy with a physician and/or midwife free of charge, including Iceland, France, Italy, Finland, Norway, Sweden, and Denmark. Preventive antenatal care ranges from information from national health authorities on appropriate lifestyle choices during pregnancy in terms of alcohol abstinence and regular exercise to, for example, free consultations with midwives, dieticians, smoking cessation counselors, etc. No European country specifically prohibits or criminalizes drinking alcohol during pregnancy, reflecting considerable emphasis on maternal autonomy throughout pregnancy and childbirth as well as very little legal recognition of foetal legal status in areas not concerning abortion. Generally, European national laws do not frame the maternal– foetal relationship in a conflict perspective, but rather tend to frame legal protection of the foetus as legal protection of the pregnant woman. Any rights of the foetus are derived through the rights of the woman. The UK Court of Appeal considered the possibility of criminalizing drinking alcohol during pregnancy in 2014 but ultimately decided against it: the case involved a girl who had been born with foetal alcohol syndrome as a result of her mother’s alcohol abuse whilst she was pregnant. Lord Justice Treacy stated that an “essential ingredient” for a crime to be committed “is the infliction of grievous bodily harm on a person—grievous bodily harm on a foetus will not suffice.”44 Predominantly, maternal autonomy is respected when it comes to lifestyle choices. Both Norway and Denmark have regulations in place allowing admission and detention on the basis of a previously expressed consent. Norway is the only European country to allow coercive deprivation of liberty of substance-­/alcohol-­abusing women for the purpose of foetal protection.45 The Municipal Health and Care Services Act46 lays down the rules on the detention of pregnant women who abuse drugs. Sections 10-­3 allow for the pregnant woman to be detained without her consent and taken to an institution designated by the regional health agency in cases where the drug abuse is of such a nature that the child is predominantly likely to be born with an injury. Studies show that the 44  CP v CICA [2014] EWCA Civ 1554. 45  Karl Harald Søvig, “Detention of Pregnant Women to Protect the Foetus: Nordic Perspectives” (2011) Nordic Health L Eur Context 158. 46  See Act no 30 of 24 June 2011 as amended by Act no 46 of 22 June 2012.

The Right to Avoid Procreation: A European Perspective   1047 detention usually takes place during the second trimester, in part because of the legal safeguards that must be observed, including the duty to consult and inform. Rooted in the ethical principle of nonmaleficence, the Norwegian welfare State found itself urged to act as public debate became framed by moral indignation and a focus on pregnant women finding themselves torn between the wish to protect the foetus and the failure to do so.47 Neighboring Scandinavian countries, equally welfare states, were more influenced by the question of the legal status of the woman vis-­à-­vis the foetus.48

2.3 Childbirth Generally, women in Europe do not have a statutory right to choose a caesarian section (c-­section). This reflects that, generally, there is no right to receive a particular medical service and that States—from a human rights point of view—have a large margin of appreciation in the organization of the healthcare system, where the level of public financing, risk of litigation, and medicalization of childbirth all play a role. There are wide variations in c-­section rates across Europe, indicating a lack of consensus about the best way of delivering children.49 Compared to World Health Organization (WHO) recommendations that 15% of births need a c-­section, around one in four births in the United Kingdom is by c-­section, while in Cyprus the figure is 52% and in Sweden 17%. In the UK, women’s decisions about childbirth are fiercely protected by the law. In St George’s NHS Trust v S (1997), the Court of Appeal made it plain that pregnant women retain the same autonomy over their bodies as any other person. They can refuse medical treatment regardless of the danger to themselves or their child. An exception applies for pregnant women who lack the mental capacity to make decisions about their medical treatment; they can be treated against their will under section 16 of the Mental Capacity Act 2005 as long as the treatment is in their “best interests.” This exception was applied in Re AA (Essex c-­section case) (2013). The court ordered a forced c-­section for a woman who was “seriously mentally ill and incapacitated” and who risked uterine rupture with a natural vaginal birth. The leading judge in the case decided that the woman lacked capacity within the terms of section 2(1) of the Mental Capacity Act to make a decision on the manner of birth of her child.

47  Kerstin Söderström and John-­Arne Skolbekken, “Pregnancy and Substance Use: The Norwegian z 10–13 Solution. Ethical and Clinical Reflections Related to Incarceration of Pregnant Women to Protect the Foetus from Harmful Substances” (2012) 29 Nordic Studies Alcohol Drugs 155. 48  Leppo A, “Illusionen om enkla lösningar [The Illusion of Simple Solutions]” (2009) 26(3) Nordic Studies Alcohol Drugs 314–317. 49  Macfarlane AJ et al., “Wide Differences in Mode of Delivery within Europe: Risk-­stratified Analyses of Aggregated Routine Data from the Euro-­Peristat Study” (2016) BJOG: An International Journal of Obstetrics & Gynaecology. Wiley Online Library. https://obgyn.onlinelibrary.wiley.com/doi/ full/10.1111/1471-­0528.13284; accessed 18 July 2018.

1048   Janne Rothmar Herrmann Other (civil law) jurisdictions have much less litigation in the medical area, and a forced c-­section would ultimately be a question of a criminal law case afterwards if charges were pressed. In, for example, Danish Criminal Law, it is debated if the foetus has legal status only from the point of being fully delivered or already from the point of mid-­delivery. It would, however, be unlikely that a doctor who had proceeded with a c-­section in spite of the woman’s wishes would be convicted of bodily harm because of the intent to save lives. Childbirth is free of charge in State hospitals in many European countries, including the UK, Spain, France, Germany, the Netherlands, and the Scandinavian countries. Home births are routine in the Netherlands in particular, but are also offered free of charge in countries like Denmark to patients who qualify from a medical point of view. Other countries do not offer the same scope for women to make their own childbirth choices50 as states currently enjoy a wide margin of appreciation in such regulation according to the ECtHR.51

3 Conclusion Whereas the right to have an abortion was established as a fundamental right derived from the US Constitution in Roe v Wade, the fundamental rights approach in the ECtHR has been hesitant and considerably less dynamic than in other areas of case law. Even though the desire to (not) procreate is one of the most intimate aspects of human ex­ist­ ence, very few elements have been accepted as fully private and personal. Rather, a high degree of state regulation and intervention is accepted due to the lack of common European consensus. On the one hand, the Court’s conservatism helps secure the support of Contracting States Parties to the ECHR in leaving sensitive matters with cultural differences and diverse legal and moral opinions to the domestic democratic mandate. On the other hand, it has placed women’s right in a position where human rights are certainly not applied as “a living instrument” or interpreted in “light of present days conditions.” The lack of European consensus is evident from the diverse national legislative responses, with the most uniformity existing in the area of contraception and the least in the area of abortion. The EU’s ECJ, by contrast, has pioneered legal protection of pregnant women in employment, and health and safety protection of pregnant women is harmonized through a minimum guarantee Directive, applicable in the 28 EU Member States. Through the nondiscrimination and equality mainstreaming in EU policies, effective rights and safeguards have been secured for European women.

50  Ivana Bacik, “Maternity Care, Midwifery Practice and Women’s Choices: Recent Legal Developments” (2015) 21 Medico Legal J Irel 4–10. 51  Pojatina v Croatia App no 18568/12 [ECHR, 4 October 2018].

The Right to Avoid Procreation: A European Perspective   1049

Bibliography Andión-Ibañez X and Zampas C, “Conscientious Objection to Sexual and Reproductive Health Services: International Human Rights Standards and European Law and Practice” (2012) 19 European Journal of Health Law 231. Bourne J and Derry C, Gender and the Law (Routledge 2018). Collected Edition of the ‘Travaux Preparatoires’ of the European Convention on Human Rights: Volume I: Preparatory Commission of the Council of Europe, Committee of Ministers, Consultative Assembly (11 May–13 July 1949) (Brill Nijhoff 1975). https://brill.com/view/ title/9554; accessed 13 April 2018. de Witte B, Muir E, and Dawson M, Judicial Activism at the European Court of Justice, Causes, Responses and Solutions (Edward Elgar Publishing 2013). http://ep.fjernadgang.kb.dk/ login?url=http://kbdk.eblib.com/patron/FullRecord.aspx?p=1164208; accessed 16 July 2018. Dworkin, R, Life’s Dominion (HarperCollins, 1993). Heino A and others, “Conscientious Objection and Induced Abortion in Europe” (2013) 18 The European Journal of Contraception & Reproductive Health Care 231. Koch L, “Eugenic Sterilisation in Scandinavia” (2006) 11 The European Legacy 299. Leppo A (2009): “Illusionen om enkla lösningar (The Illusion of Simple Solutions)” (2009) 26(3) Nordic Studies on Alcohol and Drugs 314–317. Marshall J, Personal Freedom Through Human Rights Law?: Autonomy, Identity and Integrity under the European Convention on Human Rights (Brill Nijhoff 2008). https://brill.com/ view/title/13127; accessed 4 July 2018. Milne RG and Wright RE, “The Decline of Fertility in Malta: The Role of Family Planning” (1997) 13 European Journal of Population [Revue Europeenne De Demographie] 147. Mouton M and German Historical Institute, From Nurturing the Nation to Purifying the Volk, Weimar and Nazi Family Policy, 1918–1945 (Cambridge University Press 2007). Pennings G, “Legal Harmonization and Reproductive Tourism in Europe” (2004) 19 Human Reproduction 2689. Söderström K and Skolbekken J-A, “Pregnancy and Substance Use: The Norwegian z 10–3 Solution. Ethical and Clinical Reflections Related to Incarceration of Pregnant Women to Protect the Foetus from Harmful Substances” (2012) 29 Nordic Studies on Alcohol and Drugs 155. Søvig KH, “Detention of Pregnant Women to Protect the Foetus: Nordic Perspectives” (2011) in Rynning, E. and Hartlev, M. (eds.) Nordic Health Law in a European Context p. 158. Toebes B and others (eds.), Health and Human Rights in Europe (1st ed., Intersentia 2012). Van der Sloot B, “Between Fact and Fiction: An Analysis of the Case-Law on Article 12 of the European Convention on Human Rights” (2014) 26 Child and Family Law Quarterly 397. Macfarlane AJ et al., “Wide Differences in Mode of Delivery Within Europe: Risk-stratified Analyses of Aggregated Routine Data from the Euro-Peristat Study” (2016) BJOG: An International Journal of Obstetrics & Gynaecology. Wiley Online Library. https://obgyn. onlinelibrary.wiley.com/doi/full/10.1111/1471-0528.13284; accessed 18 July 2018.

chapter 52

The R ight to Avoid Procr eation a n d the R egu l ation of Pr egna ncy A US Perspective Elizabeth Sepper

In the United States, the right to avoid procreation finds protection in constitutional and statutory law. The Supreme Court’s decisions to recognize contraception and abortion as constitutional rights have permitted generations of women a measure of reproductive freedom. Today, nearly all women will use some form of contraception during their reproductive years. One in four US women has had an abortion. The constitutionalization of contraception, and abortion in particular, has brought to the fore deeper contestation about the moral (and legal) status of the fetus, women’s place in society, and the meaning of motherhood. These same issues play out in decisions about the protection and regulation of women during their pregnancies. Within the framework of Supreme Court precedent, the fifty states have adopted varying approaches to the right to abortion—and related rights to equality, bodily integrity, self-­determination, and sex without reproduction. The trend, however, has inclined toward greater restrictions in a growing number of states, with some states reduced to a single abortion provider. At the same time, the right to avoid procreation under the US Constitution stands at the precipice at the Supreme Court. The current Court seems likely to dilute or reject the right to abortion, with consequences for contraception, protections for pregnant women at work, and criminal intervention during pregnancy.

1052   Elizabeth Sepper

1  The Constitutional Right to Avoid Procreation In the US system of federalism, states bear the primary responsibility both for regulating the medical profession and for enacting criminal laws. Provided such laws do not intrude on rights protected by the Constitution, courts apply a deferential standard of review, requiring only that the laws have a “rational” basis. When constitutional rights are at stake, however, courts give the government much less freedom to regulate. For purposes of reproductive autonomy, the Due Process Clause of the Fourteenth Amendment is particularly important in protecting liberty from infringements by the government.1 In a series of decisions in the 1960s and 1970s, the Supreme Court held that contraception and abortion were among the fundamental rights—like freedom of movement—meriting heightened constitutional protection from governmental intrusion. As it has developed, the jurisprudence of reproductive rights has balanced interests in privacy, bodily integrity, health, fetal life, medical ethics, and equality. While the contraceptive right remains intact, the legal standard governing the regulation of abortion has been lowered over time, granting states broad leeway to limit access. Recent decades have seen greater restrictions on abortion at the state level, although some states have developed relatively liberal abortion laws. The Court’s approach to the right to avoid procreation—and its reliance on a due process right to privacy in particular—has faced criticism from conservative and progressive scholars alike.

1.1 Contraception The right not to procreate originates in the Supreme Court’s 1965 decision, Griswold v. Connecticut, and it was initially conceived as a privacy right.2 While the Constitution does not enumerate a right to privacy, the Court explained that specific guarantees in the Constitution—like the prohibition on self-­incrimination, unreasonable searches and seizures, quartering of soldiers in any house, and freedoms to associate with others and to hold beliefs—create a constitutional “zone of privacy.”3 This privacy interest precluded a criminal ban on contraception to the extent that it invaded the privacy of the marital relationship. In its second major decision on contraception Eisenstadt v. Baird, the Court used the Equal Protection Clause to extend the right not to procreate to unmarried persons. Justice Brennan for the majority explained “If the right of privacy means anything, it is 1  Most abortion regulations are adopted by state governments, which are subject to the Fourteenth Amendment’s Due Process Clause. Some abortion regulations are passed by Congress or the government of Washington, DC, and these are subject to the Fifth Amendment’s Due Process Clause. Courts interpret the two due process clauses in the same manner. 2  381 U.S. 479 (1965). 3  Id. at 485.

The Right to Avoid Procreation: A US Perspective   1053 the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child”4—a phrase that would gain resonance in Roe v. Wade. Later decisions make clear that this broad right to contraception precludes state interference, but does not require state provision, and extends even to minors.

1.2  Abortion The Supreme Court has established boundaries within which states have to work when writing their abortion regulations. While predominantly and increasingly hostile to abortion, states have taken different paths in their approach to the interests at stake. Three key decisions form the signposts of the Supreme Court’s abortion jurisprudence: Roe v. Wade, Planned Parenthood of Southeastern Pennsylvania v. Casey, and Hellerstadt v. Whole Woman’s Health. In the 1973 decision in Roe v. Wade, the Court first struck down a ban on pre-­viability abortion.5 The decision made clear that the right to privacy identified in Griswold was based on the liberty protection of the Due Process Clause of the Fourteenth Amendment. It also set strict scrutiny—the highest constitutional test— as the standard for reviewing infringements on abortion. Applying strict scrutiny, the Court created the now-­famous trimester framework that barred government interference with abortion during the first trimester, allowed regulation reasonably related to maternal health in the second, and permitted bans on abortion in the third (post-­ viability) unless the procedure was necessary to preserve the life or health of a woman. When the abortion issue came back to the court in 1991 in Planned Parenthood of Southeastern Pennsylvania v. Casey,6 in an unusual joint opinion co-­authored by Justices O’Connor, Kennedy, and Souter, the Court discarded the trimester framework and relaxed its level of scrutiny. Instead, the Court held, a state may advance its interest in potential life and maternal health at any time in pregnancy unless state regulation imposes an “undue burden” on a woman’s decision to terminate her pregnancy. The Court defined “undue burden” as a law that “has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion.”7 The joint opinion said “What is at stake is the woman’s right to make the ultimate decision, not a right to be insulated from all others in doing so. Regulations which do no more than create a structural mechanism by which the State, or the parent or guardian of a minor, may express profound respect for the life of the unborn are permitted, if they are not a substantial obstacle to the woman’s exercise of the right to choose.”8 Applying this new standard, the Court upheld a 24-­hour waiting period and mandatory disclosures of information to patients but struck down a requirement that women first notify their spouses. In upholding the abortion right, the Court made clear that bans on abortion prior to

4  405 U.S. 438, 453 (1972). 5  410 U.S. 113 (1973). 7  Id. at 876–877. 8  Id. at 877.

6  505 U.S. 833 (1992).

1054   Elizabeth Sepper v­ iability constitute an undue burden on the abortion right, whereas after viability ­furthering the interest in fetal life may satisfy the strict scrutiny standard. As articulated in Casey, the undue burden test acknowledges a woman’s liberty interest in decisional privacy and autonomy as well as her equality interest. Even as it retained the privacy analysis, the Court acknowledged the link to equality, saying “[t]he ability of women to participate equally in the economic and social life of the Nation has been facilitated by their ability to control their reproductive lives.”9 At the same time, the Supreme Court has recognized governmental interests in fetal life, women’s health, and, in one case, protecting women from regret—regulatory interests that grow as the pregnancy progresses. After Casey, states varied widely in their approach. Some states retained Roe’s basic framework. Others adopted a moderate ground. Many states enacted extensive regulations. In the interest of fetal life, they imposed waiting periods stretching up to 72 hours, informed consent scripts (including notices contrary to science, for example, that abortion causes breast cancer), and ultrasounds prior to the abortion procedure. States also shifted toward passing laws related to maternal health—more specifically, asserting a goal of safe clinical environments for abortion procedures. Laws known as Targeted Regulation of Abortion Providers (or TRAP laws) ranged from mandated numbers of parking spaces to door widths to prohibitions on proximity to elementary schools. Widespread clinic closures resulted, leaving many women living far from abortion care. While courts understood Casey to prohibit an abortion ban, they upheld many of these regulations that made abortion difficult for women to access and burdensome for providers to deliver. In 2016, Whole Woman’s Health v. Hellerstedt seemed to reinvigorate the undue burden standard as a limitation on state power to regulate abortion through TRAP laws. There, Texas had imposed two new requirements: (1) that abortion clinics meet the health and safety standards for ambulatory surgical centers and (2) that any provider performing an abortion have admitting privileges at a hospital within thirty miles of the clinic where the abortion would be performed. Together, the district court found, these provisions would close “almost all abortion clinics in Texas,” creating an “impermissible obstacle” to abortion.10 The Fifth Circuit disagreed. It determined that the undue burden standard was really a rational basis test, requiring deference to the legislature. Courts, it said, could not evaluate whether the regulations “would actually improve women’s health and safety.”11 Justice Breyer’s opinion for the Supreme Court emphasized that the undue burden standard required closer constitutional scrutiny. First, the Court explained, Casey “requires that courts consider the burdens a law imposes on abortion access together with the benefits those laws confer.”12 Carefully examining the record evidence, the 9  Casey, 505 U.S. at 835. 10  Whole Woman’s Health v. Lakey, 46 F. Supp. 3d 673, 687 (W.D. Tex. 2014). 11  Whole Woman’s Health v. Lakey, 769 F.3d 285, 297 (5th Cir.), vacated in part, 135 S. Ct. 399 (2014). 12  Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2309 (2016).

The Right to Avoid Procreation: A US Perspective   1055 Court struck down the admitting privileges provision. The state could produce little ­evidence of benefit to women’s health over prior law, which already required a working arrangement with a physician with hospital privileges. By contrast, the evidence of the law’s burden was significant—half of clinics closed, meaning “fewer doctors, longer waiting times, and increased crowding,” and the numbers of women living 150 or 200 miles from a provider skyrocketed.13 Those burdens, taken together, met the undue burden standard. The surgical center standard provision likewise did not benefit patients. The record made clear that “abortions taking place in an abortion facility are safe—indeed, safer than numerous procedures that take place outside hospitals and to which Texas does not apply its surgical-­center requirements.”14 The cost of converting to a surgical center would close all but seven or eight clinics. “More fundamentally, in the face of no threat to women’s health, Texas seeks to force women to travel long distances to get abortions in crammed-­to-­capacity superfacilities. Patients seeking these services are less likely to get the kind of individualized attention, serious conversation, and emotional support that doctors at less taxed facilities may have offered”—the Court said—harming, not protecting, quality of care and health. Whole Woman’s Health was expected to constrain state regulation in the absence of health benefits. The opinion signaled that regulation justified by women’s health and informed decision-­making—requirements such as delivering informed consent scripts or permitting only physicians to dispense medication abortion—would need to reflect scientific evidence and the practice of medicine. But the retirement of Justice Kennedy, who joined the opinion, lessens Whole Woman’s Health’s import. In 2020 in Gee v. June Medical Services, LLC, the Supreme Court heard a challenge to a Louisiana admitting-­ privileges law that is indistinguishable from the Texas law struck down in Whole Woman’s Health. For reasons of stare decisis, Chief Justice Roberts joined Justices Breyer, Ginsburg, Kagan, and Sotomayor to strike down the law. With Justice Ginsburg’s death and the nomination of Amy Coney Barrett, rejection of the right to abortion seems nearly certain. States today range on a spectrum from liberal to moderate to hostile toward the abortion right. Scoring states based on their policies, the Guttmacher Institute indicates that 4 states are supportive or very supportive of abortion, 25 adopt middle-­ground positions, and 21 states are very hostile or hostile to abortion.15 Middle-­ground states will require abortions to be performed by a licensed physician rather than advanced practice clinician, permit providers and institutions to refuse to perform abortions, require parental involvement in a minor’s abortion decision, and impose gestational limits (typically, though not always, at viability) after which abortions are banned except when necessary for the woman’s life or health. More supportive states typically protect access to abortion clinics, guarantee public and private insurance for abortion, and have a state statute or constitutional provision 13  Id. at 2313. 14  Id. at 2315. 15  Guttmacher Institute, State Abortion Policy Landscape: From Hostile to Supportive (December 2018), https://www.guttmacher.org/article/2018/12/state-­abortion-­policy-­landscape-­hostile-­supportive.

