Biomedical ethics : a Canadian focus [Third edition.] 9780199022281, 0199022283

Citation preview

OXFORD

THIRD EDITION

BIOMEDICA EllICS A CANADIAN FOCUS

Roma

111

EDITED BY

FISHER RUSSELL BROWNE r BURKHOLDER

1.I

CCM ©40

THIRD EDITION

5\A I L_

A CANADIAN FOCUS

EDITED BY

JOHNNA FISHER I J.S. RUSSELL ALISTER BROWNE I LESLIE BURKHOLDER

OXFORD UNIVERSITY PRESS

11 III I

I I ill

W

OXFORD UNIVERSITY PRESS

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 in certain other countries. Published in Canada by Oxford University Press 8 Sampson Mews, Suite 204, Don Mills, Ontario M3C OHS Canada wwwoupcanada.com Copyright 0 Oxford University Press Canada 2018 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First Edition published in 2009 Second Edition published in 2013 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 licence, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Permissions Department at the address above or through the following url: www.oupcanada.com/permission/permission_request.php Every effort has been made to determine and contact copyright holders. In the case of any omissions, the publisher will be pleased to make suitable acknowledgement in future editions.

Library and Archives Canada Cataloguing in Publication Biomedical ethics : a Canadian focus / edited by Johnna Fisher, J.S. Russell, Alister Browne and Leslie Burkholder. — Third edition. Includes bibliographical references and index. Issued in print and electronic formats. ISBN 978-0-19-902228-1 (softcover).—ISBN 978-0-19-902229-8 (PDF) 1. Medical ethics Canada. 2. Bioethics—Canada. 3. Medical ethics—Canada—C se studies. I. Fisher, Johnna, 1964-, editor R724.8565 2018

174.20971 C2017-906107-0

C2017-906106-2

Cover image: Assembly/Iconica/Getty Images Cover and interior design: Sherill Chapman Oxford University Press is committed to our environment. Wherever possible, our books are printed on paper which comes from responsible sources. Printed and bound in the United States of America 1 2 3 4-21 20 19 18

1111111

Contents Preface xiii

1 Morality and Moral Decision-Making: A Brief Introduction 1 1.1 The Indispensability of Biomedical Ethics 1 1.2 The Case of Hassan Rasouli 2 1.3 Taking Morality Seriously 3 1.4 Normative Ethical Theories 7 1.5 Moral Reasoning in Biomedical Ethics 17 1.6 Professional Codes of Ethics and Law 19 1.7 Looking Ahead 20 1.8 Study Questions 21 1.9 Suggested Readings and Resources 22

2 Medical Decision-Making: Self-Determination and Deciding for Others 23 2.1 Introduction 23 2.2 Determining Decision-Making Capacity 32 Standards of Competence 32 Allen E. Buchanan and Dan W. Brock

A Relational Approach to Autonomy in Health Care 35 Susan Sherwin

2.3 Informed Consent 45 The Nuts and Bolts of Obtaining Consent to Treatment 45 L.E. and F.A. Rozovsky

The Concept of Informed Consent 47 Ruth R. Faden and Tom L Beauchamp

Transparency: Informed Consent in Primary Care 51 Howard Brody

Culture, Power, and Informed Consent:The Impact of Aboriginal Health Interpreters on Decision-Making 56 Joseph Kaufert and John O'Neil

■o i

®

LI

I I JII

III Wil

Contents

Informed Consent and Public Health 60 Onora O'Neill

2.4 Substituted Judgments 63 Enough:The Failure of the Living Will 63 Angela Fagerlin and Carl E. Schneider

Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures 76 Canadian Medical Association

2.5 Best-Interest Judgments 76 Involving Children in Medical Decisions 76 Christine Harrison, Nuala P Kenny, Mona Sidarous, and Mary Rowell

Position Statement: Treatment Decisions Regarding Infants, Children, and Adolescents 80 Christine Harrison; Canadian Paediatric Society, Bioethics Committee

Withholding or Withdrawing Life-Sustaining Treatment in Children 81 Royal College of Paediatrics and Child Health

Deciding to Forego Life-Sustaining Treatment 82 Judicial Council, American Medical Association

Ethical Relativism in a Multicultural Society 83 Ruth Macklin

2.6 Cases 91 Case 1 Scott Starson: Refusing Treatment while Incompetent 91 Case 2 No Chemotherapy for Anael: Surrogate Refusal ofTreatment for a Minor Child 92 Case 3 Do Everything for Mom: Advance Directives and a Surrogate's Right to Demand Treatment 93 Case 4 Treatment for Jehovah's Witnesses: Adults and Children 94 Case 5 Markayla Sault: Caring for an Aboriginal Patient 95 2.7 Study Questions 95 2.8 Suggested Further Reading 96

3

Management of Medical Information 98 3.1 Introduction 98 3.2 Patient Access to Information 106 On the Supposed Right to Lie from Benevolent Motives 106 Immanuel Kant