1056   Elizabeth Sepper protecting access to abortion independent of and sometimes exceeding the federal constitutional standard. For example, in 2018, the Iowa Supreme Court rejected the undue burden standard because it “allow[ed] the legislature to intrude upon the profoundly personal realms of family and reproductive autonomy, virtually unchecked.”16 Women’s liberty interest and equality interest required strict scrutiny, it held. At the hostile end of the spectrum, states typically restrict public funding and private insurance coverage of abortion, impose informed consent scripts and two trips to the facility, prohibit the use of telemedicine for medication abortion, and impose wide-­ ranging TRAP laws. They ban abortion after 20 weeks in all but rare cases. Five states now have passed 6-­week bans, seeking to outlaw abortion before most women know they are pregnant—in violation of constitutional protections. While states divide in their approach, the past two decades have seen ever-­greater restrictions in an ever-­larger number of states.

2  Critiques of the Constitutional Framework The Supreme Court’s jurisprudence on the right to avoid procreation has faced ­sustained critique. From the right, scholars and advocates attack Roe for its reliance on unenumerated rights, arguing that the Constitution should protect only those liberties explicitly listed in the Bill of Rights. “Due Process,” they say, does not mean a right to privacy, let alone a right to abortion. Others like Justice Rehnquist in his Roe dissent focus on the appropriate role of the courts and argue that abortion rights—and consideration of appropriate line-­drawing—was “far more appropriate to a legislative judgment than to a judicial one.”17 Moreover, on this view, the Court interfered with the federalist system and occupied a field best left to the states. Other scholars take the position that the Court has given insufficient weight to the interests in fetal life and should weigh the fetus’s rights as equal to those of the pregnant woman. Across the political spectrum, critics have agreed with Justice Scalia’s assessment that the undue burden test has an “inherently standardless nature,” allowing trial courts to uphold or strike down laws “[b]y finding and relying upon the right facts.”18 Feminist scholars have long argued that the Court’s focus on liberty—in the form of decisional autonomy and bodily integrity—obscures sex equality and the gender role stereotypes at play. Reva Seigel writes, “There is no language in Roe suggesting that society’s changing attitudes toward women . . . warranted a change in the constitutionality of criminal abortion statutes. . . . Only indirectly does Roe acknowledge that laws criminalizing abortion present questions of sexual freedom for women.”19 While Casey did 16  Planned Parenthood of the Heartland v. Reynolds, 915 N.W.2d 206, 240 (Iowa 2018). 17  Id. at 173–174 (Rehnquist, J., dissenting). 18  Casey, 505 U.S. 833, 992 (Scalia, J., concurring in part and dissenting in part). 19  Reva B. Siegel, Roe’s Roots: The Women’s Rights Claims That Engendered Roe, 90 B.U. L. Rev. 1875, 1899 (2010).

The Right to Avoid Procreation: A US Perspective   1057 r­ecognize considerations of equality, sex equality analysis remains largely in the ­background of abortion doctrine as fetuses and physicians take center stage.20 Much abortion regulation seems “premised on the assumption that women—both young women and older ones—make decisions about abortion lightly, or impulsively, and that if only they were made to reflect just a bit longer and with a bit more information, they would change their minds.”21 Some feminist critics have sought to more fundamentally disrupt thinking about abortion as constitutional right. Catherine MacKinnon has contended that the constitutionalization of abortion distracts from the work of addressing the social and sexual disparities that created the unwanted pregnancy and, more generally, that encompassing abortion within a “right to privacy” risks insulating from scrutiny intimate relations in which men subordinate women.22 Robin West has further critiqued a constitutional right to choose to parent or terminate a pregnancy for making parenting a private choice and cost and in turn legitimizing the absence of an adequate social welfare net.23 Applying critical legal theory, West identifies costs to the reproductive rights movement’s emphasis on courts, in particular the abandonment of more radical demands for social change that the feminist movement advanced in the 1970s. The reproductive justice movement—led by women activists of color—has advocated a more comprehensive shift. Rather than singularly focus on abortion or contraception, reproductive justice requires “the complete physical, mental, spiritual, political, social, and economic well-­being of women and girls.”24 It demands the equal importance of rights to have a child, rights to avoid procreation, and the right to parent one’s children with the recognition that these rights are intimately linked to the conditions in one’s community from environment to education to crime. Ambitious in scope, the reproductive justice movement demands radical change to eradicate the legal, social, and cultural oppression of women and women of color in particular.

3  Barriers to Access Statutory, social, and economic conditions constrain women’s ability to exercise rights to contraception and abortion. Lack of insurance coverage, unavailability of healthcare providers, youth, and stigma prove formidable barriers to access. The extent to which

20  Sylvia Law, Rethinking Sex and the Constitution, 132 U. Pa. L. Rev. 955, 986 (1984). 21  Carol Sanger, Decisional Dignity: Teenage Abortion, Bypass Hearings, and the Misuse of Law, 18 Colum. J. Gender & L. 409, 414 (2009). 22  Catharine A. MacKinnon, Feminism Unmodified: Discourses on Life and Law94–102 (1988). 23  Robin West, From Choice to Reproductive Justice: De-­Constitutionalizing Abortion Rights, 118 Yale L.J. 1394, 1409–1410 (2009). 24  Loretta Ross & Rickie Solinger, Reproductive Justice: An Introduction (2017).

1058   Elizabeth Sepper women encounter and can overcome these obstacles depends—like so much in the United States—on their race, class, and location.

3.1  Poverty and Lack of Insurance Ninety-­nine percent of US women use contraception during their reproductive years. Nearly one in four has an abortion in her lifetime. The costs of contraception and abortion are high and pose weighty burdens on low- and middle-­income women. For many women, their poverty and lack of insurance stand in the way of a meaningful exercise of the right to avoid procreation. Roe presented the possibility of mainstreaming abortion within healthcare. Many expected that Medicaid, the health insurance program for the poor, would cover abortion just as it insures pregnancy. As one district court put it, the program would ra­tion­ ally recognize that “abortion and childbirth, when stripped of the sensitive moral arguments surrounding the abortion controversy, are simply two alternative medical methods of dealing with pregnancy.”25 Just four years after Roe, however, the Supreme Court held that states had no obligation to cover nontherapeutic abortions, upholding a Connecticut law that limited public funding to abortions certified to be “medically necessary.”26 The law did not “disadvantage” any poor woman, the Court said. [S]he continues as before to be dependent on private sources for the service she desires. . . . The indigency that may make it difficult and in some cases, perhaps, impossible for some women to have abortions is neither created nor in any way affected by the Connecticut regulation.27

In 1976, Congress also put in place the Hyde Amendment to keep federal funds from being used to pay for abortions for poor women even if medically necessary, with a few narrow exceptions that have varied over time (sometimes limited to instances when the life of the mother would be endangered if the fetus were carried to term, sometimes extended to rape or incest). The Supreme Court upheld the law, concluding that the denial of funding presented “no governmental obstacles in the path of a woman who chooses to terminate her pregnancy” but merely a refusal to subsidize her choice.28 Even the choice language, embraced by the abortion rights movement, may disadvantage poor women. As Rickie Solinger writes, “What happens when the special guarantee for all women—the promise that women can decide for themselves whether and when to be mothers—is expressed by the individualistic, marketplace term ‘choice’? How can 25  Roe v. Norton, 408 F. Supp. 660, 663 n.3 (D. Conn. 1975), rev’d by Maher v. Roe, 432 U.S. 464 (1977). 26  Maher v. Roe, 432 U.S. 464 (1977). The Connecticut statute defined as “medically necessary” those abortions necessary to preserve the physical or psychiatric health of the woman. 27  Id. at 474. 28  Harris v. McRae, 448 U.S. 297, 316 (1980).

The Right to Avoid Procreation: A US Perspective   1059 users of such a term avoid distinguishing, in a consumer-­culture fashion, between a woman who can and a woman who can’t afford to make a choice?”29 Today, the ban on federal Medicaid funding of abortion remains in place, although seventeen states operate programs that cover abortions for their Medicaid populations. Title X of the Public Health Service Act, enacted in 1970, provides grants to fund family planning services, but Title X funding cannot be used “in programs where abortion is a method of family planning.” In 2010, the Affordable Care Act (ACA), which expands access to public and private health insurance programs, likewise included a ban on federal funding of abortion. While states may require individual insurance plans sold through ACA-­created exchanges to cover abortion, federal subsidies for premiums may not be used to cover the abortion-­related portion of the premium. In some states, laws extend public financing bans to all insurance plans sold in the state. Under Missouri law, for example, private insurance policies may cover abortion only in cases of life endangerment unless a policyholder purchases an optional rider at an additional cost. Only three states—California, New York, and Oregon—require insurance plans to cover abortion. Contraception—by contrast—has been subsidized. To permit access to contraception, legislatures put in place funding programs for low-­income women. Some states have adopted measures expanding access to family planning services through Medicaid, requiring provision of emergency contraception for sexual assault victims in hospital emergency departments, and permitting pharmacist prescription of oral contraceptives. Millions of low-­income people in the United States have access to contraception through Title X, Medicaid, and other public programs. Until the late 1990s, people with private or employer-­based plans often found contraceptive coverage absent or insufficient. Employer insurance had developed over the twentieth century to meet men’s health needs and frequently excluded maternity and other women’s health services as “lifestyle choices” not healthcare. But then, in 1998, the Food and Drug Administration (FDA) approved Viagra, the first erectile dysfunction drug, and virtually all insurance plans immediately covered the drug. Women’s groups nationwide protested and lobbied state capitols. Twenty-­nine states now require insurance plans that cover prescription drugs to also cover any FDA-­approved contraceptive. To remedy continuing coverage gaps and barriers posed by insurance co-­pays, the federal ACA mandated that employers that offer insurance plans cover the full range of contraceptive methods, including emergency contraception, IUDs, sterilization, and oral contraceptives, without cost-­sharing from plan beneficiaries. Despite variation across state programs, federal law provides a national floor of coverage but confronts ongoing challenges. Contraceptive mandates have sparked demands for religious exemptions, discussed in the next section. In 2019, the Trump administration imposed a gag rule prohibiting Title X clinics from discussing abortion with their patients. As a result, Planned Parenthood—a provider of a significant share of family 29  Rickie Solinger, Beggars and Choosers: How the Politics of Choice Shapes Adoption, Abortion, and Welfare in the United States 6 (2001).

1060   Elizabeth Sepper planning services and abortions nationwide—exited the program. Title X and Medicaid also face regular threats to their funding.

3.2  Religious and Other Refusals Religious and institutional refusals to deliver reproductive health procedures impede access to contraception and abortion. Women frequently find contraception, sterilization, tubal ligation, and abortion denied to them in Catholic-­affiliated healthcare facilities, which are widespread and sometimes provide the only hospital care in town. Pharmacies sometimes refuse to stock or fill prescriptions for emergency contraception. Crisis pregnancy centers (CPCs) frequently deceive women into believing that they provide medical care and/or abortions. Since 1973, legislation at the national level and in almost every state has authorized providers who object to abortion (or, frequently, sterilization) to refuse to provide or participate in the procedures. These laws were inspired by court cases requiring hospitals to permit tubal ligations following delivery and, of course, Roe v. Wade. Congress made clear that receipt of federal funds did not require an individual or institution to perform sterilizations or abortions if it “would be contrary to religious beliefs or moral convictions.”30 Legislatures across the United States followed, passing legislation— known as conscience clauses—to protect refusing providers from discrimination in ­hiring, staff privileges, or promotion; professional discipline; civil actions (typically malpractice); and regulatory or criminal sanctions. In almost every state, a doctor may refuse to provide abortion; refusals of contraception and sterilization are also commonly permitted. Under all but a few conscience clauses, a doctor or nurse may refuse care even in emergency situations. Under some statutes, healthcare providers may withhold information, refuse to refer for treatment, and deny patients care distant from the contested procedure. This legislation constitutes a significant exemption from statutory and common law. Although providers in the United States have no general legal or ethical duty to treat a patient or provide a service, federal and state statutes require that in hospital emergency rooms, patients suffering from emergency conditions or in active labor be stabilized. And, once providers have accepted a person as a patient, law requires them to treat the patient in accordance with the standard of care, inform of treatments and their risks and benefits, refer for services they are not able to provide, and continue to care for one’s patient who needs continuing medical attention. Under refusal laws, patients harmed by refusal cannot recover from refusing hospitals and providers for the harm done to them. For example, delaying an abortion procedure for a woman who is miscarrying—as Catholic hospitals often require—may contravene the standard of care, yet, with these statutes, the patient’s injuries go unremedied.

30  42 U.S.C. § 300a-­7(b) (2000).

The Right to Avoid Procreation: A US Perspective   1061 The Supreme Court recently reined in the government’s ability to regulate refusing institutions in order to ensure access to reproductive health services in two significant cases. In Burwell v. Hobby Lobby, the Supreme Court held that for-­profit corporations may exercise religion and granted objecting employers an accommodation from the ACA’s contraceptive mandate under the Religious Freedom Restoration Act (RFRA), which grants statutory protection to the exercise of religion.31 While this accommodation ultimately guaranteed employees contraceptive access, other employers demanded a total exemption and received one by regulation when the Trump administration came into office. Of far greater consequence will be the 2018 case of National Institute of Family & Life Advocates v. Becerra.32 There, California required clinics primarily serving pregnant women to provide notices: in the case of licensed clinics, about the availability of state-­ funded family planning services, including abortion and contraception; and, in the case of unlicensed clinics, that they are not licensed to provide medical services. Anti-­ abortion CPCs challenged the notices as a violation of rights to free speech and religious exercise. The Supreme Court agreed. It held that these requirements constituted impermissible content-­based regulation because they compelled the CPCs to deliver a particular notice and thus changed the content of their speech. The dissenting justices noted that this reasoning, if applied generally, would mean the end of virtually all disclosure requirements from financial regulation to health warnings. It is more likely that the case represents the abortion exceptionalism that runs through the Court’s First Amendment doctrine, whereby corporations—from tobacco companies to unlicensed ultrasound providers—and other professionals enjoy protection of their speech even as abortion providers find their speech tightly controlled.

3.3 Minority State laws often grant minors privacy from parental involvement where sensitive reproductive and sexual health issues are involved. For example, teenagers may obtain treatment for sexually transmitted infections and prescription contraceptives without parental involvement or consent. Likewise, teens can consent to medical treatment during pregnancy. Minors face additional hurdles to obtaining abortions. Most states require a minor to either notify or get consent from her parents (one or both, depending on the state) before she can get an abortion. Alternatively, she may petition a judge for permission. This process, known as a judicial bypass hearing, results from the Supreme Court’s decision in Bellotti v. Baird.33 The Court sought to strike a balance between parents’ constitutionally protected liberty interest to raise their children as they see fit and a minor’s constitutionally protected liberty interest in abortion and medical decision-­making. Parents, the 31  42 U.S.C. § 2000bb–1(a) & (b). 33  443 U.S. 622 (1979).

32  138 S. Ct. 2361, 2368 (2018).

1062   Elizabeth Sepper Court determined, must have a say in a decision with “consequences so grave and indelible,” but may not hold a veto over their daughters’ decision.34 Thus, a minor may go directly to court to make her case to a judge for why she should be granted an abortion. In practice, the process is often difficult and humiliating for teenagers.35

3.4  Social Stigma In addition to legal and financial constraints, reproductive healthcare suffers from social stigma. Women typically stay silent about their abortions (and even dilation and curettage following miscarriage), resulting in feelings of isolation and the perception that “other” women have abortions. Increasingly, pro-­choice advocates encourage women to “shout their abortion” or be transparent with friends and family in order to personalize the political issue of abortion. An ideal of inevitable and natural motherhood—what Katherine Franke terms “repronormativity”—means that young or unmarried women seeking sterilization find doctors unwilling to perform the procedure based on the conviction that they will eventually embrace maternity.36 Contraception generally is less stigmatized, widely provided, and more discussed (at least between friends and partners). Nonetheless, societal anxieties about sex, and premarital sex in particular, manifest themselves in policy. For example, a number of states limit sex education in public schools to teaching abstinence only.

4  Protections for and Regulation of Pregnancy US law both grants equality protections and limits liberty interests of pregnant people. To encourage maternity and protect pregnancy, statutory schemes ensure maternity benefits, mandate nondiscrimination, and grant family leave. But they do so in limited ways, with the result that pregnant women and new parents labor under poor and sometimes dangerous conditions. The balance of fetal and maternal interests can shift as women determine to continue their pregnancies and deliver babies. In such instances, courts have sometimes permitted arrest, incarceration, and forced medical intervention to preserve a state interest in fetal life. Low-­income and racial minority women disproportionately have borne the burdens of both government surveillance and employer demands.

34  Id. at 642. 35  See generally Sanger, supra note 21. 36 Katherine M. Franke, Theorizing Yes: An Essay on Feminism, Law, and Desire, 101 Colum. L. Rev. 181 (2001).

The Right to Avoid Procreation: A US Perspective   1063

4.1  Workplace and Antidiscrimination Law The social norm is for women to work through pregnancy, at least partly because few can afford to lose their wages. Workplace law affords some significant—but limited— protections to employees during their pregnancies. The year after Roe, the Supreme Court infamously opined that excluding pregnancy from California’s disability insurance program was not sex discrimination subject to heightened constitutional scrutiny because the program treated men and women alike, merely distinguishing between “pregnant women and non-­pregnant persons.”37 Shortly thereafter it similarly held that the Civil Rights Act of 1964 barred sex, but not pregnancy, discrimination in employment.38 The Pregnancy Discrimination Act (PDA), passed in 1978, rejected this view. The act clarifies that sex discrimination also means discrimination “because of or on the basis of pregnancy, childbirth, or related medical conditions.”39 It also applies when a pregnant worker is temporarily unable to physically perform some of her job duties. In providing that “women affected by pregnancy, childbirth, or related medical conditions shall be treated the same for all employment-­related purposes . . . as other persons not so affected but similar in their ability or inability to work,”40 however, the PDA only requires the same treatment, rather than an affirmative right to accommodation. The question then arises: What other workers are similar? Until recently, unless an employer singled out pregnancy for exclusion, courts routinely rejected claims challenging inadequate sick leave, heavy-­lifting requirements, limited bathroom breaks, or prohibitions on carrying water bottles “as a demand for preferential treatment or a subsidy inconsistent with an antidiscrimination mandate.”41 In Young v. United Parcel Services, Inc.,42 the Supreme Court adopted a standard somewhat more favorable to pregnant employees, instructing courts to compare pregnant workers not to employees with nonoccupational injuries but to the class of nonpregnant workers, whether injured on the job or off. Under this standard, employers must accommodate pregnant workers when they provide an accommodation to any other nonpregnant employee who is similar in ability to work. The Court warned that the employer’s justification for failing to accommodate pregnancy “normally cannot consist simply of a claim that it is more expensive or less con­ven­ient to add pregnant women to the category of those . . . whom the employer accommodates.”43 The Family and Medical Leave Act (FMLA) entitles employees to take twelve weeks of leave from work to care, for example, for a newborn, a newly adopted child, or an ill spouse. Employees are also entitled to take time for their own medical needs. One of the FMLA’s stated purposes in allowing reasonable leave is to “minimize[] the potential for employment discrimination on the basis of sex by ensuring generally that leave is available 37  Geduldig v. Aiello, 417 U.S. 484, 496 (1974). 38  General Electric Co. v. Gilbert, 429 U.S. 125 (1976). 39  42 U.S.C. 2000e(k) (1978). 40 Id. 41  Deborah Dinner, Strange Bedfellows at Work: Neomaternalism in the Making of Sex Discrimination Law, 91 Wash. U. L. Rev. 453, 523–524 (2014). 42  135 S. Ct. 1338 (2015). 43  Id. at 1354.