Telling the Truth to Patients: A Clinical Ethics Exploration 107 David C. Thomasma

Telling Patients the Truth 110 Robert Scott Stewart

Medical Secrecy: Patients' Right of Access to Medical Records 113 Arthur Schafer

3.3 Confidentiality of Information 117 A Defense of Unqualified Medical Confidentiality 117 Kenneth Kipnis

Contents

Breaching Confidentiality 128 Dave Unger How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales 134 Bonnie Kaplan 3.4 Genetic Information 147 Genetic Exceptionalism and "Future Diaries": Is Genetic Information Different from Other Medical Information? 147 Thomas H. Murray 3.5 Cases 153 Case 1 George VI's Cancer 153 Case 2 Arndt v Smith 154 Case 3 The Crash of Germanwings Flight 9525 155 Case 4 Personal Genome Mapping 156 3.6 Study Questions 157 3.7 Suggested Further Reading 158

4 Professional Ethics 159 4.1 Introduction 159 4.2 The Nature and Limits of Professional Autonomy and Professional Responsibility 165 Patient and Physician Autonomy. Conflicting Rights and Obligations in the Physician-Patient Relationship 165 Edmund D. Pellegrino The Problem with Futility 176 Robert D. Truog, Allan S. Brett, and Joel Frader Should Physicians Be Gatekeepers of Medical Resources? 181 Milton C. Weinstein Caring in a Crisis:The Ethical Obligations of Physicians and Society during a Pandemic 188 Canadian Medical Association Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies 191 Udo Schuklenk and Ricardo Smalling The Functions and Limitations of Professional Codes of Ethics 201 Dale Beyerstein 4.3 Cases 205 Case 1 Hassan Rasouli and Medical Futility 205 Case 2 MRS in Toronto: A Duty to Care 206 Case 3 Is It Ever Right to Violate Rules of Rationing? 207 Case 4 BC Physicians "Cherry-Picking" Patients for Ease of Care, Refusing Those in Need 207 Case 5 Helping Residents to Live at Risk 208 Case 6 When Is Treatment Futile? 209 4.4 Study Questions 209 4.5 Suggested Further Reading 210

11

II

I

ill

111111

Contents

5

Reproductive Ethics 211 5.1 Introduction 211 5.2 Starting Reproduction 218 Reproductive Freedom, Autonomy, and Reproductive Rights 218 Christine Overall

Preconception Gender Selection 226 John A. Robertson

Preconception Arrangements 233 Royal Commission on New Reproductive Technologies

5.3 Stopping Reproduction 238 Why Abortion Is Immoral 238 Don Marquis

On the Moral and Legal Status of Abortion 248 Maly Anne Warren

A Defence of Abortion 253 Judith Ja►vis Thomson

A Third Way 262 L.W. Sumner

Abortion through a Feminist Ethics Lens 268 Susan Sherwin

5.4 Completing Reproduction 277 Judicial Intervention in Pregnancy and Birth 277 Royal Commission on New Reproductive Technologies

Dissent on Judicial Intervention in Pregnancy and Birth (Royal Commission on New Reproductive Technologies) 281 Suzanne Rozell Scorsone

5.5 Cases 282 Case 1 Sex Selection in Canada 282 Case 2 Prenatal Diagnosis and Abortion or Infanticide through Declining Treatment 282 Case 3 Abortion of Suspected Female Fetus 283 Case 4 Ms G and Refusal of Treatment while Pregnant 283 Case 5 Endangering Behaviour in a Pregnant Woman 284 Case 6 Abortion of a Fetus Due to Diagnosis of Down Syndrome 284 5.6 Study Questions 285 5.7 Suggested Further Reading 286

6 End-of-Life Decision-Making 287 6.1 Introduction 287 6.2 Withholding or Withdrawing Life-Sustaining Treatment and Physician-Assisted Death 295 Withholding and Withdrawal of Potentially Life-Sustaining Treatment 295 Health Law Institute, Dalhousie University

Voluntary Active Euthanasia 297 Dan W. Brock

Contents ® Active and Passive Euthanasia 309 James Rachels

When Self-Determination Runs Amok 312 Daniel Callahan

Medical Ethics and Double Effect:The Case of Terminal Sedation 316 Joseph M. Boyle

6.3 Physician-Assisted Dying in Canada 321 Controlling the Risks of PAD 321 British Columbia Supreme Court

Legislative Background: Medical Assistance in Dying (Bill C-14) 323 Department ofJustice

Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection 328 Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying and the Special Joint Committee on Physician-Assisted Dying

6.4 The Concept of Death and Its Practical Implications 338 Defining Death 338 Alister Browne

6.5 Cases 341 Case 1 Nancy B.: Withdrawing Life-Sustaining Treatment 341 Case 2 Dr Nancy Morrison: Nonvoluntary Active Euthanasia of an Adult 341 Case 3 Tracy Latimer: Nonvoluntary Active Euthanasia of a Minor 342 Case 4 Mr McCullough: Recommending Voluntary Passive Euthanasia 344 Case 5 Elizabeth and Eric MacDonald: Assisted Suicide 344 6.6 Study Questions 345 6.7 Suggested Further Reading 345

7

Delivery of Health Care and Resource Allocation 349 7.1 Introduction 349 7.2 Access to Health Care in Canada 357 Sustaining Medicare:The Commission on the Future of Health Care in Canada 357 Roy Romanow

7.3 The Right to Health Care, Macroallocation, and Setting Priorities 367 The Right to a Decent Minimum of Health Care 367 Allen E Buchanan

Justice and the High Cost of Health 372 Ronald Dworkin

Why UndervaluingnStatisticar People Costs Lives 377 Tony Hope

QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference Do They Make for Health Care Priority Setting? 382 Bjarne Robberstad

O

Contents

7.4 Microallocation: Decisions at the Bedside 391 The Allocation of Exotic Medical Life-Saving Therapy 391 Nicholas P. Rescher

Just Caring: In Defense of Limited Age-Based Healthcare Rationing 399 Leonard M. Fleck

7.5 Increasing Resources 407 Human Organs, Scarcities, and Sale: Morality Revisited 407 R.R. Kishore

7.6 Cases 412 Case 1 Optimal Care versus Cost Containment: What Is a Doctor to Do? 412 Case 2 Having an Estate Sale of One's Organs 413 Case 3 Rationing Services to an Elder Who Is Responsible for His Medical Condition 413 Case 4 Buying a Kidney in India but Requesting Canadian After-Care 413 Case 5 Does Clifford Olson Deserve Medical Treatment? 414 7.7 Study Questions 415 7.8 Suggested Further Reading 415

8 Public Health 418 8.1 Introduction 418 8.2 Collective Action Problems 427 The Measles and Free Riders: California's Mandatory Vaccination Law 427 Katharine Browne

Free Riding and Organ Donation 431 Walter Glannon

8.3 Screening for Disease 434 Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario? 434 Timothy Caulfield

The Ethics of Screening: Is "Screeningitis" an Incurable Disease? 436 Darren Shickle and Ruth Chadwick

8.4 Harm Reduction Programs 443 Canada's Highest Court Unchains Injection Drug Users; Implications for Harm Reduction as Standard of Healthcare 443 Dan Small

Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate 451 Timothy Christie, Louis Groarke, and William Sweet

8.5 Inequality in Health 458 When Are Health Inequalities Unjust?—The Social Determinants of Health 458 Norman Daniels

8.6 Cases 471 Case 1 Health Care Workers and Flu Shots 471 Case 2 Students and Study-Enhancing Drugs 472

Contents

Case 3 Breast Cancer Screening 473 Case 4 Water in First Nations Communities 474 8.7 Study Questions 475 8.8 Suggested Further Reading 475

9 Research with Humans 476 9.1 Introduction 476 9.2 The Dark History of Human Research in Canada 485 Dr Ewen Cameron 485 Colin A. Ross

Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis— Learning from the Cases of Nancy Olivieri and David Healy 489 Arthur Schafer

9.3 Ethical Issues in the Design and Conduct of Clinical Research 495 An Ethical Framework for Biomedical Research 495 Ezekiel J. Emanuel, David Wendler, and Christine Grady

A Critique of Clinical Equipoise:Therapeutic Misconception in the Ethics of ClinicalTrials 509 Franklin G. Miller and Howard Brody

Protecting Communities in Biomedical Research 518 Charles Weyer and Ezekiel J. Emanuel

Repairing Research Integrity 523 Sandra L Titus, James A. Wells, and Lawrence J. Rhoades

9.4 Use of Vulnerable Populations in Research 526 Vulnerability in Research and Health Care; Describing the Elephant in the Room? 526 Sarnia A. Hurst

Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies 537 D. Avard, L Black, J. Samuel, G. Griener, and BM. Knoppers

9.5 Cases 547 Case 1 Dr Roger Poisson: Fraud in Breast Cancer Study 547 Case 2 Mr Halushka: Human Research and Harm to Participants 548 Case 3 Luka Magnotta and the Legal Recognition of Research Confidentiality 549 Case 4 Jesse Gelsinger: Research Conflicts and Ethical Review 549 9.6 Study Questions 550 9.7 Suggested Further Reading 550