1064   Elizabeth Sepper for eligible medical reasons (including maternity-­related disability) and for compelling family reasons, on a gender-­neutral basis.”44 The impact of the FMLA, however, is softened by two significant limitations. First, FMLA leave is unpaid, making it financially out of the reach of many workers. Second, approximately 40% of workers are not covered by the FMLA due to the size of their employer, length of their tenure at the workplace, or inadequate number of hours worked.

4.2  Lifestyle Choices During Pregnancy During pregnancy, a web of cultural norms seeks to control a pregnant person’s behavior. Legal activities invite opprobrium. Pregnant people find their consumption of food, participation in athletics, and occasional drink of alcohol policed by their fellow citizens. Having chosen to continue their pregnancies, they are considered to assume moral, if not legal, duties to the fetus. Healthcare providers face particular dilemmas when their pregnant patients use drugs. Illicit as well as prescription drugs may pose risks for the health and development of the fetus. In these instances, maternal autonomy may conflict with fetal well-­being. Alternatively, one might model the problem of addiction in particular as presenting not a conflict but an alignment between maternal and fetal well-­being. A woman seeking to become a mother and the fetus yet to be born share an interest in treatment for drug use, adequate prenatal care, and the establishment of an early attachment between mother and child. Pregnant women’s illicit drug use can result in the involvement of law enforcement. While illicit drug possession is criminalized, drug use alone does not result in criminal penalties unless the user becomes pregnant. Before 2014, no state had enacted a law that directly criminalized the use of drugs by a pregnant woman. But, since the late 1970s, many state prosecutors had relied on generally applicable criminal statutes—for neglect or chemical endangerment of a child, delivery of drugs to a minor, involuntary manslaughter, and so on—to arrest and incarcerate women for using drugs during pregnancy. Many courts expressed concern about the broad readings of criminal statutes involved in these cases. One court noted that any woman who experienced a miscarriage “would be at risk of a criminal indictment for virtually any perceived self-­ destructive behavior during her pregnancy…, to wit: smoking or drinking heavily; using illegal drugs or abusing legal medications; driving while under the influence of drugs or alcohol; or any other dangerous or reckless conduct.”45 By contrast, other courts defined “child” under state child endangerment statutes to include fetuses, endorsing pregnant women’s prosecution.46 In 2014, Tennessee specifically criminalized the use of drugs by a pregnant woman and prosecuted ninety-­five women before the law expired in 2016. And in several states, personhood measures, which would make a fetus a legal person for purposes of all state law, have been proposed—thus far unsuccessfully. 44  29 U.S.C. § 2601(b)(4) & (2). 45  Hillman v. State, 503 S.E.2d 610, 613 (Ga. App. 1998). 46  Whitner v. State, 492 S.E.2d 777, 778 (S.C. 1997); Ex Parte Ankrom, 152 So.3d 397, 408 (Ala. 2013).

The Right to Avoid Procreation: A US Perspective   1065 The risk of harm to the fetus has served to justify the arrest and incarceration of pregnant women for drug use. Yet scientific studies show that lawfully smoking, taking certain prescription drugs, and drinking alcohol have an equal or greater negative impact on a developing fetus.47 Cortney Lollar has contended that racism is reflected in the use of the law enforcement model: “When anyone is prosecuted for their activities while pregnant, they are primarily, if not only, poor and African American women.”48 Even where law enforcement is not involved, pregnant women who use drugs may lose custody of their newborns. Twenty-­four states plus the District of Columbia consider drug use while pregnant child abuse.49 On suspicion of fetal drug exposure, child protective services may deprive often-­vulnerable women of their newborns during a time important for attachment and development of the child. Poor and African American women are most likely to be targeted. Errors also can occur. For example, a Pennsylvania couple was separated from their newborn after a poppyseed bagel the mother had eaten prior to labor indicated drug use in a test to which she had not consented. In some states, a medical, instead of law enforcement, model applies in the form of drug treatment programs designated for pregnant women or priority access to state drug treatment programs. Research indicates that both rehabilitation and social service coordination generally provide for better outcomes for both the mother and the fetus.50 The most comprehensive study—which admittedly undercounts—found hundreds of cases in which “pregnant women were denied a range of fundamental rights normally associated with constitutional personhood, including the right to life, physical liberty, bodily integrity, due process of law, equal protection, and religious liberty, based solely on their pregnancy status.”51 In nearly one in five cases, not adhering strictly to medical advice was cited as a factor to justify arrest, detention, or forced medical interventions. In most cases, no adverse pregnancy outcome occurred. The women arrested, detained, prosecuted, or subject to forced medical intervention were significantly more likely to be poor and/or African American.

4.3  Decision-­Making During Labor The legal and ethical framework governing decision-­making during labor reflects many of the same considerations related to regulating activities during pregnancy but pre-­ labor. Having determined to carry the pregnancy to term and give birth, a woman may be considered to have elevated moral or even legal responsibilities toward the fetus. 47 Cortney E. Lollar, Criminalizing Pregnancy, 92 Ind. L.J. 947, 949–950 (2017). 48  Id. at 953. 49  Seema Mohapatra, Unshackling Addiction: A Public Health Approach to Drug Use During Pregnancy, 26 Wis. J. L. Gender & Soc’y 241, 267 (2011). 50  Rebecca Stone, Pregnant Women and Substance Use: Fear, Stigma, and Barriers to Care, 3 Health & Justice 1, 3 (2015). 51  Lynn M. Paltrow & Jeanne Flavin, Arrests of and Forced Interventions on Pregnant Women in the United States, 1973–2005: Implications for Women’s Legal Status and Public Health, 38 J. Health Pol., Pol’y & L. 299 (2013).

1066   Elizabeth Sepper Hospitals sometimes seek court orders to force pregnant women unwilling to undergo surgery to have a forced cesarean delivery (c-­section). In some instances, the woman safely delivers vaginally, consistent with her wishes, during the time it takes to seek a court order. Courts have disagreed on the extent to which a woman’s choice of childbirth procedure must be honored. Some courts have held that the state’s interest in protecting fetal life at delivery must be balanced against, and can outweigh, a woman’s right to refuse treatment, including forced c-­sections.52 In reaching this conclusion, courts often rely on Roe v. Wade and reason that because the state’s interest in the potential life of the fetus suffices to bar women from aborting in the third trimester, it similarly justifies forced c-­section or other delivery interventions. Other courts hold that the fact of pregnancy does not alter women’s rights to refuse medical treatment. The DC Court of Appeals en banc, for example, recognized that “if a patient is competent and has made an informed decision regarding the course of her medical treatment that decision will control in virtually all cases.”53 Setting aside the issue of forced interventions, childbirth options are limited in the United States. The vast majority of women give birth in hospitals, which until recently have offered little access to midwives, doulas, and nonmedical pain relief techniques and tools. Non-­hospital births have increased dramatically in number (e.g., the number of babies born in birth centers increased 83% between 2006 and 2016) but remain around 1.5% of births nationwide. The ACA has aided access to midwives and non-­ hospital care, for example by requiring insurers to reimburse nurse midwives as they do ob-­gyns. Childbirth is costly in the United States, with the average fee for delivery of a vaginal birth exceeding $10,000. Unlike abortion, maternity care is covered under the federal Medicaid Act for low-­income women, which pays medical expenses for 45% of births in the United States. Privately insured women, however, may face significant cost-­sharing in the form of insurance deductibles. C-­sections drive these costs. US c-­section rates exceed those of similar countries, without clinical benefit. Hospitals vary significantly, with c-­section rates ranging from 7.1% to 69.9% of deliveries, a tenfold difference likely explained by physician practice patterns rather than maternal preference.54 Childbirth also holds significant risks in the United States, which ranks as the most dangerous country in the developed world for a woman to give birth. US women are three times more likely to die than Canadian women and six times more likely than Scandinavian women during pregnancy and in the first year after delivery or termination

52  See, e.g., Pemberton v. Tallahassee Mem’l Reg’l Med. Ctr., Inc., 66 F. Supp. 2d 1247 (N.D. Fla. 1999); Jefferson v. Griffin Spalding Cty. Hosp. Auth., 274 S.E.2d 457 (Ga. 1981). 53  In re A.C., 573 A.2d 1235 (D.C. 1990) (en banc); see also Illinois v. Mother Doe, 632 N.E.2d 326, 332 (Ill. App. Ct. 1994) (“The woman retains the same right to refuse invasive treatment, even of lifesaving or other beneficial nature, that she can exercise when she is not pregnant”). 54  Katy Backes Kozhimannil, Michael R. Law & Beth A. Virnig, Cesarean Delivery Rates Vary 10-­Fold Among US Hospitals, 32 Health Aff. 527 (2013).

The Right to Avoid Procreation: A US Perspective   1067 of the pregnancy.55 Black women suffer a maternal mortality rate three to four times that of white women—comparable to women in Mexico and Uzbekistan.56 Many more experience life-­threatening injuries. Myriad factors contribute to maternal mortality in the United States, including lack of access to care, increases in maternal age, chronic disease rate, c-­section delivery rate, and postpartum care consisting of only one follow-­up weeks after delivery.57 Ultimately, maternal mortality reflects disparities and unconscious biases in healthcare across race, gender, and socioeconomic lines.58

5 Conclusion Since 1973, US constitutional doctrine has protected the right to avoid procreation. But statutory and practical hurdles have rendered the right illusory for many women, especially those marginalized by their race, gender, sexual orientation, geography, or class. And debates over women as mothers and fetuses as persons have raged in legal, political, and social fora. In 2018, Justice Kennedy—the co-­author of Casey—resigned from the Supreme Court. And in 2020, Justice Ruth Bader Ginsburg, a champion on reproductive freedom, died. Their replacements stand to provide votes to reject abortion. The Court could overturn Roe explicitly. In that case, access to abortion would vary even more widely by state. Approximately twenty states likely would ban abortion outright (some have “trigger bans” on the books to automatically criminalize abortion if Roe falls).59 In about the same number of states, the right to abortion would remain protected, con­sist­ ent with or exceeding what the federal constitution had required. Alternatively, the Court could stop short of overruling Roe but dilute the undue burden test to a rational basis test, which—as Whole Woman’s Health explained—would in effect block access for many (and perhaps most) women. Under both circumstances, courts would remain in what Justice Scalia called “the abortion-­umpiring business.”60 Even if states were constitutionally permitted to prohibit abortions within their own borders, federal courts would face new, difficult, and “morally freighted” questions—especially, as Richard Fallon pointed out a decade ago, if states sought to prohibit abortions outside their borders or to target speech facilitating abortions.61 55  Nina Martin & Renee Montagne, Black Mothers Keep Dying After Giving Birth: Shalon Irving’s Story Explains Why, npr, https://www.npr.org/2017/12/07/568948782/ black-­mothers-­keep-­dying-­after-­giving-­birth-­shalon-­irvings-­story-­explains-­why. 56 Id. 57 Id. 58 Id. 59  Center for Reproductive Rights, What If Roe Fell?, https://www.reproductiverights.org/ what-­if-­roe-­fell. 60  Richard H. Fallon, Jr., If Roe Were Overruled: Abortion and the Constitution in A Post-­Roe World, 51 St. Louis U. L.J. 611, 613 (2007) (quoting Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 995–996 (1992) (Scalia, J., concurring in the judgment in part and dissenting in part)). 61  Id. at 626–640.

1068   Elizabeth Sepper Of course, in a post-Roe America, abortion access might not differ widely based on state citizenship. Congress might legislate to limit or protect access to abortion nationwide. While some precedent for federal power in this area exists (Congress, for example, enacted the so-­called partial abortion ban on a specific late-­term abortion procedure), comprehensive federal lawmaking on abortion would upend the long-­standing authority of states to regulate the practice of medicine. Reversal or dilution of constitutional protections for abortion would pose risks for contraception, antidiscrimination law, and physical liberty as well. Anti-­abortion activists have fueled a conflation of abortion and certain forms of contraception in the judicial and public imagination. For example, due in no small part to the message that emergency contraception is medication abortion, legislators across the country passed laws permitting pharmacists to refuse to fill prescriptions for emergency contraception. In the contraceptive mandate controversy, proponents of exemptions often insisted that Hobby Lobby and some other employers objected only to emergency contraception and intrauterine devices (IUDs)—which they labeled “abortifacients”—not other forms of contraception. In his confirmation hearings to replace Justice Kennedy on the Supreme Court, Justice Kavanaugh likewise called emergency contraception and IUDs “abortifacients.” The accommodation granted by the Court and total exemption by the Trump administration, of course, extended to all forms of contraception. While consequences would likely fall primarily on providers, the criminalization of abortion would give the embryo or fetus priority over the person of the woman, with wide-­ranging consequences. Government actors could assert enhanced authority to subject pregnant women to surveillance, incarceration, forced surgery, and other deprivations of liberty—that is, to subjugate the interest of the woman to the interest of the fetus. Employers could seek to fire or restrict pregnant workers due to concerns for the fetus. Defenders of rights to abortion and contraception are not entirely without options. State constitutions may again hold promise. In 2018 and 2019, respectively, the supreme courts of Iowa and Kansas held that their state constitutions safeguarded a right to abortion and required strict scrutiny of regulations infringing on this right to safeguard privacy and equality. In the long term, such sex equality arguments may prove a firmer normative foundation for constitutional protection than the liberty arguments adopted in Roe did. There is also the opportunity to advance religious and conscience arguments now monopolized by the religious right. The Supreme Court’s recent recognition of corporate free exercise of religion in Hobby Lobby and Masterpiece Cakeshop62 opens the door for institutions providing reproductive care to assert religious grounds to resist barriers to serving their patients. Some women seeking abortions or restricted forms of contraception could powerfully argue that their religious beliefs and conscience drive their choice. Indeed, when people expected Casey to result in the downfall of Roe in 62  Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Comm’n, 138 S. Ct. 1719 (2018) (deciding, without discussing the issue of corporate religion, a for-­profit business’s claim to free exercise under the Constitution).

The Right to Avoid Procreation: A US Perspective   1069 1992, religious anti-­abortion groups blocked the enactment of RFRA, fearful that women might successfully use the statute to secure conscience exemptions to an abortion ban. Ultimately, reversal of Roe and Casey would mean more women forced to carry unwanted pregnancies by reason of state citizenship and status. Poor people, teenagers, and women of color would suffer the brunt of the shift. Women who have come of age expecting contraception and abortion to be available would find that motherhood is the role that the state has chosen for them to play.

T H E L AW OF DE AT H A N D DY I NG

chapter 53

Decisions at the En d of Life David Orentlicher and Judit Sándor

1 Introduction Advances in medical technology have increased the ethical and legal dilemmas at the end of life. It is difficult to know what is a “good death” or what constitutes a dignified ending of life.1 Nor is the proper role of the law in regulating end-­of-­life medical care clear. To what extent should these matters be worked out between physicians and patients (or families), and to what extent should legal rules guide practice? And who should determine the rules: Legislators issuing laws of broad application or judges developing nuanced standards on a case-­by-­case basis?2 In this chapter we explore the similarities and differences between US and European legal thinking and jurisprudence. Of course, on a number of issues, there is no single US or European approach. Accordingly, while we will give special attention to the jurisprudence of the US Supreme Court and the European Court of Human Rights (ECtHR), we also will consider the significant variations among different US states and different European countries. In recent decades, courts and legislatures have increasingly respected the right to self-­determination of patients or their families in deciding about the way people want to die. In most jurisdictions, patients may have unwanted medical treatment withheld or withdrawn, even when the treatment would be or is life-­sustaining. Hospice and other palliative care are important alternatives, although in many cases suffering may still persist. In a small but growing number of jurisdictions, patients may receive a lethal dose of medication to bring about death. 1  Shai J. Lavi, The Modern Art of Dying (2005). 2  John Keown, Euthanasia, Ethics and Public Policy: An Argument Against Legislation (2002).

1074   David Orentlicher and Judit Sándor The increasing freedom to make healthcare decisions that may hasten death reflects the confluence of several factors. Both in the United States and in European countries, the law had previously recognized a right of “informed consent.” If patients can decide whether to receive a recommended treatment, they have a corollary right to refuse that treatment. Also, advances in medical technology have permitted physicians to save many lives that once were lost to disease or accident. In some cases, however, the person survives with a very poor quality of life and no hope of recovery. While medical care can maintain the person’s life for weeks, months, or even years, the greatly diminished quality of life and the burdensomeness of the treatment mean, to some individuals, that the treatment is not desired. And even for patients not dependent on medical treatment for survival, the irreversibility of their condition and the magnitude of their suffering may make continued life unwanted. Quality of life matters, not just quantity of life. In addition, life-­sustaining advances in technology have become very costly. If the life that can be preserved has a very poor quality and society’s limited resources could be used for other patients who would have a better quality of life and who want treatment, then it is not clear that the state has an interest in always maintaining a patient’s life. As with other rights, the right to make medical decisions is subject to limits. In deciding the extent of rights to make end-­of-­life decisions, the law has recognized that while seriously ill patients have the greatest stake in their medical care, family members, healthcare providers, and the state also have important interests. Accordingly, health law has had to balance a number of competing interests in determining when it is permissible to discontinue medical care or take other actions that will shorten life, and the scope of the right to decide has been much debated. On one hand, the patient’s interests in self-­determination and human dignity point toward a broad freedom to take life-­shortening action. As the US Supreme Court has observed, “at the heart of liberty is the right to define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life.”3 And for patients who lack the ability to make medical decisions, there is an important interest in being treated with respect and having their suffering minimized. When there are competing interests and the benefits and detriments of medical care must be weighed, who better than patients or their families to decide how the balance should play out? Some people may place greater weight on length of life, while others will place greater weight on quality of life. There is no one “correct” way to balance the two interests. Respect for autonomy and dignity are especially important since the patient’s bodily integrity is involved. The law imposes a number of duties on individuals, such as paying taxes or providing a good home for their children, but it rarely forces people to accept unwanted intrusions into their body, as would be the case if a patient were forced to accept unwanted medical care. Similarly, personal interests in autonomy and dignity are heightened by the possibility that continued life will be accompanied by severe and irreversible suffering. Some fates 3  Planned Parenthood v. Casey, 505 U.S. 833, 851 (1992).

Decisions at the End of Life   1075 seem worse than death. Accordingly, the law seeks to ensure that patients need not endure great suffering, and physicians observe a fundamental professional duty to relieve their patients’ suffering. The state’s interests may reinforce the individual interest in refusing treatment or taking a lethal dose of medication. As indicated, healthcare at the end of life may be very costly and divert resources that can provide treatment desired by other patients and that can do much more to promote the public’s interest in better health. On the other hand, there are important arguments against a broad freedom to make decisions that will shorten life. For example, in order to preserve personal autonomy and dignity, the law generally disfavors individual choices that foreclose freedom of choice in the future. Thus, for example, people may not agree to serve as slaves or to renounce their right to a divorce. In addition, it is not clear in which direction dignity points. Under an active notion of human dignity, giving patients and their families wide discretion to make decisions about the timing of death provides the best protection of dignity. But according to a more passive version of human dignity, society fosters human dignity by protecting the disabled and other disadvantaged persons from disfavor with a legal regime that promotes the prolongation of life.4 Patients with serious disabilities might refuse treatment or seek aid in dying not because their suffering is irremediable but because society has failed to fund the services necessary to ensure a good quality of life. In this regard, under Article 25 of the United Nations Convention on the Rights of Persons with Disabilities, “States Parties shall take all appropriate measures to ensure access for persons with disabilities to health services that are gender-­sensitive, including health-­related rehabilitation.” More specifically, under paragraph (f), countries shall “Prevent discriminatory denial of health care or health services or food and fluids on the basis of disability.” Moreover, the state has interests in limiting patient autonomy, most importantly an interest in preserving life. Indeed, the preservation of life is one of the most compelling interests that government promotes. As a corollary, the state has a strong interest in fostering respect for the sanctity of life. If people may choose death, it becomes easier for others to think they ought to choose death. Analogous to the state’s interest in preserving life is the duty of physicians to preserve the lives of their patients. For some physicians, withdrawing life-­sustaining treatment may violate their sense of professional integrity. Considerations of professional integrity become especially sharpened when the question is whether patients may receive a prescription for a lethal dose of medication or a lethal injection from their physicians. But even accepting a broad freedom to make end-­of-­life decisions leaves matters unsettled. For example, before deferring to a patient’s decision, physicians need to ensure that the decision reflects a genuine expression of the patient’s self-­determination. A patient’s refusal of care or a request for lethal medication might be the result of careful

4  Deryck Beyleveld and Roger Brownsword, Human Dignity, Human Rights, and Human Genetics, 61 Modern L. Rev. 661, 662 (1998).