Glossary 553 Index 557

G

Pr ice We are honoured to continue the work of our late colleague, Johnna Fisher, by editing the third edition of her Biomedical Ethics: A Canadian Focus. Johnna Fisher was an outstanding teacher and a respected bioethicist who, tragically, was just coming into her own as a young academic when she died. Her editorship of the first two editions of this book established a benchmark for biomedical ethics texts in Canada. The current edition is substantially revised over previous editions, and this requires some comment. We amended and extended Biomedical Ethics in response to helpful and often copious advice from peer reviewers and also from Oxford University Press's Leah-Ann Lymer. Each of the new editors has also brought his perspective to the book. The years since the second edition was published have produced many important developments in biomedical ethics in Canada that required careful attention. As a consequence, each chapter introduction has been written anew, two chapters have been added, and 60 per cent of the readings are new to this edition. But we have retained many of Johnna's outstanding case studies (not surpassed anywhere, in our view) and the classic readings she had selected. We thank all of the individuals who contributed to the revision of this book, and we hope we have extended Johnna's work in fruitful ways. J.S. Russell, Alister Browne, and Leslie Burkholder May 2017

Reviewers We wish to thank the reviewers, both named and anonymous, whose thoughtful comments helped to shape this edition. Ken Kirkwood, Western University Chris Lowry, University of Waterloo Jeff McLaughlin, Thompson Rivers University Andrea Nicki, Simon Fraser University Michael Poellet, University of Saskatchewan Milly Ryan-Harshman, University of Ontario Institute of Technology Andrew Sneddon, University of Ottawa Curtis Sommerlatte, Concordia University

MoraIlty and MoraII Dedision-Makring A Brief Introduction 1.1 The Indispensability of Biomedical Ethics

-1 :

Health care poses diverse, complex moral challenges that touch nearly everyone at some time in their lives. These challenges tend also to be at the very forefront of public debate and controversy. Euthanasia, abortion, organ donation, human experimentation, genetic therapies, access to health care, the role of children in medical decision-making, sex selection of fetuses, and treatment of those responsible for their illness are just a few examples of issues that are encountered in the lives of patients and health care providers and that require the most thoughtful public engagement. They test our moral understanding at a deep level. They also pose difficult conceptual, metaphysical, psychological, economic, political, and sociological problems. (What is death? What is futility? What is the capacity to consent? What can a society afford to spend on health care? Will sex selection lead to social unrest?) There is no escape from grappling with these sorts of questions. Indeed, the practical moral problems that we confront in health care frequently cannot be separated from ideas, issues, and theories covering a range of disciplines. And then there are the specifically medical questions. (What is the patient's diagnosis? Will the treatment work? What are the risks?) All these questions place obvious, often onerous, demands on patients, their families, health care providers, and public policy-makers.(lt explains why a discipline of biomedical ethics has emerged and is indispensable to the proper conduct of health care. This text attempts to provide a systematic introduction to biomedical ethics that will educate students about the range of moral issues and problems that health care faces today. Its focus is primarily Canadian. It takes a Canadian look at the field by including, whenever possible, standards, cases, and articles that address biomedical moral issues and problems in a Canadian context. As such, it will be of particular interest to anyone wanting to come to grips with moral issues and problems that arise in health care in Canada. While the articles and cases tend to focus primarily on health care provider-patient relations, they also address conceptual issues in health care that will be of interest to anyone with a deeper philosophical interest in biomedical ethics and the public policy questions it raises. The current chapter provides an introduction to moral theo6,) and 'oral reasorl.ng that will help to ground an understanding of issues and debates-in biomedicarelMait is a vital start to a proper appreciation of biomedical ethics and becoming an informed observer and participant in related discussions and debates. Those who are new to studying morality are sometimes perplexed and intimidated by the abstractness of some of these ideas and

,=-

un

II

I

:,111

111111

1 Morality and Moral Decision-Making

by the diversity of apparently conflicting views. Some of this is understandable. There are important disagreements among many of the moral theories that we will examine. These disagreements are often reflected in the debates that are addressed in later chapters. That is one good reason to have a sense of them, since they deepen understanding of those debates and what is at stake in them. But another reason to begin with a treatment of moral theories is to see the extent to which they overlap and can even be complementary. In this sense, the diversity of theoretical views in morality is no different from that in other spheres of inquiry where such disagreement reflects the difficulty of accurately describing and understanding complex phenomena. Competing theories often purport to capture content that is not covered or as well explained by other theories. In this sense, moral inquiry at more theoretical levels is no different from other subject areas, like economics, psychology, chemistry, and physics. It also helps to explain why we should regard morality as a Kational sphere, of inquiry:By this, we mean an area of discourse and investigation that is 'evidence-based, one in which problems are identified and solutions attempted (at least provisionally) in arguments that are based in the principles of logical reasoning and the best, most relevant ideas, information, and theories that can currently be found. As with any sphere of rational inquiry, we are looking for reasons, supported by the best evidence, for any claims that are made. In stating that morality is evidence-based, we are not supposing that moral reasoning is strictly empirical in nature. That is a matter of dispute within philosophy. But as we will see in the rest of this chapter, this dispute is consistent with recognizing that moral decisions are grounded in reasons that are themselves drawn from the best evidence that can be found to support those decisions. We begin with a Canadian case involving end-of-life decision-making to draw out ideas about morality and moral reasoning and to illustrate their practical relevance. Later in the chapter, we give some further guidance about how to think of the relationship between moral theories and biomedical ethical issues. We will also offer specific remarks about how to approach moral decision-making in biomedical ethics and will discuss briefly the role of professional codes of ethics and law in biomedical ethics. The chapter concludes with some topics for discussion and a list of suggested readings and resources on moral theory and biomedical ethics.