1076   David Orentlicher and Judit Sándor and informed thinking, but it also may be clouded by the side effects of medication, the early stages of dementia, or pressures from family members. And for patients who clearly have lost their ability to make medical decisions, the law has to decide the basis on which decisions should be made. In some cases, patients may have left clear instructions about their preferences in an advance directive or through conversations with family members or providers of healthcare. That may seem decisive, but what about the possibility that the patient may have had a change of heart? And in many cases, the patient may have said little. When a patient has left little, no, or conflicting guidance, how should physicians proceed? Should they turn to the family for a decision, should they decide based on a best interests comparison of the benefits and detriments of the different options, should they follow the preferences of a majority of the public, or should they err on the side of life and provide care in the absence of clear evidence that the patient would not want life sustained? Another factor is the extent to which the freedom to decide rests on the specific interest in preserving bodily integrity as opposed to more general notions of self-­determination and human dignity. If the primary goal is to protect against unwanted intrusions into one’s body, then patients would enjoy a stronger right to have unwanted medical treatment withheld than to receive palliative medication that might shorten life, to swallow a lethal dose of medication, or to have the lethal dose injected by a physician. But if a more general notion of patient autonomy is critical, and if patient self-­ determination is best realized through patient action, then patient authority should be stronger for self-­administration of a lethal drug than for euthanasia or even than for withdrawal of life-­sustaining treatment once the patient has lost decision-­making capacity. While much of the public focus has been on the question of life-­shortening actions, questions also arise with regard to life-­extending actions. Cases often arise in which the patient or the family wants “everything done” to prolong a patient’s life, but physicians, hospitals, or the public believe that further care is unwarranted. The care might provide little benefit at great cost, whether financial or in terms of patient suffering. Does the physician’s authority to practice medicine or society’s interest in allocating its limited resources fairly and prudently mean that some care should not be provided even though the cessation of care would shorten the patient’s lifespan? In the following sections, we discuss the different principles guiding end-­of-­life decisions as they apply in a comparative context to four types of decisions at the end of life: 1. The right of competent patients to have medical care withheld or withdrawn; 2. Withholding or withdrawing treatment from patients who lack decision-­making capacity 3. The right of patients to receive a prescription for a lethal dose of medication for themselves to take or to have a physician inject into them (medical aid in dying); and 4. The authority to withhold desired medical care from patients on grounds of medical “futility.”

Decisions at the End of Life   1077

2  The Right of Competent Patients to Refuse Medical Care: US and Europe The right to have medical care withheld or withdrawn is most straightforward when the patient is an adult possessing decision-­making capacity. In such cases, patients can speak for themselves and directly promote their interest in self-­determination. In general, for such patients, courts and legislatures have recognized that patients are legally entitled to refuse unwanted treatment, even if doing so would result in the patient’s death. Individual interests in autonomy broadly, and in bodily integrity more specifically, mean that the government should not be able to impose unwanted treatment. But there are differences across jurisdictions in the extent of the right to refuse life-­ sustaining treatment. As discussed, while there are important reasons to recognize such a right, there also are countervailing concerns, particularly the state’s fundamental interest in preserving life. Accordingly, in some jurisdictions, the right to refuse treatment depends on the severity of the patient’s illness and the prospects for recovery. In other jurisdictions, even though arguments in favor of a right to refuse typically invoke concerns about the severity of the patient’s illness and the dim prospects for recovery, the right is not qualified by the patient’s prognosis. All competent patients may refuse life-­sustaining treatment. An unqualified right reflects a number of considerations. It is difficult to draw lines in this area. When for example does suffering cross the line from tolerable to intolerable? As one US court wrote, As in all matters, lines must be drawn at some point, somewhere, but that decision must ultimately belong to the one whose life is in issue. . . . It is not a medical decision for her physicians to make. Neither is it a legal question whose soundness is to be resolved by lawyers or judges. . . . It is a moral and philosophical decision that, being a competent adult, is hers alone.5

Also, on matters involving the quality of life, it is dangerous to give government the power to decide whose lives must be preserved and whose need not be. Given inbuilt historical and institutional biases, the state is likely to exercise such a power in ways that favor some patients and disfavor others based on race, income, sex, or other inappropriate factors. Allowing patients to decide for themselves need not put patient welfare at risk. Mentally competent people typically do not refuse life-­sustaining treatment unless they are suffering from irreversible illness. And when they refuse without having an irreversible

5  Bouvia v. Superior Court, 225 Cal. Rptr. 297, 304–305 (Ct. App. 1986).

1078   David Orentlicher and Judit Sándor illness, it is usually on account of religious scruples, as when Jehovah’s Witnesses reject blood transfusions. In the United States, a right to refuse treatment had been gradually adopted from state to state starting in the 1970s.6 Courts generally looked to federal or state constitutional provisions protecting rights to make important personal decisions7 or to the common law right of “informed consent,” under which physicians may not administer treatment to a patient without first informing the patient about the treatment and obtaining the patient’s consent to the treatment. If patients must consent to treatment, then they have a right not to consent. Ultimately, in 1990, the right to refuse treatment was adopted nationwide by the US Supreme Court in the Cruzan case.8 According to the Court, the widespread recognition of the right to informed consent provided a strong basis for interpreting the Constitution’s right to liberty to encompass a right to refuse life-­ sustaining treatment. While the Cruzan opinion suggested that the state’s interest in preserving life might limit the extent of the right to refuse treatment, as when the patient has a treatable illness,9 US law has not relied on the interest in preserving the patient’s life to impose any practical limits on the right to refuse for competent adults. As mentioned, when competent patients refuse life-­sustaining treatment, they typically do so because of irreversible illness or religious objections to treatment. It is a rare case in which a competent patient refuses treatment that will restore good health, other than when treatment would violate religious beliefs. Hence, only in some very early cases did courts override refusals of treatment, and neither the US Supreme Court nor any state supreme court has mandated the provision of life-­sustaining treatment over the objections of a competent adult, absent third-­party interests. In cases in which a court requires treatment, the patient has a communicable disease that threatens public health, the patient is pregnant and treatment is needed to preserve the life or health of the fetus, or the patient is a prisoner and there are concerns about the disruption of institutional order.10 In European health law, too, informed consent is a basic principle. According to Article 8(1) of the European Convention on Human Rights (ECHR), people possess a right to respect for private life. In interpreting that principle and similar declarations of

6  In the first major decision in this area, the Quinlan case in 1976, the New Jersey Supreme Court did qualify the right to refuse in terms of the patient’s medical condition, but it subsequently abandoned that approach in favor of a generally unqualified right. In re Conroy, 486 A.2d 1209, 1223-­25 (N.J. 1985) (the right to refuse treatment outweighs countervailing state interests unless the discontinuation of treatment would “affect the health, safety, or security of others”). 7  In the US Constitution, there is a right to liberty. State constitutions have similar provisions. 8  Cruzan v. Director, Missouri Department of Health, 497 U.S. 261 (1990). 9  Id. at 280. The Nevada Supreme Court has stated that the right to refuse life-­sustaining treatment applies only in the context of irreversible illness, although it did so in a case involving an irreversibly injured patient whose right to refuse treatment was recognized by the court. McKay v. Bergstedt, 801 P.2d 617 (Nev. 1990). 10  There are also cases in which trial courts have ordered blood transfusions for parents with young children, but these decisions have been reversed on appeal.

Decisions at the End of Life   1079 human rights,11 the Council of Europe adopted Article 5 of the Oviedo Convention on Human Rights and Biomedicine, which stipulates that An intervention in the health field may only be carried out after the person concerned has given free and informed consent to it. This person shall beforehand be given appropriate information as to the purpose and nature of the intervention as well as on its consequences and risks. The person concerned may freely withdraw consent at any time.

In elaborating on Article 5 in a parliamentary resolution, the Council of Europe reiterated the principles of personal autonomy and consent, according to which no one can be compelled to undergo any medical treatment against his or her will.12 The resolution lays down guidelines for national parliaments in relation to advance directives, living wills, and continuing powers of attorney. But these requirements for patient consent do not mandate a right to refuse life-­ sustaining treatment. As indicated, one can derive a right to refuse life-­sustaining treatment from the ECHR’s Article 8(1) right to respect for private life, and the ECtHR suggested as much in 2002, in the Pretty case,13 when it wrote that “the imposition of medical treatment, without the consent of a mentally competent adult patient, would interfere with a person’s physical integrity in a manner capable of engaging the rights protected under Article 8(1) of the Convention.” However, Article 8(2) of the Convention recognizes state authority to limit the reach of 8(1) when necessary “for the protection of health.” Ultimately, the ECtHR’s “margin of appreciation” for countries on bioethical matters means that formal recognition of a right to refuse life-­ sustaining treatment has depended on national courts or legislatures. Thus, for example, the right to refuse life-­ sustaining treatment was established in the United Kingdom in 1993 in the Bland case,14 in Germany by the Federal Court of Justice in 2014, and in Italy by the Constitutional Court in 2007.15 In Switzerland, court decisions also recognized a right to refuse treatment (passive Sterbehilfe).16 For examples of legislatively enacted rights, there are Belgium, where the right to refuse life-­sustaining treatment was codified in Article 8(4)

11  For example, the Council on Europe invoked the UN’s Universal Declaration of Human Rights, which includes a right to liberty (Article 3). 12  Parliamentary Assembly Resolution 1859 (2012). 13  Pretty v. the United Kingdom, ECtHR, Application no. 2346/02, judgment of April 29, 2002, final judgment of July 29, 2002. 14  Airedale NHS Trust v. Bland [1993] AC 789. 15  Englaro case at the Corte di Cassazione, https://www.biodiritto.org/Biolaw-­pedia/Giurisprudenza/ Corte-­di-­Cassazione-­sez.-­I-­civ.-­Caso-­Englaro-­interruzione-­dei-­trattamenti-­e-­incapacita, accessed on August, 25, 2019. 16  Georg Bosshard, Switzerland 463, 465 in John Griffiths, Heleen Weyers, and Maurice Adams, Euthanasia and Law in Europe (2d ed. 2008).

1080   David Orentlicher and Judit Sándor of the 2002 Law on Patients’ Rights17; Finland, as part of the 1992 Act on the Status and Rights of Patients; and France, via the 2002 Law on Patients’ Rights, Leonetti Law in 2005, and the Leonetti-­Claeys Law in 2016.18 But not all European countries recognize a broad right of competent adults to refuse life-­sustaining treatment. In Hungary, the Health Care Act, adopted in 1997, introduced the right to refuse life-­saving and life-­sustaining treatment under specific restrictions. For example, the right to refuse life-­sustaining treatment applies only “if the patient suffers from a serious illness which, according to the current state of medical science, will lead to death within a short period of time even with adequate health care, and is incurable.”19 Poland also limits the right to refuse, and it does so by distinguishing between the withholding of treatment and the withdrawal of treatment. While patients may decline the initiation of life-­sustaining treatment, the withdrawal of life-­sustaining treatment is considered to be an illegal act of assisted suicide.20 As the examples of Hungary and Poland suggest, the right to refuse life-­sustaining treatment is much weaker in the Central and Eastern European countries. The issue is not a topic of regular public discussion, and there are relatively few judicial cases in this field. While the right to refuse treatment is well-­established in many countries for the adult possessing decision-­making capacity, the right is less robust when moving beyond this core right. When patients lack decision-­making capacity, jurisdictions tend to be more restrictive in the extent to which they permit the withholding or withdrawing of treatment.21

3  Withholding or Withdrawing Treatment from Patients Who Lack Decision-­M aking Capacity In all legal systems, patients who lack decision-­making capacity are treated differently from those who have mental capacity. When people no longer can speak for themselves, 17  Geneviève Schamps, A Role for Loved Ones to Promote Patients Autonomy at the End of Life 133, 136–137, in Brigitte Feuillet, Kristina Orfali, Therese Callus (eds.), Families and End-­of Life Treatment Decisions: An International Perspective (2013); Maurice Adams, Belgian Law on Euthanasia and Other MBPSL 295, 298, in Griffiths et al., supra note 16. 18  In the 2005 “Loi Leonetti,” France approved the withholding or withdrawal of treatment from patients with terminal illness or other hopeless condition when the harms of providing treatment were disproportionate to the benefits. The language of a 2002 French law provides for a broad right to refuse life-­sustaining treatment, but the French courts have not always applied the law’s provisions faithfully. Stephanie Hennette-­Vauchez, France 371, 380–381, in Griffiths et al., supra note 16. 19  Hungarian Health Care Act no CLIV of 1997 §20. 20  Christian Weber et al. Existential Decision-­Making in a Fatal Progressive Disease: How Much Do Legal and Medical Frameworks Matter?, 16(80) BMC Palliative Care (2017), doi.org/10.1186/ s12904-­017-­0252-­0256. 21  Later in this chapter, we will see that when patients request medical aid in dying, jurisdictions vary in the extent to which they recognize a right.

Decisions at the End of Life   1081 it is not clear how their families or healthcare providers can best promote their interests. And for children, end-­of-­life decisions are even more challenging since the patients never possessed the capacity to make healthcare choices, assessment of suffering is more complicated, and parents often try to do everything to save their children’s lives, even when treatment seems futile. We discuss each category separately here.

3.1  United States: Adults While the US Supreme Court established a nationwide right to refuse treatment in the Cruzan case, it permitted states a good deal of leeway in deciding how the right to refuse should be implemented for patients who cannot make medical decisions for themselves. In such cases, someone else has to decide on the patient’s behalf, and a surrogate decision-­maker is inevitably an imperfect representative for the patient. As states have grappled with the question of how to deal with this imperfection, they have reached a consensus on some questions and taken divergent paths on other questions. Agreement across states exists that if patients have left clear evidence of their preferences regarding healthcare at the end of life, those preferences must be recognized. Importantly, patients can document their preferences in writing through an advance directive. An advance directive can take the form of a treatment directive (commonly called a “living will”), in which individuals indicate how they want to be treated in the event they lack decision-­making capacity. For example, patients may write that in the event of a terminal illness, they would not want to be given life-­prolonging treatment. Alternatively, an advance directive can take the form of a power of attorney, in which an individual appoints a family member or friend as the surrogate decision-­ maker. Surrogate decision-­makers must decide for the patient based on their understanding of the patient’s preferences and the patient’s best interests. Documenting one’s preferences in writing is an important way to provide clear evidence, but not the only way. Courts in the United States will recognize oral statements that patients made in conversation with family members, friends, or healthcare providers. In assessing the statements, courts will consider how recently, how thoughtfully, and how consistently they were made. The specificity of the statements also can be important. Did the patient talk generally about not wanting treatment “if I can’t be myself,” or did the patient express a particular desire not to be artificially fed in the event of advanced dementia? Courts also generally will consider other kinds of evidence. For example, patients’ religious views or their patterns of practice when making medical decisions are relevant. Being guided by clear evidence of a patient’s wishes is not only a common principle of state laws, it likely is required by the US Constitution that states respect a patient’s clearly expressed preferences. In the Cruzan case, by a 5–4 decision, the Court affirmed the earlier ruling of the Supreme Court of Missouri and held that it was acceptable to require “clear and convincing evidence” of a patient’s wishes for removal of life support. In addition, as the concurring opinion of Justice Sandra Day O’Connor suggested, while states

1082   David Orentlicher and Judit Sándor can adopt that strict standard, they would not be able to impose a stricter standard, such as the beyond-­a-­reasonable-­doubt standard of proof for criminal convictions.22 The Court’s decision led to increased efforts to promote the use of advance directives as a way for patients to provide clear evidence of their wishes. Advance directives are important, but they do not always solve the problem. Many advance directive forms give instructions in general terms (e.g., refusing treatment that would “serve only to artificially prolong the dying process”) that can lead to interpretive uncertainty when specific treatment decisions need to be made. Greater discussion of advance directives can be found in the chapter by Donnelly and Berg in this volume. If clear and convincing evidence of a patient’s wishes is lacking, then states take different approaches. In many states, either by legislation or court decision, the law gives family members authority to decide. In other words, in these states, it is as if the patient appointed a family member via a power of attorney. As one court observed about family members, Almost invariably the patient’s family has an intimate understanding of the patient’s medical attitudes and general world view and therefore is in the best position to know the motives and considerations that would control the patient’s medical decisions. . . .  [In addition o]ur common human experience informs us that family members are generally most concerned with the welfare of a patient. It is they who provide for the patient’s comfort, care, and best interests, and they who treat the patient as a person, rather than a symbol of a cause.23

Family decision-­makers are expected to decide based on their understanding of the patient’s preferences, with consideration of the patient’s best interests to resolve uncertainties about the patient’s preferences. Family decision-­making states vary in their definition of “family.” Some include only spouses, parents, adult children, and adult siblings, while others include a broader range of blood relatives. Some states also include close friends or even any “adult who has exhibited special concern for the patient, who is familiar with the patient’s personal values.”24 Objective standards are usually applied when there is insufficient evidence about the patient’s wishes. Many patients do not discuss their treatment preferences before losing decision-­making capacity, and other patients never possessed decision-­making capacity. In such cases, which include decisions made by family members or court-­appointed guardians, the patient’s medical care should be based on an objective assessment of the patient’s best interests. Under a best-­interests analysis, decision-­makers balance the benefits and detriments of treatment to see if the advantages outweigh the disadvantages. Relevant considerations include the extent to which treatment can extend life, the 22  Cruzan, 497 U.S. at 289–292 (O’Connor, J., concurring). 23  In re Jobes, 529 A.2d 434, 445 (N.J. 1987). 24  18-­A Me. Rev. Stat. Ann. § 5-­805(c).

Decisions at the End of Life   1083 likelihood that treatment will be effective, and the quality of life both with and without treatment. Another common approach is for states to vary their rules based on the patient’s condition. If the patient is terminally ill or permanently unconscious, then states tend to allow families to decide, again based on their understanding of the patient’s preferences or, when those are uncertain, in terms of the patient’s best interests. Thus, for example, the New Jersey Supreme Court permitted the husband of Nancy Jobes to authorize withdrawal of her feeding tube after she became permanently unconscious. But if the patient is neither terminally ill nor permanently unconscious, as with patients who suffer from advanced Alzheimer’s disease or a severe neurologic injury, states often require treatment to be provided. Given the uncertainty in determining what the patient would want, these states err on the side of preserving life. An illustrative example is provided by the case of Michael Martin who became seriously disabled intellectually after an automobile accident and dependent on artificial nutrition and hydration. Although his wife testified to numerous discussions in which Mr. Martin had made her promise not to keep him alive if he could not recover from a serious illness or injury, the Michigan Supreme Court did not believe his statements constituted clear and convincing evidence and ordered the continuation of his treatment.25 Note how the patient’s inability to decide about treatment changes the nature of the right to refuse treatment. As mentioned earlier, the right to refuse for a patient with decision-­making capacity generally exists regardless of the severity of the patient’s medical condition. But once the patient loses decision-­making capacity, then the right can vary with the patient’s condition. The more dismal the prognosis, as with terminal illness or permanent unconsciousness, the less stringent the procedural standards for withholding or withdrawing treatment.

3.2  Europe: Adults When the law allows the termination of life-­saving or life-­sustaining treatment, it prioritizes the human element over the medical-­technocratic one. In Europe—perhaps because of the historical experience of the Nazi eugenic practices and the misuse of the term “euthanasia”—not only was this term discredited but also any external assessment of the quality of life. Consequently, autonomy and personal dignity play the most important role in end-­of-­life decisions, and the assessment of the value of life even via medical means has been delegitimized. Instead, the question of where the law allows the termination of life-­saving treatment is resolved through the subjective standard of unbearable suffering. In Europe, as in the United States, the patient’s previously expressed wishes are important in that assessment. Under Article 9 of the Convention on Human Rights and Biomedicine (Oviedo Convention), “the previously expressed wishes relating to a medical 25  In re Martin, 538 N.W.2d 399 (Mich. 1995).

1084   David Orentlicher and Judit Sándor intervention by a patient who is not, at the time of the intervention, in a state to express his or her wishes shall be taken into account.” No further specification of what “taken into account” requires is available, leaving states a wide margin of discretion to adopt different approaches. When a patient’s wishes are unclear, European legal principles point toward family decision-­making. Article 6 of the Oviedo Convention stipulates that “[w]here, according to law, an adult does not have the capacity to consent to an intervention because of a mental disability, a disease or for similar reasons, the intervention may only be carried out with the authorization of his or her representative or an authority or a person or body provided for by law.” In European countries, implementation of this obligation is usually on the basis that the “representative” is, in general, a family member. In addition, the individual patient concerned shall as far as possible take part in the authorization procedure. The Council of Europe has published a Guide on the Decision-­Making Process Regarding Medical Treatment in End-­of-­Life Situations, which addresses these issues. This Guide was drawn up by the Committee on Bioethics of the Council of Europe, with the intention of facilitating the implementation of the principles enshrined in the Oviedo Convention. Another effort to provide a common approach was the promotion of advance directives. In 2009, the Recommendation CM/Rec(2009)11 on “Principles Concerning Continuing Powers of Attorney and Advance Directives for Incapacity” was published, along with a more detailed explanatory memorandum, with the intent to provide some common standards in the field of advance directives. The Guide specifies that treatment covers not only interventions whose aim is to improve a patient’s state of health by acting on the causes of the illness, but also interventions which have a bearing only on the symptoms and not on the etiology of the illness, or which are responses to an organ dysfunction. As the Guide observes, artificial nutrition and hydration would satisfy this concept of treatment and are considered to do so in many countries. However, in other countries, artificial nutrition and hydration are viewed as meeting a patient’s basic needs “which cannot be withdrawn unless the patient, in the terminal phase of an end-­of-­life situation, has expressed a wish to that effect.” While the Guide is based on the Oviedo Convention’s principles, the Convention does not deal expressly with end-­of-­life decisions. In the case of the refusal of treatment by patients with decision-­making capacity, the margin of appreciation on bioethical matters leaves regulation of withdrawal of treatment from incapacitated patients largely to national courts or legislatures. In the countries of the European Union and the Council of Europe, there is no consistent position on the legal recognition of, and limitations on, end-­of-­life decisions. Thus, for example, some countries follow the principles in the Oviedo Convention and base decisions on the previously expressed wishes of the patient, with reliance on family members or other representatives of the patient to authorize medical decisions when the patient’s wishes are not clearly expressed. In the Netherlands, for example, the Medical Treatments Contract Act of 1995 provides for recognition of a patient’s advance

Decisions at the End of Life   1085 directive.26 In the absence of an advance directive, the Act calls for decisions to be made by a legal guardian; a surrogate appointed by the patient; or a spouse, partner, parent, child, or sibling; and all of these agents are obligated to “involve the patient as much as possible” in the decision-­making.27 Under Finland’s 1992 Act on the Status and Rights of Patients, “the patient’s legal representatives, a close relative, or other person closely ­connected with the patient, must give their consent to the treatment. In giving their consent, the patient’s legal representatives, close relative, or other person closely connected with the patient must respect the patient’s previously expressed wishes or, if no wishes had been expressed, the patient’s well-­being.” Similarly, in Germany, healthcare providers would look first to a patient’s advance directive, and, if none exists, then the patient’s family or other surrogate is authorized to decide after consultation with the patient’s physician. The surrogate and physicians would try to identify the patient’s presumed wishes based on prior statements, including conversations about life in general and ethical or religious views.28 Other countries modify the Oviedo principles by giving the patient’s physician a greater role in the decision-­making process. The French Leonetti-­Claeys law takes such an approach.29 If the patient left an advance directive, it would guide treatment, absent situations in which it would be inappropriate to do so. As an alternative to an advance directive, patients can designate a “trusted person” to represent the patient’s wishes. If the patient’s wishes are not clearly expressed in an advance directive or through the trusted person, then healthcare providers, the trusted person, and/or family members engage in a shared decision-­making process for life-­sustaining treatment. An in­de­pend­ ent physician also participates in the process. Ultimately, the healthcare provider team informs the family of its decision, with the family being able to appeal the decision in the administrative courts. Sweden also employs a shared decision-­making process, with a prominent role for the patient’s physician, when the patient’s preferences are not clear. In such cases, healthcare decisions “are made by healthcare professionals, who try to reconstruct the patient’s presumed will in discussion with his or her relatives. The final decision is made by the responsible physician.”30 In some countries, as in some US states, the patient’s right to refuse treatment narrows in the event of incompetence. For example, in Italy, treatment can be withdrawn only if there is clear and convincing evidence that the patient would not want treatment. For most patients, that means having written an advance directive. The development of this standard was tortuous. In the 2007 case of Eluana Englaro, for example, Italy’s Corte di Cassazione31 recognized a right to have treatment withdrawn in the event of 26  Article 450(3), discussed in John Griffiths and Heleen Weyers, Dutch Law on Euthanasia and Other MBPSL 51, 58, in Griffiths et al., supra note 16. 27  Article 465(3), (5), discussed in Griffiths and Weyers, supra note 26, at 57. 28  Weber et al., supra note 20. 29  Jonathan Messika et al., Strengths of the French End-­of-­life Law as Well as its Shortcomings in Handling Intractable Disputes Between Physicians and Families, 26 New Bioethics 53 (2020). 30  Weber et al., supra note 20. 31  Italy’s Supreme Court for civil and criminal matters.

1086   David Orentlicher and Judit Sándor incompetence under two conditions: the patient is permanently unconscious, and it must be clear that the patient would not have wanted the treatment from the patient’s prior statements or other evidence. Even as carefully circumscribed, the case caused considerable political turmoil. As in the Terri Schiavo case in the United States,32 all three branches of state power became involved—the Government, the President, the Constitutional Court and the ECtHR—causing a political crisis.33 Ultimately, the Constitutional Court affirmed the lower court in 2009. Because of the strong influence of the Vatican,34 Italy had been slow to recognize rights to have treatment withdrawn despite constitutional provisions supporting such rights. Euthanasia, both active and passive, was regarded as a taboo. Indeed, Englaro had been unconscious for 17 years before her treatment was withdrawn, and her father’s request to terminate treatment had been rejected multiple times by the courts. In 2017, advance directive legislation expanded the rights of the incompetent patient in Italy. While the 2017 law preserves the Englaro requirement of clear evidence of the patient’s wishes, an advance directive would apply once the patient loses decision-­making capacity, not just upon permanent unconsciousness. The legislation also allows for the appointment of a surrogate decision-­maker (or trustee). Even when countries recognize family decision-­making, other questions remain. In the Lambert and Others v. France35 case (see further discussion in the section on futility and obstinacy), one group of relatives alleged that the withdrawal of Vincent Lambert’s artificial nutrition and hydration was a breach of the state’s obligation to protect the right to life under ECHR Article 2 and constituted ill-­treatment and torture within the meaning of Article 3. One question that this case raised was who should represent the wish of the unconscious relative when there are sharp disagreements between different groups of relatives. Should ideological or religious similarity or difference between relatives matter? Do all relatives have a legal standing? Who is to decide between different positions of the different relatives? It seems that in such cases, the Court will rely on medical opinion on the benefits for the patient as assessed by medical standards rather than the preferences of family members. Similar issues arise in the United States, and while some statutes provide guidance, others do not. As a general matter, one can turn to the fiduciary duties of family members to resolve these questions. Family members have a responsibility to decide based on the patient’s preferences and, to the extent those are uncertain, the patient’s best interests. Healthcare providers should follow the wishes of those family members who are acting most in accordance with their duties to the patient.

32  Mark A. Hall et al., Health Care Law and Ethics 562 (9th ed. 2018). 33  Right-­to-­Die Case in Italy Turns into Political Turmoil Crisis, https://www.welt.de/english-­news/ article3163452/Right-­to-­die-­case-­in-­Italy-­turns-­into-­political-­crisis.html, last accessed March 30, 2020. 34  Sacred Congregation for the Doctrine of the Faith, Declaration on Euthanasia, 1980, available at http://www.vatican.va/roman_curia/congregations/cfaith/documents/rc_con_cfaith_doc_19800505_ euthanasia_en.html, accessed on June 20, 2020. 35  Lambert and Others v. France, ECtHR, Application no. 46043/14, judgment of June 5, 2015.

Decisions at the End of Life   1087 While there are still different approaches to the withdrawal of life-­sustaining treatment in Europe, efforts to achieve a common approach have been most notable in a few areas. Access to pain relief is understood as a basic patient’s right, and there are guidelines for the provision of palliative care. In addition, most European countries have now developed a legislative framework for the rights of the terminally ill that includes provisions on advance directives (e.g., the Health Care Act in Hungary). Furthermore, many countries have adopted special laws on advance directives with or without an appointment of a person whom the patient can trust: for instance, the personne de confiance (trusted person) in France or the patient’s medical representative in Spain.36 In Austria, the Power of Attorney in Health Matters (Vorsorgevollmacht in Gesundheitsangelegenheiten) has a similar function.37

3.3  Europe: Minors As with adults, the Council of Europe’s “margin of appreciation” for countries on bioethical matters means that the recognition of withdrawing treatment from minors depends on the national laws. The assessment of capacity for minors follows different methods in which age, the relevance of the intervention, suffering, and parental decision are taken into account.38 Article 6 of the Oviedo Convention prescribes that “[w]here, according to law, a minor does not have the capacity to consent to an intervention, the intervention may only be carried out with the authorisation of his or her representative or an authority or a person or body provided for by law.” In case of a conflict between the best interests of the child and parental decision-­ making authority, often courts can decide. In Switzerland, for instance, the capacity of discernment serves as the major criterion in the recognition of the right to withdraw medical treatment.39 The tragic Gard case40 from the United Kingdom is illustrative. That case concerned the withdrawal of treatments sustaining the life of a baby (Charlie Gard) whose brain was irreversibly damaged by a rare and apparently incurable disease. His parents wished for him to undergo experimental treatments and opposed the doctors’ opinion, according to which any further therapy would be futile and would only prolong his suffering. 36  Denard Veshi and Gerald Neitzke, Advance Directives in Some Western European Countries: A Legal and Ethical Comparison between Spain, France, England, and Germany, 22 Eur. J. Health L. 321 (2015). 37  Roberto Andorno, Nikola Biller-­Andorno, and Susanne Brauer, Advance Health Care Directives: Towards a Coordinated European Perspective?, 16 Eur. J. Health L. 207 (2009). 38  Brigitte Feuillet-­Liger, Ryuichi Ida, Thérese Callus (eds.), Adolescents, Autonomy, and Medical Treatment: Divergence and Convergence across the Globe (2012). For further discussion, see the chapter by Berg and Cave in this volume. 39  Dominique Manaï, The Right of Adolescent Patients in Switzerland 251, 265, in Feuillet-­Liger et al., supra note 38. 40  Gard and Others v. the United Kingdom, ECtHR, Application no. 39793/17, declared inadmissible on June 28, 2017.

1088   David Orentlicher and Judit Sándor Under UK law, in the event of a dispute between doctors and parents as to the treatments to be administered to a minor patient, it is for the courts to resolve the matter in light of the “child’s best interest” criterion. Applying this criterion, domestic courts had upheld the doctors’ opinion and rejected the parents’ appeals. Before the ECtHR, Charlie Gard’s parents complained under Article 2 ECHR (right to life) of the fact that the hospital was blocking access to life-­sustaining treatments, and under Article 8 ECHR, of the disproportionate interference with their parental rights. The applicants (the parents) complained on their own behalf and on behalf of their child under Articles 2 and 5 of the Convention. They argued that the hospital had blocked life-­sustaining treatment to Charlie Gard in violation of the positive obligation under Article 2. Furthermore, they stated that Charlie was deprived of his liberty under Article 5. The domestic courts had concluded that it would be lawful for the hospital to withdraw life-­sustaining treatment because it was likely that Charlie would suffer significant harm if his present suffering was prolonged without any realistic prospect of improvement and the experimental therapy would be of no effective benefit. The ECtHR bore in mind the considerable “wide margin of appreciation” left to the authorities in the sphere concerning access to experimental medication for the terminally ill and in cases raising sensitive moral and ethical issues, reiterating that it was not for the Court to substitute itself for the competent domestic authorities. The Court also considered that it was most likely Charlie was being exposed to continued pain, suffering, and distress and that undergoing experimental treatment with no prospects of success would offer no benefit and continue to cause him significant harm. Another case involving conflict between parents and medical professionals, this time involving a much older minor child, Afiri and Biddarri v. France,41 concerned the decision to withdraw the life-­sustaining treatment being administered to a 14-­year-­old girl in a vegetative state following acute cardiorespiratory failure. The applicants, her parents, complained in particular of the fact that the decision to withdraw the treatment from their minor daughter ultimately lay with the doctor despite the fact that the parents opposed it. They argued that they should have a right of co-­decision under the collective procedure, in their capacity as the parents and persons with parental responsibility. The ECtHR found that the legislative framework in force complied with Article 2 (right to life) of the Convention and that, even though the applicants disagreed with the outcome of the decision-­making process undertaken by the doctors, the proc­ess had satisfied the requirements of that Article. The Court also found that French law had provided for a judicial remedy that satisfied the requirements of Article 2. The Court considered that, despite the parents’ objections, the continuation of treatment would amount to unreasonable obstinacy (futility) and that the decision of July 21, 2017 had not constituted a serious and manifestly unlawful breach of a fundamental freedom. As can be seen, while the specific Oviedo Convention on bioethics is silent about end-­ of-­life decisions, the general human rights framework can help countries to adopt a

41  Afiri and Biddarri v. France, ECtHR, Application no. 1828/18, rejected on January 23, 2018.

Decisions at the End of Life   1089 careful but gradual approach to these sensitive questions involving end-­of-­life decision-­ making for or by minors.

3.4  United States: Minors In the United States, parents generally have authority to make medical decisions, including decisions about life-­sustaining medical treatment, on behalf of their children. As the US Supreme Court has recognized, parents have a fundamental right to make decisions about the upbringing of their offspring, both because parents are most concerned about the well-­being of their children and also to avoid undue efforts by the state to control the destiny of its citizens. Not all medical decisions for minors must be made by their parents. States typically allow minors to agree to a number of interventions without parental involvement. These interventions generally involve decisions for which it is clearly in the minor’s interest to receive care, as for contraception or treatment for sexually transmitted diseases, drug abuse, or mental illness.42 There also is a concept of “mature minors” for older adolescents who develop decision-­making capacity before the age of adulthood (18 in the United States). While some states have recognized a right to refuse life-­sustaining treatment for mature minors,43 it is still a limited number of states.44 States have not developed a robust mature minor doctrine for decisions that would shorten life, leaving parental authority intact.45 But even when parental authority is recognized, parental rights are subject to reasonable limits, as is the case with rights generally. In particular, the government “can intervene in the parent–child relationship where the health and safety of the child . . . are in jeopardy.”46 Thus, under their child abuse and neglect laws, all US states provide for legal authority to override parental decisions that would constitute medical neglect. The classic cases are those involving families of Jehovah’s Witnesses where the children need blood transfusions and their parents decline the transfusions on religious grounds. In those cases, the courts have held that the children must be given the transfusions.47 In ordering treatment, the courts typically cite Prince v. Massachusetts, where the Court wrote: 42  Sydni Katz, A Minor’s Right to Die with Dignity: The Ultimate Act of Love, Compassion, Mercy, and Civil Liberty, 48 Cal. West. Intl. L. J., 219, 219, 235 (2018). 43  Kristina Orfali, The Rights of Adolescents in Health Care in the USA: A Complex Legal Patchwork and an Evolving Situation 289, in Feuillet-­Liger et al., supra note 38. 44  Ann Eileen Driggs, The Mature Minor Doctrine: Do Adolescents Have the Right to Die?, 11 Health Matrix: The Journal of Law-­Medicine 687, 696–702 (2001). 45  For more discussion, see the chapter by Berg and Cave in this volume. 46  Newmark v. Williams, 588 A.2d 1108, 1116 (Del. 1991). 47  See, e.g., In re McCauley, 565 N.E.2d 411 (Mass. 1991) (8-­year-­old child needed blood transfusion as part of treatment for an acute leukemia); In re Cabrera, 552 A.2d 1114 (Pa. Super. Ct. 1989) (6-­year-­ old child with sickle cell anemia who had already suffered two strokes needed weekly blood transfusions over at least a year’s time to reduce the risk of a recurrence of her strokes from 70% to 10%).

1090   David Orentlicher and Judit Sándor Parents may be free to become martyrs themselves. But it does not follow they are free, in identical circumstances, to make martyrs of their children before they have reached the age of full and legal discretion when they can make that choice for themselves.48

The hard question in this area is where the limits of parental authority end and when a refusal of treatment becomes child neglect or abuse. At either end of the spectrum, the law is fairly clear. Courts have had no trouble holding that, when a child can readily be restored to good health or when treatment poses little risk, parents may not refuse a life-­ sustaining treatment, such as a blood transfusion,49 antibiotics,50 or surgery to repair an intestinal perforation.51 Similarly, courts have had no trouble holding that parents may refuse life-­ sustaining ventilators or feeding tubes when the child is irreversibly unconscious.52 Other decisions, however, are not so clear. Some treatments will have a low likelihood of success and may carry substantial risks. As the case law indicates, the courts have not come to a consensus on the extent of parental discretion. Consider, for example, the following two cases. In one of the cases, a 3-­year-­old boy was diagnosed as having an aggressive malignancy of the immune system, Burkitt’s lymphoma. By the time a diagnosis was made during surgery for an intestinal obstruction, the cancer had spread elsewhere in the boy’s body. His physician recommended that he be treated with a heavy regimen of chemotherapy. According to the medical testimony, there was a 40% chance of a “cure.” Without treatment, he would likely die within 6–8 months. Medical testimony also indicated that the treatment itself was quite toxic and might prove fatal. The boy’s parents declined the chemotherapy in favor of care from a Christian Science practitioner. Given the substantial risks of treatment and the less than 50% chance of success, the court upheld the parental refusal of treatment.53 In the other case, a 12-­year-­old girl was diagnosed with Ewing’s sarcoma, a bone cancer for which the girl had a 25–50% chance of long-­term remission with treatment. Without treatment, she would likely die within 6–9 months. Her father, a lay minister of the Church of God of the Union Assembly, refused treatment on religious grounds. Given the apparent certainty of death without treatment and the reasonable possibility 48  321 U.S. 158, 170 (1944). 49  McCauley and Cabrera, supra note 47; In re L.S., 87 P.3d 521 (Nev. 2004). 50  Walker v. Superior Court, 763 P.2d 852 (Cal. 1988). 51  Commonwealth v. Twitchell, 617 N.E.2d 609 (Mass. 1993). 52  See, e.g., In re L.H.R., 321 S.E.2d 716 (Ga. 1984) (permitting the removal of a ventilator from a several-­month-­old child in an irreversible coma); In re P.V.W., 424 So. 2d 1015 (La. 1982) (permitting the removal of artificial life supports from an infant who had been irreversibly comatose since birth); In re Guardianship of Crum, 580 N.E.2d 876 (Ohio Prob. Ct. 1991) (permitting the removal of a feeding tube from a 17-­year-­old who was irreversibly unconscious); In re Guardianship of Myers, 610 N.E.2d 663 (Ohio Prob. Ct. 1993) (permitting the removal of a feeding tube from a minor, apparently an adolescent, who was irreversibly unconscious). 53  Newmark, supra note 46.