1.2 The Case of Hassan Rasouli Hassan Rasouli and his family immigrated to Canada from Iran in 2010. Not long after their arrival, Mr Rasouli was diagnosed as having a small non-cancerous tumour in his brain. The tumour was successfully removed at a Toronto-area hospital. But complications arose following the surgery, and Mr Rasouli had a heart attack and a serious stroke. He soon became comatose and was put on life support. Early in 2011, he was diagnosed as being in a continuing unconscious state (or "persistent vegetative state"), the initial step toward a diagnosis of permanent vegetative state. The doctors believed that Mr Rasouli was unconscious and would not recover. His family disagreed. The movements that he showed from time to time with his fingers and eyes convinced them that he was conscious. They thought that he had shown signs of improvement and had even given them a "thumbs up" sign, indicating his contentment with his situation. The family's Islamic religious views held that human life is sacred, and they wanted treatment to continue. Mr Rasouli's situation was re-examined later in 2012. It was thought that he might be "minimally conscious," one step above being in a persistent vegetative state. His doctors, however, thought the results were equivocal, as was any claim that there had been an improvement in his medical condition. They reported to Mr Rasouli's family a decision to remove his life support. The family then applied to the courts to prevent the doctors from acting. The doctors argued in court that life support was not conferring any medical benefit to Mr Rasouli and so was futile. As such, neither life support nor its withdrawal constituted treatment, and so the doctors did not require the consent of the family to remove it.

1.2 The Case of Hassan Rasouli 0

The case made its way to the Supreme Court of Canada in 2013. The Court rejected the argument that life support and its removal were not treatments.' It recognized that Mr Rasouli's consent was required to withdraw treatment, or if his wishes were unknown as the court believed and the doctors and the family disagreed over whether further treatment was in Mr Rasouli's best interests, the doctors must approach Ontario's Capacity and Consent Board to resolve the dispute. To date, Mr Rasouli's doctors have not taken this step, and he remains on life support?

1.3 Taking Morality Seriously Is Morality Relative? A response to difficult cases like Hassan Rasouli's by individuals who are new to studying morality or biomedical ethics is to wonder whether their resolution ever reflects anything more than mere opinion:, Moral judgments, it is suggested, are not backed by reasons or evide,--ice in any genuine sense. They reflect only moral beliefs or preferences and nothing more Philosophers describe this sort of position as(relativism about moral value. One way of expressing this idea that is often attractive te students is that morality depends wholly on what a culture or religion believes is right. Thus, whatever moral standards a culture or religion believes in or endorses are morally correct, and that is all that can be said in determining what is right. This position is commonly described as cultural relativism (implicitly extending the notion of culture to include religious groups). Morality, in this view, is purely conventional and arbitrary, and disagreements between cultural or religious outlooks are, ultimately, not amenable to rational evaluation and resolution. It may be tempting to tliink-that the Rasouli case is a reflection of cultural relativism. Rasouli's family had recently immigrated to Canada from Iran where they had been immersed in Islamic culture, so perhaps the disagreement in this case simply reflects different cultural and religious attitudes toward end-of-life decision-making. Philosophers and ethicists generally agree that this is too superficial a way of thinking about morality. What follows in the next few paragraphs is an explanation for this rejection of cultural relativism and relativism about morality generally, along with the beginning of an extended argument for taking morality seriously as a sphere of rational inquiry. It is a striking fact about the Rasouli case that neither the doctors nor Mr Rasouli's family claims authority for their position simply on the ground that "this is what my culture believes." Each party states evidence for their view as part of a dialogue over what should be accepted by all parties as relevant to resolving their dispute. Despite their disagreement over withdrawing Mr Rasouli's life support, the parties agree on a great deal. For example, no one wants to cause Mr Rasouli unnecessary suffering, so they share the values of not doing harm and preventing suffering. The family believes Mr Rasouli is conscious and content, and so keeping him alive preserves his well-being and dignity and recognizes the sacredness of his life. It is doubtful that the doctors think that Mr Rasouli's well-being or dignity is unimportant, so here are other values that they share. The doctors also agree with the Rasouli family that human life is a profoundly important value. Indeed, what else would explain the measures they took to preserve it? It is likely, too, that some of the doctors agree that human life is sacred. But the doctors have doubts that Mr Rasouli is conscious. They believe that even if he were conscious, it would be consciousness of a very low quality, most likely to the point where he would have no awareness of self. If this is true, they would perhaps understandably argue that talk about Mr Rasouli's well-being and dignity is largely misplaced. Given their assumptions about Mr Rasouli's mental state and their belief that continued life support is futile, the doctors also believe that treating him involves a misuse of scarce resources that could be better used elsewhere to preserve the life, well-being, and dignity of others. They would undoubtedly argue that even accepting that human life is sacred unavoidably requires making difficult decisions when resources are scarce and not every human life that requires those resources can be preserved. A difficult choice needs

cultural relativism what is

morally right is determined by whatever moral standards a culture endorses.