Decisions at the End of Life   1091 of long-­term success with treatment, the court overrode the parental refusal of treatment. The court seemed to consider the fact that the family’s religious sect did not refuse all kinds of medical treatment.54 In other illustrative cases, courts have permitted a mother to refuse aggressive antiviral drug therapy to treat HIV infection in her 4-­year-­old son and deferred to a parent’s refusal of antipsychotic drug for a 3-­year-­old child, but denied the parents’ request to withdraw a ventilator from a 14-­year-­old child who had a lethal genetic disease and a life expectancy of no more than 2 years but who was still “aware and enjoy[ed] TV and videos.”55 In other cases, the courts imposed treatment for cancers that can be quite responsive to treatment.56 Many states provide exemptions in their child abuse and neglect laws for parents who refuse medical treatment on religious grounds. In some of these states, the exemptions excuse parents from liability when their children suffer from the denial of care, though generally not when the child’s life is endangered. In addition, the exemptions typically apply to child abuse and neglect laws but not so often to other laws under which parents may be held accountable, such as involuntary manslaughter statutes.57 And, of course, the exemptions speak to the ability of the state to punish parents. Even if a court cannot sanction the parents, it still may order that treatment be provided to the child. Indeed, the federal regulation that led to the adoption of the religious exemptions explicitly drew a line between holding parents liable and ensuring access to care for children. The regulation, which no longer is in effect,58 tied federal funding for child abuse programs to the enactment of religious exemptions. But the regulation also stated that the exemptions “shall not preclude a court from ordering that medical services be provided to the child, where his health requires it.” State law exemptions typically track this requirement. In practice, though, the judicial power to order treatment does not always provide adequate protection for children. There are many commonalities between the US and European legal approaches to both refusal of medical treatment and to withholding or withdrawing treatment. Each recognizes the autonomy and liberty of competent adults but exercises considerably more caution where noncompetent adults or minors are concerned. Irrespective of the legal approach taken, it is also important to recognize the practical, non-­legal barriers that can exist in the withholding or withdrawing of life-­sustaining treatment. Some arise from physician practices which traditionally rest on the professional imperative of preserving life even when prolonged artificially. Professional integrity seems to require different therapeutic options and various forms of palliative care. This is all the more so where patients are not able, or legally competent, to speak for themselves. In theory, the 54  In re Hamilton, 657 S.W.2d 425 (Tenn. Ct. App. 1983). 55  In re Nicholas E., 720 A.2d 562 (Me. 1998); In re Martin F. v. D. L., 820 N.Y.S.2d 759 (N.Y. Sup. Ct. 2006); In re Gianelli, 834 N.Y.S.2d 623, 629 (NY. Sup. Ct. 2007). 56  In re Cassandra C., 112 A.3d 158 (Conn. 2015); P. J. ex rel Jensen v. Wagner, 603 F.3d 1182 (10th Cir. 2010). 57  State v. Neumann, 832 N.W. 2d 560 (Wis. 2013). 58  Critics of the regulation effectively lobbied government officials to remove the regulation.

1092   David Orentlicher and Judit Sándor proper resolution of a difference of opinion between medical practitioner and patient in terms of end-­of-­life care should be legally resolved by transferring care to a different physician. But this is not always feasible in practice or enforceable by law. End-­of-­life practices in Hungary, for example, illustrate the gap between legal rights and the reality of everyday health practice. Since the right to refuse treatment was only reluctantly accepted by the medical institutions, and people did not know how to enforce their rights, many of these provisions remained a mere legal possibility without actually being implemented in practice. While decisions to withdraw treatment have become much more routine over the past several decades, more “active” measures that shorten life remain controversial. In the next section, we take up the practice of medical aid in dying, which includes practices also known as physician-­assisted suicide or euthanasia.

4  Medical Aid in Dying In her novel on the death of her mother, Simone de Beauvoir stated that, at the end of life, there is a competition between suffering and death. If death wins the game, there will be less suffering. But if suffering takes the lead and death is prolonged, then humans lose their dignity and they go through a cruel journey of dying. Legally speaking, in this situation, the right to life competes with the right to dignity. While doctors’ primary duty is to preserve life, their role in offering dignified death based on the patient’s request is still limited. Traditionally, much weight has been placed on the distinction between “passive” and “active” practices that might hasten a patient’s death. On one hand, courts and legislators were generally receptive to the claim that patients should be able to refuse unwanted medical treatment even though they might die as a result. On the other hand, lawmakers were more reluctant to permit the more active step of ingesting a lethal dose of medication prescribed by a physician or arrangements to have a physician inject a patient with a lethal dose (the two main types of medical aid in dying). The distinction between the passive and the active is common in ethics. Acts are generally more blameworthy than omissions. But some omissions also are unacceptable, legally as well as morally. Starving an infant until it dies is a crime just as is feeding it poison. Moreover, withdrawing a ventilator or other life-­sustaining treatment from a patient is a morally relevant act. Or consider a medical treatment, such as palliative pain relief, that might shorten life. Under the principle of “double effect,” it is permissible to provide such care even at the risk of hastening death. So the distinction between withdrawing treatment and providing medical aid in dying requires justification beyond the difference between the passive and the active. One likely reason for the distinction lies in the justifications for a right to refuse life-­ sustaining treatment. Recall that arguments for the right reflect concerns about patients suffering from serious and irreversible illness. In such circumstances, it is thought,

Decisions at the End of Life   1093 patients should not have to accept medical treatment that will only prolong their suffering without curing their illness. But lethal doses of medications can be ingested by people who are not in fact suffering from irreversible illness. Allowing aid in dying runs a greater risk than does withdrawal of treatment that people’s lives will be ended prematurely. While allowing aid in dying risks premature deaths, banning aid in dying risks intolerable suffering. There are patients suffering from serious and irreversible illness who do not need medical treatment to sustain their lives. For them, the right to refuse treatment does not address their needs. Jurisdictions that have faced the question of medical aid in dying have responded primarily in two different ways. The more common approach is to forbid medical aid in dying entirely. In these jurisdictions, the risks of permitting aid in dying seem to exceed the risks of prohibiting the practice. This was the view taken by the US Supreme Court in the Glucksberg59 and Quill60 cases when it rejected a constitutional right to aid in dying.61 In terms of constitutional doctrine, the Court observed that while the right to refuse life-­sustaining treatment grew out of a “long legal tradition protecting the decision to refuse unwanted medical treatment,” the law had long rejected a right to aid in dying.62 Similarly, in the Pretty case,63 the ECtHR observed that the authority of the state to protect health under Article 8(2) of the European Convention extended to bans on aid in dying. At stake was the ability of the government to “safeguard life” against the risk that a right to aid in dying would pose to the “weak and vulnerable.” In Pretty, Diane Pretty sought aid in dying because she was suffering from a serious motor neurone disease, a degenerative disease that gradually affects the muscles. Given that the final stages of the disease are distressing and undignified, she wished to be able to control how and when she died. Because of her disease, the applicant could not commit suicide alone and wanted her husband to help her. But although it is not a crime in English law to commit suicide, assisting a suicide is a crime. Because the authorities refused her request, the applicant complained that her husband had not been guaranteed freedom from prosecution if he helped her die. As indicated, the European Court relied on the state’s authority to protect health under Article 8 of the Convention. In addition, observed the Court, it would distort the idea of Article 2’s right to life to interpret it as conferring the diametrically opposite right: namely, a right to die. The Court also held that there had been no violation of Article 3 (prohibition of inhumane or degrading treatment) of the Convention. Even if it 59  Washington v. Glucksberg, 521 U.S. 702 (1997). 60  Vacco v. Quill, 521 U.S. 793 (1997). 61  Rosamond Rhodes, Physicians, Assisted Suicide, and the Rights to Live or Die 165, in Margaret P. Battin, Rosamond Rhodes, and Anita Silvers (eds.), Physician Assisted Suicide: Expanding the Debate (1998). 62  In rejecting a constitutional right to aid in dying, the Court left the states free to legalize aid in dying by statute or via recognition of a right under the state constitution. 63  Pretty, supra note 13.

1094   David Orentlicher and Judit Sándor could not but be sympathetic to the applicant’s apprehension that without the possibility of ending her life, she faced the prospect of a distressing death, nonetheless, the positive obligation on the part of the state which had been invoked would require that the state sanction actions intended to terminate life, an obligation that could not be derived from Article 3. The Court lastly held that there had been no violation of Articles 9 (freedom of conscience) or 14 (prohibition of discrimination) of the Convention. Although this Court case did not provide recognition of a right to aid in dying, the Pretty case is considered a landmark decision because it recognized that the right to die exists within the framework of respect for private life. In subsequent cases, the EtCHR followed its decision in Pretty, rejecting other claims for a right to aid in dying. For example, in Haas v. Switzerland,64 the applicant, who had been suffering for about 20 years from a serious bipolar affective disorder, wished to end his life and complained of being unable to obtain the lethal substance required for that purpose without a medical prescription. The Court held that there had been no violation of Article 8. The applicant had twice attempted suicide and had stayed in psychiatric hospitals on several occasions. On July 1, 2004, he became a member of Dignitas, an association which offers, among other services, assisted suicide. Taking the view that his illness, for which treatment is difficult, made it impossible for him to live with dignity, the applicant asked Dignitas to assist him in ending his life. He approached several psychiatrists to obtain the necessary lethal substance, namely 15 grams of sodium pentobarbital, which is available only on prescription, but was unsuccessful. Similarly to other cases in matters of bioethics, the Court concluded in this case that states have a wide margin of appreciation in such matters. Although the Court further accepted that the applicant might have wished to commit suicide in a safe and dignified manner and without unnecessary pain, it nevertheless considered that the requirement under Swiss law for a medical prescription in order to obtain sodium pentobarbital had a legitimate aim, namely to protect people from taking hasty decisions and to prevent abuse, the risks of which should not be underestimated in a system that facilitated access to aid in dying. The Court considered that the requirement of a prescription, issued on the basis of a thorough psychiatric assessment, was a means of satisfying the obligation on states to put in place a procedure capable of ensuring that a person’s decision to end his or her life did in fact reflect his or her free will. As regards the question whether the applicant had had effective access to a medical assessment that might have allowed him to obtain sodium pentobarbital (if not, his right to choose when and how he died would have been theoretical and illusory), the Court was not persuaded that it had been impossible for him to find a specialist willing to assist him as he had claimed. In Koch v. Germany,65 the applicant alleged that the refusal to allow his wife (who was paralyzed and needed artificial ventilation) to acquire a lethal dose of medication so that 64  Haas v. Switzerland, ECtHR, Application no. 31322/07, judgment of January 20, 2011, final judgment of June 20, 2011. 65  Koch v. Germany, ECtHR, Application no. 497/09, judgment of July 19, 2012, final judgment of December 17, 2012.

Decisions at the End of Life   1095 she could take her own life had breached her right, and his, to respect for their private and family life. He also complained of the domestic courts’ refusal to examine his complaints on the merits, and the Court found a violation of Article 8 on that point only. The facts of the Sanles Sanles v. Spain case66 remind us of the movie Mare Dentro (in English “The Sea Inside”) directed by Alejandro Amenábar, where a person who had been tetraplegic following an accident committed suicide 30 years later with the help of a friend. The applicant in Sanles, who was appointed by her brother-­in law, requested the right to dignified life or a dignified death or non-­interference with her brother-­in-­law’s wish to end his life. The case was held inadmissible because the applicant had not been directly affected by the alleged violations of the Convention and therefore could not be considered a victim. Though a right to aid in dying has been rejected at the constitutional or human rights level in the United States and almost all of Europe, there are countries in which courts have reached a different conclusion. For example, in the Carter case, the Canadian Supreme Court held that when the government bans aid in dying for competent adults “who seek such assistance as a result of a grievous and irremediable medical condition that causes enduring and intolerable suffering,” the law violates the rights to life, liberty and security under Section 7 of the Canadian Constitution’s Charter of Rights and Freedom.67 According to the Court, The right to life is engaged where the law or state action imposes death or an increased risk of death on a person, either directly or indirectly. Here, the prohibition deprives some individuals of life, as it has the effect of forcing some individuals to take their own lives prematurely, for fear that they would be incapable of doing so when they reached the point where suffering was intolerable. The rights to liberty and security of the person, which deal with concerns about autonomy and quality of life, are also engaged. An individual’s response to a grievous and irremediable medical condition is a matter critical to their dignity and autonomy. The prohibition denies people in this situation the right to make decisions concerning their bodily integrity and medical care and thus trenches on their liberty. And by leaving them to endure intolerable suffering, it impinges on their security of the person.

Subsequent to the Carter decision, the Canadian government passed legislation to implement the right. Under the statute, patients may choose aid in dying when (a) they are adults possessing decision-­making capacity; (b) they have a serious, advanced, and irreversible illness; (c) their suffering has become intolerable; and (d) their death has become reasonably foreseeable.68 Germany followed a cautious piecemeal approach to aid in dying. The historical burden of the misuse of the term of “euthanasia” before and during World War II for murder cases and for various eugenic practices made it long impossible to raise the issue of 66  Sanles Sanles v. Spain, ECtHR, application no. 48335/99, declared inadmissible on October 26, 2000. 67  Carter v. Canada, [2015] 1 SCR 331. 68  Statutes of Canada 2016, ch. 3.

1096   David Orentlicher and Judit Sándor accepting some forms of euthanasia which had been discredited during the Nazi era. In Germany, the judiciary played a very important role in accepting some forms of aid in dying. Since 2014, the Federal Constitutional Court legalized passive euthanasia by means of the withdrawal of life support to patients who request it. Aid in dying remained illegal. The further step occurred in 2017, when the Federal Administrative Court ruled that, in extreme circumstances, it was illegal for an agency to deny access to life-­ending substances. The constitutional basis was the right to personal self-­determination. In 2019, the Federal Court of Justice stated that doctors have no obligation to prevent the death of a person who, pursuant to a valid autonomous decision, attempts suicide. The most recent development was the decision of the Federal Constitutional Court on February 26, 2020, which stated that penalizing assisted suicide services is unconstitutional and void because it would violate the fundamental right to personal self-­ determination. German’s Federal Constitutional Court thus recognized a right to aid in dying for terminally ill or other gravely ill patients. The option for patients to go to other countries for aid in dying initiated in 2017 a new debate in Italy about the legality of aid in dying and led to recognition of a right in Italian law via judicial decision. After a car accident, a prominent disc jockey, Fabiano Antoniani, became blind and tetraplegic. Since he could not receive assistance in his request to die in Italy, he traveled to Switzerland for aid in dying, assisted by pro aid-­in-­ dying campaigner Marco Cappato who was prosecuted for offering this assistance. The legal debate on Antoniani’s case continued even after his death. In 2019, Italy’s constitutional court decided that aid in dying should be permitted by law in the country in certain circumstances, including those in which a patient’s irreversible condition was “causing physical and psychological suffering that he or she considers intolerable.” Absent a constitutional or human right to aid in dying outside of Canada, Germany, and Italy, rights to aid in dying have been recognized on a statutory basis, jurisdiction by jurisdiction. Several European countries and US states have adopted aid-­in-­dying laws.69 In the United States, about 10 states have legalized aid in dying, and all of them have limited the right in the same manner: patients must be adults with decision-­making capacity, must self-­administer the lethal medication, and must be terminally ill (typically defined as having an expected remaining life of no more than 6 months).70 These three requirements serve important purposes. First, they are designed to minimize the risk that aid in dying would be used by someone who is not suffering from a serious and irreversible illness. Indeed, by definition, a person with a terminal illness has a serious and irreversible illness. In addition, by limiting aid in dying to adults with mental capacity who can self-­administer the medication, states can better ensure that the choice of aid in dying reflects a genuine expression of self-­determination by the patient. The degree or nature of the patient’s suffering is not part of the assessment. Eligibility for aid 69  In the US state of Montana, the state supreme court held that there was no legal prohibition against aid in dying for terminally ill patients by self-­administration. Baxter v. State, 224 P.3d 1211 (Mont. 2009). 70  Patients also must be residents of the state.

Decisions at the End of Life   1097 in dying turns on life expectancy, mental capacity to make medical decisions, and physical ability to self-­administer medication. While the three safeguards are consistently adopted from state to state, they are not without controversy. As critics observe, the requirement of self-­administration denies aid in dying to patients who are suffering from a terminal illness but physically unable to take the lethal dose. Or patients might be suffering from a serious and irreversible illness and have a life expectancy of more than 6 months and therefore not be considered terminally ill under an aid-­in-­dying statute. States supplement the three basic safeguards with other protections. After the treating physician has confirmed the patient’s eligibility, a second physician also must confirm eligibility. Patients must reaffirm their desire for aid in dying over at least a 15-­day period and document their wishes in writing. If there is reason to believe that a psychological condition is influencing the patient’s decision-­making, then referral for counseling is required. In Hawaii, all aid-­in-­dying patients must undergo evaluation by a mental health specialist. Oregon has the longest experience with aid in dying in the United States, dating back to 1998.71 Fewer than 1% of deaths in the state result from aid in dying. The most common diagnosis has been cancer, and the patients have been similar to other dying patients with regard to sex, race, urban or rural residence, health insurance coverage, and hospice enrollment. The level of education has been higher for patients who choose aid in dying. The most common concerns of patients choosing aid in dying have been loss of autonomy, decreased ability to participate in enjoyable activities, and loss of dignity. In Europe, the Netherlands was the first country that openly acknowledged and gradually decriminalized medical aid in dying—provided that specific conditions are met. Aid in dying is defined as deliberate termination of the life of a person on his or her request by another person. As a result of this process, it does not constitute a criminal offence for physicians to perform aid in dying provided they comply with the due care criteria specified in the Act on Termination of Life on Request and Assisted Suicide (Review Procedures). Physicians have to report each case. In accordance with the due care criteria, the physician must, among other things, be satisfied that the patient’s request for aid in dying is voluntary and well-­considered and that the suffering is unbearable with no prospect of improvement. The legislative development in the Netherlands was based on published reports on the use of aid in dying in the Netherlands, notably the Remmelink Report 1991.72 Ten years later, the first law came into force. Since then, ­compliance with guidelines for aid in dying has improved. The main rationale behind the 71  The Oregon Death with Dignity Act (DWDA) was originally approved by voters in 1994, blocked by legal challenges, and reapproved by voters in 1997. It went into effect in 1998, and it allows terminally ill adult Oregonians to obtain and use prescriptions from their physicians for self-­administered, lethal doses of medications. 72  Jan Remmelink was Attorney General of the High Council of the Netherlands. He headed the Dutch government committee on euthanasia that released the first official study of the practice of euthanasia in the Netherlands in 1991.

1098   David Orentlicher and Judit Sándor legislation was to make medical decisions transparent. It has become evident that aid in dying, both physician-­administered and patient-­administered, are regular medical practices in the Netherlands. However, it is also important to emphasize that aid in dying generally can be performed only at the patient’s own request, not at the request of relatives or friends.73 In addition to cases where patients suffer from an incurable disease, there has been also a public debate on the issue of whether people who consider their life as completed should be given help in ending their life in a dignified manner and at a time of their own choosing. The debate concerns people for whom life has lost its meaning and become too great a burden for them to continue living. These are generally people of an advanced age who no longer see any way to make their lives meaningful; who have difficulty coping with their loss of independence and loss of mobility; who feel lonely, partly due to the loss of loved ones; and who suffer from general fatigue, physical decline, and loss of personal dignity. It is about people who no longer want to live, despite receiving attention and help from others as well as good-­quality care. European countries have made efforts to proceed gradually and cautiously in developing laws on end-­of-­life decisions. In Belgium, for example, there are about 2,300 cases a year in which doctors terminate the life of a patient through aid in dying. The law permitted aid in dying first for adults in the Belgian Law of May 28, 2002,74 which allows a doctor to perform aid in dying based on the patient’s explicit request and provided that strict conditions are met. In 2014, the scope of the right to aid in dying was extended to some minors. Aid in dying is allowed for minors only if they are capable of making a legally competent and conscious decision and when they are in the condition of constant and unbearable physical or mental suffering resulting from a serious and incurable disorder caused by illness or accident. The Belgian law offers also the option of advance directives for those cases when patients would like to make a decision for the time when they will no longer able to decide themselves. Cautious legislation notwithstanding, it is inevitable that certain hard cases shake the public, both in Europe and in the United States, and these cases might shape—even distort--the law or have a significant impact on its implementation. Just as the US case of Brittany Maynard spurred the adoption of aid in dying legislation in California and other states, the case of Fabiano Antoniani in Italy had an impact on transforming the thinking on aid in dying. The emotional effect of young people dying mobilizes public discussion. European countries have taken different approaches than have the US states to the legalization of aid in dying. In the Netherlands, Belgium, and Luxembourg, for example, patients need not be terminally ill. The recent Court decision in Germany may also

73  There are limited aid-­in-­dying provisions for minors that allow for parental request in exceptional cases. For older minors (starting at age 16), only the minor’s consent is required. For younger minors, parental consent also is required, and, for the youngest minors, parental consent is sufficient. 74  Belgian Law of 28 May 2002 on Euthanasia, amended by the Law of February 13, 2014.