11 11 I

O

W

II

all

111141

1 Morality and Moral Decision-Making

non-moral facts states of affairs that can be wholly or adequately described separately from moral values. For example, biological, psychological, social, physical facts.

to be made that best respects the value or sacredness of all human life. The family never rejected this argument. But they are confident that Mr Rasouli has an awareness of self and is the sort of person who would be quite content with even a low level of consciousness. They also believe that the doctors are too pessimistic about his prognosis and so treatment is not futile and resources are not being misspent or misused. Perhaps the only thing that the parties clearly and fundamentally disagree about is Mr Rasouli's mental status and whether it had progressed and will continue to progress. These disagreements are not about moral values per se. They are disagreements about certain non-moral facts, principally certain psychological facts and facts about Mr Rasouli's prognosis. Thus, the parties to this case appear to agree on much, maybe all, of what is important to making a decision. They just disagree about what the evidence is for some of the things they all agree are relevant to making a decision. Unfortunately, perhaps their most important disagreement cannot be resolved in a definitive manner at the present time, since we do not yet have (and may never have) the tools for assessing the psychological states of non-responsive, or mainly non-responsive, individuals. Unless one side can make a reasonably convincing case that the evidence regarding some of the non-moral facts favours its side, the evidence seems to support a moral tie of sorts. In effect, the Supreme Court recognized the need for further adjudication of this evidence by requiring the doctors to present their case for withdrawing treatment to Ontario's Consent and Capacity Board. Why Philosophers and Ethicists Are Generally Not Convinced by Relativism It is possible to see in this discussion why philosophers and ethicists are not impressed by cultural relativism. Despite the evident multicultural character of the exchanges in the Rasouli case, there is no evidence that any of the parties was merely pressing culturally or religiously specific values and opinions. In fact, there is agreement about several familiar values, and this raises different, non-relativist questions about their status. Why do people from different cultural backgrounds agree on them? If they are not treated merely as personal or cultural preferences, do they have a status that is independent of what people happen to believe about them? Are there values, then, that transcend culture? Moreover, even a modestly careful analysis such as conducted here suggests that much that passes as moral disagreement is not disagreement about moral value per se. Indeed, the contested non-moral facts in the Rasouli case are not limited to whether he is conscious. They include issues such as whether lie is experiencing suffering, what his chances of recovery are, and conceptual questions such as what futility and treatment are. Such non-moral disagreements can also extend to metaphysical questions about whether a supreme being exists or not. Recognizing the central role of non-moral facts presents a richer account of moral reasoning, one that turns out to be particularly important in biomedical ethics. Often, we can underestimate the extent of agreement about moral values by failing to notice that much moral disagreement depends on disputes about non-moral facts. Indeed, the emotional heat in moral debate is often generated by such disputes rather than by disputes about values per se. Certainly this was true in Mr Rasouli's case. His family adamantly disagreed with the doctors' assessment of his mental status and prognosis. By contrast, people rarely debate the fundamental importance of prevention of suffering, the promotion of well-being, the protection of human dignity, or the importance of human life. Perhaps a cultural relativist will say this only reflects a fortunate agreement between cultures about value. But this claim cannot just be assumed to be true. It stands in need of support, in part because the striking nature of the shared values, apparently across cultures, invites inquiry about whether something more can be said in defence of them that would explain why they are widely shared.

13 Taking Morality Seriously 0

Before turning to some attempts to describe sources for these shared values in section 1.4, it will be useful to explain other ways in which relativist ideas about morality can be misleading and problematic. People who are attracted by relativistic ideas in ethics (or want to see them discussed) often have the following evidence in mind: 1. The presence o f widespread disagreement in morality 2. The belief that relativism is a good theory for promoting tolerance and respect between cultures and religions 3. Relativism is a good way to resolve the question about "who gets to decide" what is right and what is wrong. None of these claims represents good evidence for cultural relativism. Some of them contradict the very idea of it. To begin with claim 1, the existence of disagreement, even widespread and difficult to resolve disagreement, does not in itself offer any evidence for relativism. Disagreement between individuals or cultures is consistent with there being a fact of the matter that is independent of what people or cultures believe. There were differing beliefs about the shape of the earth for millennia. There was nevertheless a fact of the matter about this. The same may be true about difficult-to-resolve moral disagreements. As well, disagreement is consistent with the evidence weighing more heavily, even convincingly, in favour of one position over another. This is often the case in moral reasoning, as we shall see throughout this text. Moreover, as we have already seen, the Rasouli case suggests ways in which moral disagreements can be overstated. It is an interesting (and recommended) exercise to consider whether it is disagreement about value that lies behind many apparently vexing moral controversies from past to current times. Bad biology, bad sociology, and bad psychology probably do more to explain the denial of women's rights and the justification of slavery over millennia than does disagreement about specific moral values. Finally many moral disagreements are simply not worth taking seriously. The moral, psychological, biological, and sociological evidence we have now overwhelmingly favours recognizing women's rights, for example. In cultures where these rights are not recognized, moral disagreements over women's rights cannot be taken seriously any longer, though complex political and social problems obviously remain about what to do in light of opposition to them. Claims 2 and 3 contradict relativism and actually provide evidence against it. Cultural relativism cannot be defended on the ground that it is a tolerant, respectful outlook, for tolerance and respect are moral values. That would mean relativism is committed to specific moral values that apply to all cultures, contradicting the idea that it is the culture that gets to decide what its moral values are going to be. A consistent reading of cultural relativism would have to remain neutral between tolerance and intolerance and respect and disrespect. Indeed, far from being a tolerant outlook, relativism about moral values looks extraordinarily bleak. It places the morality of Nazis and head-hunters on an equal status beside the morality of peaceful multicultural democratic communities. Even the issue of "who gets to decide" betrays a commitment to moral values, if it is motivated by concern, as it typically is, that those who decide will improperly impose their views on others. This also implies a general belief in the values of tolerance, open-mindedness, autonomy, and individual choice and so contradicts relativism. By contrast, if we take an evidence-based approach to morality, the answer to the "who gets to decide" question is, roughly, that no one gets to decide. It is the best available evidence that should be followed in answering questions about right and wrong. This is reflected in the Supreme Court's insistence that the doctors in the Rasouli case take their plan to the Ontario Consent and Capacity Board so that the evidence has the best opportunity to be properly presented and then evaluated by experts. This is also why we need to tolerate and encourage open debate about morality so that we can continually test moral ideas and claims to make sure they are as well supported as possible, revising and improving them

I I In

I0

111111

1 Morality and Moral Decision-Making

as the evidence changes or is clarified. Morality, including biomedical ethics, like any other sphere of rational inquiry, needs to be committed to principles of tolerance and fair- and open-mindedness in order to encourage productive, ongoing discussion and evaluation of evidence. In fact, it is ironic that cultural relativists would be concerned about the "who gets to decide" issue, since cultural relativism itself gives an answer that seems so clearly at odds with concern about intolerant, authoritarian imposition of moral values. What goes in ethics, according to the cultural relativist, is just whatever a culture agrees to endorse, including deeply oppressive moral standards. Two Alternatives to Relativism: Subjectivism and Moral Realism

subjectivism morality is fundamentally about the expression of feeling or emotion.

moral realism moral facts exist Independently of the evidence for them and about which we can have at least approximate knowledge.

The lesson from this discussion is that the relativism often found among beginners to moral philosophy is a superficial, almost certainly wrong account of the status of morality. Those who believe they are relativists about morality are typically confused about what it means. Usually, they are better described as liberals or multiculturalists rather than relativists, people who believe in general values of tolerance, respect, and open-mindedness. The same objections apply to arguments that are given for an individually based relativism that claims morality is just whatever moral standards an individual happens to endorse, since reasons given for individual moral relativism tend to be the same as those given for cultural relativism. A position that seems close to an individual moral relativism is the view that moral judgments are in some sense fundamentally expressions of human emotions. Philosophers sometimes refer to this position as subjectivism.3 It is similar to moral relativism in claiming that morality is a matter of preference or opinion in some sense. But its formulation is different from moral relativism. Moral relativism claims that morality is whatever an individual or culture believes is moral. Subjectivism thinks that morality is not about beliefs per se. Rather, moral claims use language to express feelings, in particular "pro" and "con" feelings toward certain states of affairs. "Abortion is wrong" thus expresses a negative feeling toward abortion (something like "down with abortion!" or "bah abortion!") and encourages others to adopt a similar attitude. Subjectivism has had somewhat more credibility among moral philosophers, in part because it does not commit the basic errors that are associated with the student's moral relativism. But it is also controversial. It is far from clear that the moral decisions and values in the Rasouli case are simply expressions of human emotion. To the extent that evidence can be given for values reflected there (discussed further below), this does not seem a promising position. Moreover, we seem to debate moral matters to try to get beyond our emotions and come to agreement on values and non-moral facts that are supported by the evidence. Thus, morality is apparently about identifying reasons for certain choices that all should be able to accept independently of their feelings or beliefs. Subjectivists have responses to these positions that are beyond the scope of this introduction. Suffice it to say that philosophical subjectivists also generally agree that moral debate is an evidence-based sphere of inquiry. It proceeds as if something like moral realism is true. Moral realism is the view that there are moral facts that exist independently of the evidence for them and about which we have some, at least approximate, knowledge. Subjectivist views reject moral realism, but they acknowledge that moral debate proceeds very much like evidencebased discussions in other disciplines.4 Moral realism has evidence for it in the way that moral debate is conducted, and it is fair to say that subjectivism struggles to accommodate these features of moral reasoning. It is not necessary, however, to choose between moral realism and subjectivism (and various positions on a continuum between these outlooks) for the purposes of engaging in the material in this text. All that is necessary is to acknowledge that moral discourse has roughly the same sort of general shape as discourse in other areas of inquiry. This is simply to say that, after all is said and done, we want an answer to a problem, reasons why that answer was selected, and reasons why alternatives were not. Nothing short of this should