Decisions at the End of Life   1099 point in this direction.75 Rather, a patient is eligible for aid in dying if the patient is incurably ill and experiencing unbearable suffering that cannot be relieved. According to Dutch law,76 for example, the patient must have “unbearable suffering, without prospect of improvement,” and there must be “no reasonable alternative” to relieve the symptoms. Compared to the United States, this is more of a subjective approach. Instead of asking how long the patient has to live, the question is how severe is the patient’s suffering. The European ethical-­legal approach considers the matter from the patient’s perspective, how much the patient may bear, and where are the limits of his or her dignity. Still, even the most radical Dutch and Belgian laws apply some medical standards in the assessment of the requests for aid in dying. Both laws prescribe a consultation with another physician in assessing the patient’s condition. There is good reason to think that a more “objective” requirement of terminal illness provides a better safeguard than does a more “subjective” requirement of unbearable suffering. By requiring a terminal illness, the law can make it more likely that the patient really is suffering from a serious illness that is incurable and irreversible. It also helps ensure that the desire for aid in dying reflects a genuine expression of self-­determination rather than the irrational choice of someone suffering from mental illness. Many controversial cases of aid in dying, both carried out legally in Europe and illegally in the United States, have involved patients who were not terminally ill and who appeared to need psychiatric care. But some patients may have irremediable psychological suffering. Aid in dying in psychiatry constitutes a special case. It follows from the standard of unbearable suffering that there might be cases where the reason for suffering originates not from a somatic but a mental condition, and indeed the links between these are as yet poorly understood. Dierickx et al. described 179 psychiatric and dementia patients, with no comorbid physical illnesses motivating their request, who were accepted for aid in dying in Belgium between 2002 and 2013.77 At the time of writing this chapter, Mortier v. Belgium78 is still a pending case. It concerns aid in dying for a psychiatric patient without the relatives’ consent. In this case a doctor administered lethal medication to the applicant’s mother, who was in a state of chronic depression, without informing the applicant or his sister beforehand. The ECtHR gave notice of the application to the Belgium Government and put questions to the parties under Articles 2 (right to life) and 8 (right to respect for private and family life).

75  German Court Repeals Ban on Assisted Suicide Services, The Guardian, February 26, 2020, available at https://www.theguardian.com/society/2020/feb/26/german-­court-­repeals-­ban-­assisted-­ suicide-­services accessed June 26, 2020. 76  Dutch Termination of Life on Request and Assisted Suicide (Review Procedures) Act. 77  Sigrid Dierickx et al., Euthanasia for People with Psychiatric Disorders or Dementia in Belgium: Analysis of Officially Reported Cases, 17(203) BMC Psychiatry (2017) doi.org/10.1186/ s12888-­017-­1369-­0. 78  Mortier v. Belgium, ECtHR, Communicated Case no. 78017/17 (at the time of writing this chapter, the decision is still pending).

1100   David Orentlicher and Judit Sándor In another important difference with the United States, the Benelux countries do not restrict aid in dying to patient self-­administration. Rather, patients or their physicians may administer the lethal dose. Other aspects of the Benelux laws mirror the US approach. A patient’s request must be “informed” and “voluntary and well considered”; doctors must meet procedural requirements and they must apply the lethal medications properly; and they must consult an independent physician to confirm the patient’s eligibility for aid in dying. It is well known that Switzerland does not prohibit self-­administration, though it does prohibit physician administration. Aid in dying is legal if it is offered without selfish motive to a person with decision-­making capacity. One specificity of this system is that aid in dying is not medically aided as most assistance is provided by lay right-­to-­die associations, not medical professionals. Not only do Swiss citizens choose this form of death, but patients with serious suffering come from other countries as well. Canada’s approach to aid in dying follows the US model in some respects and the Benelux model in others.79 The current legal solution was achieved through a hybrid course, a combination of a constitutionally based court decision and legislation required by the decision to legitimize aid in dying. As mentioned, and in common with the United States, Canada restricts aid in dying to adults who possess decision-­making capacity and who are terminally ill, though Canada has a different definition of terminal illness (the patient’s death must be “reasonably foreseeable” rather than expected within 6 months). Like the Benelux countries, Canada allows physicians and nurse practitioners to administer the lethal dose rather than requiring self-­administration. And health practitioners administer the medication in virtually all cases. The same is true in the Netherlands. We do not know exactly why few cases involve self-­administration when physicianor nurse-­administration is an option, but a Canadian government report suggests some factors: many physicians and nurses “are less comfortable with self-­administration due to concerns around the ability of the patient to effectively self-­administer the series of medications, and the complications that may ensue.” As a result, aid in dying protocols at some Canadian health institutions do not encourage self-­administration. An important question for states in the United States that have legalized aid in dying is whether physician administration also should be allowed. It may be that the requirements of a terminal illness and decision-­making capacity would provide sufficient protection against abuse, with physician administration protecting against complications. In addition, physician administration allows patients who are physically unable to self-­ administer to choose aid in dying. In sum, then, on the two major differences between the legalizing European countries and the US states, Canada splits the difference. As in the United States, Canada prefers a more objective standard for determining when patients are sufficiently ill that they can 79  The Supreme Court of Canada recognized a right to aid in dying as part of the right to liberty under the Constitution of Canada’s Charter of Rights and Freedoms, and the Parliament subsequently passed enabling legislation.

Decisions at the End of Life   1101 choose aid in dying. Patients must be terminally ill, rather than suffering greatly from an irreversible condition. As in Europe, Canada permits health providers to administer the lethal drug. While there are important differences in the ways different countries have legalized aid in dying, overall, the majority of European countries and US states are on the same page: aid in dying is prohibited. While countries remain divided on whether suffering patients should have access to aid in dying, there is much more commonality on the need to ensure that these patients have optimal access to palliative care. Thus, for example, France’s Leonetti-­Claeys Law stipulates that everyone has the right to have a dignified end of life and the best possible relief of suffering and that health professionals must use all means at their disposal to ensure that this right is respected. The law also prescribes that patients should have full access to painless medical intervention. The right to palliation specifically includes a right to palliative sedation. For patients with a serious and irreversible condition whose suffering is resistant to treatment, it is permissible to administer a deep and continuous sedation and withdraw life-­sustaining treatment. Similarly, when the US Supreme Court rejected a constitutional right to aid in dying, a concurring Justice Sandra Day O’Connor observed that a “patient who is suffering from a terminal illness and who is experiencing great pain has no legal barriers to obtaining medication, from qualified physicians, to alleviate that suffering, even to the point of causing unconsciousness and hastening death.” Some states have gone further and recognized a right to palliative care in their legislatures and courts. Florida, California, New York, and other states have passed statutes requiring physicians to advise terminally ill patients of their options for end-­of-­life care, including hospice care, and treatment to relieve pain and other symptoms.80 In addition to these statutory rights, patients may be able to recover damages from doctors and other healthcare providers on the basis of tort law claims for the failure to provide adequate pain relief.81

5  From Futility to Unreasonable Obstinance While court cases about end-­of-­life medical decisions generally have involved situations in which the patient or the family has chosen to hasten death, cases also arise in which the patient or family wants to continue with treatment, but the physicians or hospital want to withdraw care or withhold additional care. In such cases, the patient or family asserts the patient’s right to make medical decisions, but the physicians or hospital argue

80  Alan Meisel, Kathy L. Cerminara and Thaddeus M. Pope, The Right to Die: Law of End-­of-­Life Decisionmaking (2004), §6.03[I] (3d ed. 2015 Supp.); Fla. Stat. §765.1103; Cal. Health & Safety Code §442.5; N.Y. Pub. Health Law §2997-­c. 81  Tolliver v. Visiting Nurse Ass’n of Midlands, 771 N.W. 2d 908 (Neb. 2009).

1102   David Orentlicher and Judit Sándor that the treatment is medically “futile,” that it does not provide sufficient—or any— medical benefit and therefore ought not to be provided. Examples of futility cases have included permanently unconscious patients who needed artificial nutrition or artificial ventilation (the Lambert case in France and the Wanglie case in the United States) or a patient with end-­stage disease for whom doctors wanted to withhold cardiopulmonary resuscitation (CPR) in the event of a cardiac arrest (the Jane Doe case in the United States).82 In such cases, one could say, patients have the right to decide whether to accept or refuse treatment offered to them, but they do not have the right to decide which treatments should be offered. That would effectively have them engaged in the practice of medicine. And, as discussed earlier, the right to refuse life-­sustaining treatment grew out of the right of informed consent, not a more general right to make medical decisions. Even in Europe, where there is greater recognition of rights to healthcare, there is no right to a specific treatment. Rather, questions of the specific treatments to be provided are matters for the medical profession or national health systems. On the other hand, one also could say that life-­sustaining treatment always is appropriate healthcare. If a ventilator, dialysis, or feeding tube could keep the patient alive longer, it ought to be offered. Otherwise, the right to die could evolve into a duty to die. And some states in the United States have created advance directives in which people can express their preference for the provision of all medical care that would extend life rather than only using advance directives for the refusal of life-­sustaining treatment. Similarly, while some European countries use advance directives to express preferences not to receive care, others use advance directives to express a preference either for or against care.83 “Futility” is generally analyzed under two rubrics: qualitative and quantitative futility. Under qualitative futility, the claim is that medical treatment cannot provide a sufficient benefit to justify its use. Some commentators argue that there is qualitative futility when the patient would not recover enough to go home from the hospital, some find qualitative futility when the patient is permanently unconscious, and others believe there is futility only when treatment cannot provide a physiological benefit. In this third view, treatment is not futile as long as it can prolong life or affect the quality of life in any way. Under quantitative futility, the claim is that there is too low a likelihood that medical treatment will have its desired effect. Here, too, there is disagreement as to when futility exists. Some would find futility when a particular treatment has been consistently unsuccessful for at least 100 tries. Others would place it at different likelihoods of success, and the acceptable likelihood might turn on one’s view about the qualitative benefit

82  For discussion of US futility cases, as well as other aspects of futility, see Hall et al., supra note 32. 83  Ruth Horn, Advance Directives in English and French Law: Different Concepts, Different Values, Different Societies, 22 Health Care Analysis 59 (2014) (published online May 4, 2012); Council of Europe Recommendation CM/Rec(2009)11 on principles concerning continuing powers of attorney and advance directives for incapacity.

Decisions at the End of Life   1103 to be gained. The greater the potential qualitative benefit, the lower the likelihood of benefit before a treatment would be considered futile. There are two basic approaches to questions of futility in the law. Some policies take the view that as long as life can be prolonged, patients should have that option. Other policies take the view that the standard of care for dying patients is a matter for the medical profession to decide and that it is not always appropriate to do everything that might possibly prolong life. In the United States, there has not been a resolution of this issue at the national level. Accordingly, some states take one of the two basic views, some states take the other view, and other states have yet to address the issue. In a Louisiana case involving the discontinuation of artificial ventilation and dialysis in a patient who was comatose, quadriplegic, and in kidney failure, the state court of appeals observed that while physicians have “an obligation to present all medically acceptable treatment options” for consideration by the patient or surrogate decision-­maker, “this does not compel a physician to provide interventions that in his view would be harmful, without effect or ‘medically inappropriate.’ ” The court further observed that whether a treatment would be medically inappropriate is a matter of professional standards to be determined by “a consensus of physicians practicing” in the same medical specialty.84 In contrast, in other cases involving seriously compromised patients, including individuals who were permanently unconscious, courts have not allowed hospitals to withhold artificial ventilation over the family’s objection. A number of states have enacted statutes recognizing authority to withhold or withdraw futile treatments according to professional standards of care. These statutes typically permit physicians to deny “medically inappropriate” or “medically ineffective” care (e.g., Louisiana, Maryland, Texas), with some states expanding on that language to also include the denial of “health care contrary to generally accepted healthcare standards” (e.g., Arkansas, California, Maine, New Mexico, Tennessee).85 Many of the statutes recognizing futility are based on language in the Uniform Health-­Care Decisions Act of 1993, which refers to “medically ineffective health care or health care contrary to generally accepted health-­care standards.” In the comments to the Act, “medically ineffective health care” is defined as “treatment which would not offer the patient any significant benefit.” While the state statutes generally use similar language to define futile care, at least one of the statutes elaborates it in a way that narrows the scope of the futility principle. Idaho defines “medically inappropriate” care as care that will not improve the condition of a terminally ill patient whose death is expected “within hours or at most a few days” or care for which withholding would “not result in or hasten the patient’s death.”86 84  Causey v. St. Francis Medical Center, 719 So. 2d 1072 (La. Ct. App. 1998). 85  Ark. Code § 20-­26-­109; Cal. Prob. Code §4735; La. Rev. Stat. 40:1151(A)(4); Me. Stat. Tit. 18-­C § 5-­808(6); Md. Health Gen § 5-­611(b)(1); N.M. Stat. § 24-­7A-­7(F); Tenn. Code § 68-­11-­1808; Tex. Health & Safety Code Ann. §166.046; Va. Code Ann. §54.1–2990. 86  Idaho Code §39-­4514(6).

1104   David Orentlicher and Judit Sándor Some states also include procedural requirements for healthcare providers to follow before they withhold or withdraw treatment on grounds of futility. The Texas statute is notable for the procedures it sets out. In Texas, a physician’s refusal to honor a request for treatment requires review by an ethics or medical committee, and the patient or surrogate decision-­maker is entitled to attend the committee’s meeting and receive a written explanation for the committee’s decision. If the patient or family disagrees with a committee’s decision affirming the refusal, the physician must make a reasonable effort to transfer the patient to a physician who will provide the desired treatment. Although treatment must be provided pending transfer, there is no obligation to provide treatment beyond the tenth day after a committee decision in favor of refusing treatment. While some view the Texas approach as a model, it has provoked considerable controversy in some cases when doctors and hospitals have invoked it, and family members have been able to delay the withdrawal of treatment through court challenges. Moreover, in 2020, a state court of appeals rejected on constitutional grounds the application of the Texas statute to a decision to withhold life-­sustaining treatment from an 8-­month old infant over the objections of the infant’s mother.87 The ethical and legal concerns have combined such that the statute is used infrequently to discontinue treatment over family objections. While the Texas statute clearly allows for the withholding or withdrawal of treatment, other statutes require the provision of treatment pending transfer to a facility that is willing to provide care.88 But even in states with statutes that seem to clearly authorize the discontinuation of futile care, some hospitals see uncertainty in the statutory au­thor­ i­za­tion and are unwilling to deny care. Other hospitals have implemented futility policies without incident. This kind of uncertainty led Virginia to amend its futility statute, making the authority to withhold futile care clear. It is difficult to be too definitive about futility in the United States. Though the issue comes up regularly, a small number of courts have addressed the issue to date. To the extent that one can identify a trend, courts have been much more comfortable upholding futility decisions that have been implemented by healthcare providers than giving providers approval in advance to withhold life-­sustaining treatment that has been requested.89 In Europe, there is guidance at the supranational level, as a result of the Lambert case in France, but that case also leaves room for variation across the different countries. Under Lambert, countries may, but need not, implement limitations on the provision of

87  T.L. v. Cook Children’s Medical Center, 607 S.W.3d 9 (Tex. Ct. App. 2020) (finding violations of procedural due process and issuing a temporary injunction against the discontinuation of life-­ sustaining treatment). 88  Alaska Code 13.52.060(g)(2); N.Y. Pub. Health L. §§2994-­f(3). Some statutes allow for discontinuation of treatment if a transfer cannot be arranged. Cal. Prob. Code §§4735-­4736; Md. Code Ann., Health–General §5-­613(a). 89  Compare Alexander v. Scripps Memorial Hospital, 232 Cal.Rptr.3d 733 (Ct. App. 2018) (upholding futility decision) with In re Baby K, 16 F.3d 590 (4th Cir. 1994) (rejecting futility argument).

Decisions at the End of Life   1105 life-­sustaining treatment when continued care would be futile. Accordingly, just as the different US states take different views on futility, so do the different European countries. The Lambert case reached the ECtHR, and, as with other end-­of-­life questions, the Court recognized the authority of Member States to define the boundaries of patient rights. The question of futility arose in a challenge to France’s end-­of-­life Léonetti Act.90 The Léonetti Act prohibits so-­called unreasonable obstinacy and authorizes the withholding or withdrawal of treatments when they appear “useless, disproportionate or having no other effect than solely the artificial preservation of life” (acharnement thérapeutique). In the case of Lambert and Others v. France,91 Vincent Lambert suffered a head injury in a traffic accident, and, as a result, he was tetraplegic and in a permanently unconscious condition. The applicants were the parents, a half-­brother, and a sister of Lambert. They complained about the judgment of the French Conseil d’État, which had concluded, based on medical opinion, that it was lawful to withdraw artificial hydration and nutrition (and withdrawal was supported by Lambert’s wife). The applicants submitted that discontinuing artificial nutrition and hydration would be in breach of the State’s obligations under Article 2 of the Convention’s right to life. In addition, depriving him of nutrition and hydration would constitute ill-­treatment amounting to torture within the meaning of Article 3 of the Convention. The Court held (by 12 votes to 5) that there was no violation of Article 2 of the Convention in the event of implementation of the Conseil d’État judgment. In other words, termination of treatment would not constitute a violation of right to life.92 Following this judgment, in July 2019 Lambert died. He had been on life support for more than a decade and his medical condition was assessed many times. This case has become a symbol of the right to “Let Somebody Die” in France. As in France, other European countries permit or even require physicians to withhold or withdraw life-­sustaining care under the doctrine of medical futility. For example, in the Netherlands, it is considered a violation of the professional standard of care for a physician to provide life-­sustaining treatment that is futile, even if the patient or family insists on the treatment.93 Other countries take a very narrow view of futile care. In Italy, for example, the Medical Ethics Code prohibits physicians from delivering care that does not provide a health benefit or improvement in the quality of life. But in the Englaro case, the court did not consider it futile to provide artificial nutrition and hydration to a patient who had been permanently unconscious for 15 years, and other treatments that would prolong life would not be seen as futile. Similarly, in Poland, physicians are obligated to offer treatments that would prolong survival (or improve the patient’s quality of life). Under 90  Law No. 2005–2370 of April 22, 2005 concerning the Patients’ Rights at the End of Life (Léonetti Act). This law is discussed in Hennette-­Vauchez, supra note 18, at 382–383. 91  Lambert, supra note 35. 92  The Court held that the applicants lacked standing to raise the Article 3 claim. 93  Griffiths and Weyers, supra note 26, at 59–60.

1106   David Orentlicher and Judit Sándor Hungarian law,94 even if the seriously ill patient has requested the termination of ­life-­sustaining and life-­saving treatment, a three-­member committee consisting of doctors should assess his or her medical condition. The law refers to the current state of the medical science as a guidance in this assessment.95 In both Europe and the United States, when the law recognizes professional authority to withhold life-­sustaining treatment on grounds of futility, it typically requires a careful process before treatment is withdrawn. Physicians are expected to consult with colleagues, and often with an ethics committee, and allow the family to discuss the matter, seek a second opinion, and arrange transfer of care to a willing provider.

6 Conclusion End-­of-­life decisions have posed many challenges during the past several decades for many parts of the world. The very moment of death has become questionable and a subject of medical-­technological assessment that can be prolonged and sustained. In countries where developed healthcare offers a huge variety of intensive care, dignified death has been repeatedly reinterpreted. The question of treatment at the end of life has both ethical-­legal and medical implications. The end of life is always emotionally disturbing: it is one of the most dramatic events in human life; it is about illness, suffering, and death. No one can escape from these situations, and no one can turn away from the issues of life and death. One day each of us will be dead, and therefore end-­of-­life decisions have an unavoidable impact on those who are left behind. No one can be saved from pondering that one day he or she or their loved ones will have to make such decisions. Even if the decision on aid in dying or the termination of some life-­sustaining treatment seems to be within the individual’s self-­ determination only, it has an effect on the whole society. End-­of-­life cases are, therefore, very emotional, which makes any legal involvement—especially of courts—even more difficult. This emotional involvement is significantly diminished when a legislator sets the norms in advance. Still, as a result of unforeseen technological development or social changes, even a benevolent law may fail. Furthermore, even the most carefully designed law could fail if medical professionals consider it not feasible or contrary to their professional integrity. If the terminally ill patient and their family has no trust in the healthcare system, the law will not work. Perhaps that is why even civil law countries like Italy rely on a case-­by-­case assessment, and only a few countries make an attempt to have a straightforward, explicit statutory law on aid in dying. Of course, there is fear that if aid in dying were made fully legal, then it would become the norm and physicians and fami94  Hungarian Health Care Act, Section §20.() [Act No CLIV of 1997]. 95  Judit Sándor, Ethical and Legal Debates on a Dignified End-­of-­Life and the Role of the Family in Hungary 211, in Feuillet, et al., supra note 17.

Decisions at the End of Life   1107 lies would give up their struggles too easily. And perhaps patients would feel that they are a burden on the healthcare system and that they should find an easy and fast exit once they encounter serious and incurable conditions. While dignity plays a crucial role in European legal philosophy and law, liberty interests also serve as useful tools in promoting the rights of the terminally ill. In Europe, the legalistic, regulatory approach is more dominant in shaping the laws related to end-­of-­ life decisions, while in the United States medical interpretations of the end of life seem to prevail, especially in the lack of advance directives. Plurality of ethical norms, however, contributes to legal diversity, where legal and ethical standards are often borrowed from other jurisdictions.

Index

For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. abortion  28, 1035–42, 1053–62 fetal genetics  932–3 abusive conduct, see ‘antitrust law’ access to healthcare  19–22, 27 ‘abandonment’  239–41, 270–1 abortion and contraception  1057–62 affordability  27–8, 56–7, 209–10, 245–6 assisted reproduction  1006–7 coverage  248–64, 269–71, 317 duty to treat  228–38, 252–61 emergency care  231 inequity  225–6, 235–8, 241, 263–5, 860–5, 872, 989–90, 994, 1000, 1006, 1010 price controls  210–11 rationing  211, 249–50, 256–9, 266–9, 860–5 uninsured patients  216–17 accreditation, see ‘health professionals’ adult patients, see ‘consent’ advance directives, see ‘aid in dying’, ‘consent’, ‘medical futility’, ‘withdrawal of life-sustaining treatment’ advanced therapy medicinal products, see ‘medicines regulation’ affordable care  27–8, 56–7, 209–10 aid in dying  5–6, 11, 1092–101 active versus passive  1092–3 advance directives  1098 constitutionality of  1093, 1095–6 human rights and  1093–5 legalization of  1096–101 anti-bribery law, see ‘fraud’ anti–fraud activities, see ‘fraud’ antitrust law  9–10 abusive conduct  320–2 exclusion/foreclosure 322–5

horizontal collaboration  325–7, 336–7 concentration, see also ‘mergers’  327–30, 357–61 affordability of healthcare and  311–12 healthcare service markets  313–14, 350–1, 357–61 market definition  328, 357–61 market power  312–13, 315, 318–19, 361–3 mergers  313–14, 320–1, 357–61 scope  331–6, 349–51 state providers and  318–19, 331–6 ARTs, see assisted conception/reproductive technologies assisted conception/reproductive technologies  990–2, 994 children born from  993–4 cross-border care  1002–6, 1018–19 law reform, and  1004–5 donors anonymity  997, 1002 compensation  995, 1001–2 parentage 1017 egg freezing  1001 egg sharing  1006–7, 1023 embryo disputes  1018 equality in access  994, 1000, 1002–3, 1006–7, 1010 EU Tissues and Cells Directive  995 funding  992, 997–8, 1002, 1006–7 age limits  1006 cycle limits  1006 infertility-based 1006 insurance 1023–4 Human Fertilisation and Embryology Act (HFEA) 1001–2

1110   index assisted conception/reproductive technologies (continued ) in vitro fertilization (IVF)  990, 996–7 internet-assisted 1005 posthumous reproduction  999–1000, 1018 preimplantation genetic diagnosis (PGD)  932–3, 996–7, 1001 prisoners  990, 999 regulation of  994–8, 1001–2, 1016–22 professional 1019–21 tort liability  1021 right to procreate  989–92, 998–1001, 1011–16 constitutional basis  1011–16 litigation of  998–1000 negative versus positive rights  989–90, 992, 998, 1011 single embryo transfer (eSET)  997 storage time limits  1001 surrogate motherhood  1000, 1002, 1011–13, 1017 constitutionality 1014–15 assisted suicide, see ‘aid in dying’ autonomy  4, 7, 10–11, 13–14, 25, 461, 1028 best interests  459–74, 1082–9 bias, see ‘conflicts of interest’ biobanks  10, 657 access for researchers or others  678–80 consent 660–8 broad consent  663–4, 666–7 Common Rule  661 GDPR 666 national regulation in Europe  665 state laws in US  664 waiver of  663, 666 engagement with public and other stakeholders 675–6 European examples  677–8 US practices  676–7 examples of  657–8 governance CoE recommendations  674 OECD guidelines  673 UK biobank  675 US approaches  673–4 identifiable specimens  662 privacy protections  668–73

Common Rule  668 European Conventions  672 GDPR 670–2 HIPAA Privacy Rule  669 national regulation in Europe  672 state laws in US  670 regulatory variations  659–60 blood 893–5 donation 83 payments for  893–5, 903–4, 908 bodily materials (non-organ)  7, 809–14 human rights model  898–902 incomplete commodification of  884–95, 914–15 intellectual property rights  829–36 no property rule  814–15, 819–25 payments for  875–81, 902–9 arguments against  895 blood  893–5, 903–4, 908 eggs  889–93, 908–15 legal rules  881–4 sperm  887–9, 908–15 tissues  904–5, 908 property rights  818–19 quasi-property rights  825–9 third-party rights in detached body parts 829–33 capacity, see ‘consent’ capitation, see ‘fraud’ cell products, see ‘bodily materials’ regulation  718–19, 748–51 charitable organizations  49–52 children  6, 10, 459–74 abortion and contraception  473, 1061–2 best interests  459–63, 465–71 capacity 459–74 mature minors  463, 467–70, 1089 religion  462–3, 468, 470 children’s rights  10–11, 461–5, 1087–92 circumvention tourism, see ‘cross-border health care’ clinical trials regulation  694, 728–9 communicable diseases  72–3, 106–18 isolation and quarantine  92–3, 111–14 preparedness  86–7, 114–16

index   1111 risk communication  89, 108–9 surveillance  87–9, 107–8 travel restriction  93–4, 111–14 vaccination  89–92, 102, 109–11 compensation for defective products  91 for injuries to participants in research  611, 620, 623, 649 competition 32–4 conflicts of interest  5, 505, 537 bias  506, 508, 537–8 definition  506–7, 514–16, 541–2 disclosure  509, 511, 523–5, 527–8, 530–1, 547–50 drawbacks of disclosure  509–10, 527–9, 532–5 ethical dimensions  508, 539 fiduciary duties and  518–20, 544–5 financial conflicts  507, 515–16 non-financial conflicts  507, 515–17 perception 506 right to health and  542–4 risks from  507–8, 510, 516–17, 520 self-regulation and  525–6 consent  4, 459, 477–8 adults 477–503 duty to disclose  482, 484–6 exceptions to autonomy  483 right to decide  482, 484–6, 488 children 459–74 vulnerable adults  477–503 advance directives  493–6 capacity  479, 488–91 family members  498–500 Oviedo Convention  487 substituted judgment  479–80 surrogate decision-maker  499–502 UNCRPD  480–1, 486–7, 490–1 contraception  469–70, 1041, 1052–3 corpses as property  826–9 corruption, see ‘fraud’ coverage, see ‘access to healthcare’ COVID-19  6, 8–9, 13–14, 22–3, 25, 28–32, 36–7, 98 cross-border health care  6, 27, 259–61, 992, 1002–6, 1016–19, 1024–5

data privacy, see ‘medical data/information’ dietary supplements  703–4 discipline, see ‘health professionals’ disclosure, see ‘conflicts of interest’ e-health  776–7, 802–3 assisted conception/reproductive technologies 1005 embryo disputes  999–1000, 1018 end-of-life decisions  1073–107 end-of-life medical care  1073–107 ethical oversight, see ‘ethical values’ under ‘research with human participants’ EU agencies  78 European Centre for Disease Prevention and Control  85, 88 European human rights model  10 euthanasia  5–6, 11, 1092–101 fertility fertility tourism  909–11, 1002–6 fertility treatment, see ‘assisted conception/reproductive technologies’ fetal health  1031, 1035–41, 1043–8, 1056, 1064–7 financing health systems  11, 19–20 see also ‘access to treatment’ firearms 136–9 fraud  8, 11, 365 abuse  367–9, 377–8, 400–1 anti-bribery laws  384–5, 390–2 anti-fraud activities  8, 11, 367–72, 375–7, 403–6 capitation  376, 395, 407–8 corruption  368, 375–7, 381–2, 401–8 gifts and informal payments  387–9, 408–12 penalties  369–70, 378–81, 383–92, 395–400, 402–3 seeding trials  386–7 waste 378 gene therapy  942–3, 974, 982–3 regulation  717–18, 748–51 genetic counselling  920, 953–7

1112   index genetic discrimination  936–42, 967–71, 976, 982 Genetic Information and Nondiscrimination Act  936–9 genetic privacy  926–7, 936–42, 971–4 genetic research  964–7 genetic testing  919–21 adults  924–7, 953–7 children  927–30, 957–9, 981 confidentiality 926–7 direct-to-consumer  773–5, 800–1, 935–6, 963–4 informed consent  924–7, 953–7, 981 liability 926–7 newborn screening  927–30, 959–60, 981–2 prenatal testing  930–3, 960–3, 982 regulation  935–6, 952–3, 982 genetics precision medicine  944–5, 974–6 germline modifications  943–4 health disparities, see ‘access to healthcare’ and ‘public health law’ health facilities, see ‘healthcare institutions/ providers’ health inequities, see ‘access to healthcare’ and ‘public health law’ health justice  37–9 health professionals  10, 23–4 accreditation and licensing  276, 278, 281, 294, 296, 299–300, 355–6 board certification  302 complaints 287 discipline  286, 296–7 misconduct  276, 281, 284–8 mobility  276, 284–6 patient safety  280, 282, 294, 298 provider conscience  277–8, 1042–3, 1060–1 provider culpability  275 reimbursement 301–2 ‘right touch regulation’  279–80, 301–3 self-regulating health professions  23, 45–7, 277–8, 282–3, 296–9, 352–5 transparency 302–3 health research  10 health services, see ‘access to healthcare’ health system regulation  10, 19, 25–6

‘right touch regulation’  279–80 healthcare fraud, see ‘fraud’ healthcare institutions/providers  23, 39–52 licensure  305, 307–8 independence  289–91, 304 public funding and  289, 305–6 ‘right touch regulation’  279–80, 304 sanctions  290–1, 305–6 transparency  303–4, 309 healthcare service markets, see ‘antitrust law’ human body parts  7, 809, 897 human organs  10 see also ‘organ allocation’, ‘organ donation’, ‘organ transplantation’ human research participants, see ‘research with human participants’ human rights  10, 81–2 human tissues, see ‘tissue products’ individual liberty  25–30 information disclosure, see ‘conflicts of interest’ informed consent conflicts of interest  520–2, 546–7 genetic testing  920 withdrawal of treatment  1077–80 institutional review boards (IRBs), see ‘research with human participants’ insurance  5–6, 19–20, 23, 27–8 private insurance  217 social insurance  217–18, 244–6 blurred public/private insurance  221 voluntary health insurance  244 international human rights, see ‘human rights’ liability  559–97 (see also ‘genetic testing’, ‘medical malpractice’, ‘medical devices’, ‘medicines regulation’, ‘tobacco’) compensation 559–97 negligence  565–70, 583–4 breach  565–6, 577, 582 burden of proof  565–6 causation  565, 568, 577, 589 corporate negligence  571

index   1113 damages  568–9, 577, 582–3, 591 duty 565–6 vicarious liability  570 ‘no fault’ compensation  576–7, 583–4 patient injury  559–97 reform 574–6 Scandinavian ‘no fault’ compensation system  560–1, 585–90 tort doctrine  559–60 market power, see ‘antitrust law’ market solutions, assisted reproductive technology  1022–5 marketing authorization, see ‘medicines regulation’ medical data/information General Data Protection Regulation (GDPR)  88, 611, 627–31 Health Insurance Portability and Accountability Act (HIPAA)  644–5 medical devices  10 advertising  772–3, 799–800 off-label  772–3, 799 definition of devices  756–8, 781–3 Classes I–III  757–8, 782–3 combination products  757, 782 incentives to innovate data exclusivity  771 marketing exclusivity  770–1, 798 liability 691–2 premarket requirements  758–63 501(k) notification  760–2 CE marking  783–4 Class I devices  762 Class II devices  761–2 Class III devices  758–60, 762 clinical investigation  759, 784–5 custom devices  765, 787 de novo classification  762–3 direct-to-consumer genetic testing  774–5, 785–6, 800–1 expanded access  763 humanitarian use devices  764, 787 in vitro diagnostic tests  785–6 notified bodies  790–1 public oversight, Europe  783 smart consumer products  776–7, 802–3

three-dimensional printing  775–6, 801–2 risk management  766–7 adverse event monitoring  768, 795–6 cybersecurity  769, 796–7 labeling  766, 768 limits on access  766–7, 793 manufacture and distribution  784, 788 removal from market  768–70, 795–6 special controls  766–7 traceability 793–5 safety and effectiveness  759 smart consumer products  776–7, 802–3 three-dimensional printing  775–6, 801–2 medical futility  1101–6 advance directives  1102 medical malpractice conflict of interest and  522–3 liability 559–97 medicines regulation  9–10 advanced therapy medicinal products  717–18, 725–6, 735, 748–51 advertising  712, 747–8 direct-to-consumer  712–13, 747 off-label  712–13, 748 definition of medicines  692–3, 722–4 evergreening 714 gene therapy  717–18, 748–51 generic substitution  713 incentives to innovate  688–9, 709–17, 741–5 data exclusivity  711 marketing exclusivity  709–12, 741–3 new condition of use  711, 715 orphan medicines  711–12, 745–6 pediatric medicines  746–7 public-private cooperation  744 small to medium-sized enterprises  744 liability 91 parallel trade  743–4 premarket requirements  700–4 abbreviated application  700–1 biosimilar biologics  701, 735 generic medicines  700–1, 734–5 hybrid medicines  735 excluded medicines  702–4 customized compounding  702–3, 735–6

1114   index medicines regulation (continued) dietary supplements  703–4 generally recognized as safe and effective 702 homeopathic medicines  703 expanded access (compassionate use)  698–700, 733 expedited approval  696–700, 729–34 standard approval process  694–6, 724–9 public health crises  686 risk management  704, 736 adverse event monitoring  707–9, 738 labeling  705, 707–8, 736–7 limits on access  704–6, 737–9 manufacture and distribution  739 pharmacovigilance  707–9, 738 postmarketing study  706 risk evaluation and mitigation strategy 705 risk management plans  736–7 safety and effectiveness  686–7, 693–704 microbiome-based products regulation  719–20, 724, 744–5 mitochondrial replacement therapy  1016 negligence-based culpability  11, 238–9 noncommunicable diseases  73–5 commercial free speech  139–41 ‘health in all policies’  145–7 paternalism  120–3, 196 state regulation  123–7, 145–7 opioids 136–9 organ allocation  860–5 organ donation  11 deceased ‘donors’  854–60 actual vs. presumed consent  854–9 advance directives  858 mandated choice  860 routine request  859 financial incentives  867–72 living donors  847–50 donor capacity  850 informed consent  851–2 protection of  852–4 organ shortage  841–2 organ transplantation  10

barriers to access  864–5, 872 confidentiality 865–7 legal regulation  845–7 racial disparities in  865 trafficking 871–2 palliative care  1073, 1076, 1087, 1091–2, 1101 patient compensation funds, see ‘liability’ patient injury, see ‘liability’ patient safety, see ‘health professionals’, ‘healthcare institutions/providers’ and ‘liability’. pharmaceuticals, see ‘medicines regulation’ physicians, see ‘health professionals’ posthumous reproduction  999, 1018 precision medicine  944–5 pregnancy prenatal genetic testing  930–3 regulation of  1047–8, 1065–7 privacy  9–10, 417–57 access to medical records  419, 452–3 anti-discrimination 449–51 data processing  434 data security  418–19, 437–40, 451–2 HIPAA security rule  438 GDPR security rules  438–40 de-identified data  423, 438–40 HIPAA de-identification  440–2 GDPR anonymisation  442 enforcement 442–9 genetic data  429 predictive health information  455–7 protected health information  419, 421 exceptions 422–3 research data  453–5 right to data protection  425 Council of Europe  426 European Union  427 material scope  428 personal data concerning health  428, 430 personal scope  431 territorial scope  430 surveillance and reporting of communicable diseases  87–9, 107–8 provider conscience  1042–3, 1060–1

index   1115 provider culpability, see ‘health professionals’, ‘liability’ public and private actors  42–5, 47–9 public health law, see also ‘communicable diseases’, ‘noncommunicable diseases’ definition 70–2 health inequities  75–6, 102, 104, 171–203 isolation and quarantine  92–3, 111–14 paternalism 120–3 preparedness  86–7, 114–16 risk communication  89, 108–9 surveillance  87–9, 107–8 travel restriction  93–4, 111–14 vaccination  89–92, 102, 109–11 reproductive freedom  1027–8, 1051 reproductive genetic testing  996–7, 1001 reproductive justice  1057 reproductive technology, see ‘assisted conception/reproductive technologies’ research with human participants  607–54 Common Rule  608–10, 634–6 compensation for injury  611, 620, 623, 649 conflicts of interest  651–4 data protection  611, 614–15, 627–31, 643–5 Declaration of Helsinki  615–16 ethical values  615–25 informed consent  615, 620, 623–4, 639–42 waiver of consent  641–2 protection from risks  620–1, 623–4, 638–9 ethics and scientific review ethics review  610–11, 620–4, 638–9 institutional review boards (IRBs)  608–10, 633–4, 636–8 scientific review  610–11, 622–3 EU Clinical Trials Directive  609, 619 EU Clinical Trials Regulation  609, 622 harmonisation of regulation  613–21, 631–2 inclusion of women and minorities  647–9 non-clinical research  625 research ethics committees (RECs) 625–6 Nuremberg Code  608, 615

recruitment of participants  642–3 registration of trials  619, 624, 649–51 reporting of data  632, 649–51 social science research  625–7 special populations  645–7 children 645–6 pregnant women and fetuses  645–6 prisoners 645–6 use of previously gathered data  630–1, 640–1 right not to procreate  4, 11, 28, 1034–70 right to health  7, 21–2, 25–30, 163–6, 175–6, 210–11, 227–8, 247–8, 257–8 right to procreate  11, 989, 998–1001, 1011–16 negative versus positive rights  989–92 regulation of  994–8, 1001–2, 1019–22 Scandinavian compensation system, see ‘liability’ scope of practice, see ‘health professionals’ self-regulating health professions, see ‘health professionals’ sex equality research participants  647–9 right not to procreate  1029, 1056–7 social determinants of health  7, 24, 75–6, 171–203 Council of Europe  176–9 health inequalities  173–4, 189–93 sperm donation  995, 1002–6, 1017 sterilization  1041–2, 1060–2 surrogate motherhood  1000, 1002, 1017 three-dimensional printing  775–6, 801–2 tissue products regulation  718–19, 748–51, 882–3, 902–9 tobacco  5–6, 10–11, 127–33, 148–52 commerce clause regulation  124 e-cigarettes  131–3, 148–9 European Union competence  146–7, 149–52 Federal Food and Drug Administration (FDA)  129, 131–2 Framework Convention on Tobacco Control  128, 149–50 liability of tobacco manufacturers  122–3 Tobacco Control Act  130, 132 tort doctrine, see ‘liability’

1116   index undue influence, see ‘fraud’ unhealthy foods  133–6, 152–61 calorie labelling  122, 136, 154–8 commerce clause regulation  124 Federal Food and Drug Administration (FDA) 135 nutrition and health claims  154–8, 164 paternalism 133–4 state regulation  134–5, 158–61

vulnerable adults, see ‘consent’ withdrawal of life-sustaining treatment competent patients  1077–80 incapacitated patients  1080–92 adults 1081–7 advance directives  493–6, 1079, 1081–2, 1084–7 minors 1087–92