13 Taking Morality Seriously 10

satisfy anyone in any area of inquiry. We have already seen quite a bit of evidence for this in the discussion in this section. We now turn to a consideration of some general theoretical views about morality that will provide further support for this position and will suggest further ways in which specific moral debates in biomedical ethics can be profitably framed.

1.4 Normative Ethical Theories The discussion in the previous section focused partly on important questions about the status of morality, in particular whether morality is merely a matter of preference or opinion or feeling in some sense. These questions are part of what philosophers refer to as metaethics. As important as metaethics is for understanding morality and moral reasoning, normative ethics is the main focus of this book. Normative ethics inquires more directly and systematically into what the standards are that determine how to act morally (or ethically) and lead a moral (or ethical) life. (The terms "morality" and "ethics" will be used interchangeably throughout this text.) A rough definition of normative ethics, then, is that it is the inquiry into what makes right actions right or wrong. Normative ethics can be divided into general theoretical spheres, which try to identify broad theories that explain the whole, or large parts, of what it is to act morally, and applied ethics, which addresses particular moral issues and problems, sometimes by drawing on the resources of more general theories and often by drawing on moral principles found in everyday moral reasoning (such as those discussed in the Rasouli case) that the more general theories seek to explain and justify. We will begin now by reviewing some of the major general theories of normative ethics. We shall see that these theories can help to explain and make better sense of many of the moral ideas found in the Rasouli case, in health care generally, and in various cultural and religious settings. Toward the end of the chapter, we discuss applied ethics and its relationship to these more general theories. A major division in normative ethics is between consequentialist and deontological theories. This abstract, perhaps intimidating, jargon expresses fairly simple ideas. Consequentialist morality defines rightness as a function of promoting intrinsically good states of affairs, or good consequences, for short. Deontological morality, by contrast, claims that rightness is not exclusively, and perhaps not at all, a matter of promoting good consequences. The task for deontologists, then, is to explain what rightness is as something partly or wholly independent of good consequences. Today, the most prominent consequentialist theory is utilitarianism, and the most prominent deontological theory is Immanuel Kant's ethics, both of which are discussed below. Commitments to each of these positions are suggested in the discussion of the Rasouli case. As we shall see, the parties' interest in preventing suffering and preserving and promoting well-being most dearly reflects utilitariantype reasoning. The idea of respecting Mr Rasouli's dignity appears to be a Kantian ideal. We will now present these theories and then consider some objections to them and their overall relevance to biomedical ethics.

metaethics the inquiry into the nature and status of morality.

normative ethics the inquiry into the standards that determine how to act morally and lead a moral life.

applied ethics the application of moral theories, principles, and ideas to specific moral problems.

consequentialism rightness is a function of promoting good consequences.

deontology rightness is not exclusively, and perhaps not at all, a function of promoting good consequences.

Utilitarianism

There are many formulations of utilitarianism, but the most common formulation is that right actions are ones that maximize overall happiness. Jeremy Bentham (1748-1832) was the first Western thinker to give a systematic account of utilitarianism. John Stuart Mill (1806-73) gives perhaps its classic formulation, or at least the most read and cited formulation, in his essay Utilitarianism.5 Both authors defend versions of hedonistic utilitarianism, the view that happiness consists in pleasure and the absence of pain and what is right is what maximizes net overall pleasure in the world. It is worth emphasizing that this pleasure must be impartially considered. For both Bentham and Mill, in deliberating about what we ought to do, we must take into account our own pleasure and that of others

utilitarianism right actions are those that maximize overall happiness or well-being.

In I

I I UM

1111.0

® 1 Morality and Moral Decision-Making

and be as impartial between them as a disinterested spectator would be. Everyone counts for one, no one for more than one, and it is only the amount of pleasure that is morally relevant. Mill seems to have qualified Bentham hedonism, however, by arguing that there are higher and lower pleasures—for example, intellectual and aesthetic pleasures as contrasted with bodily pleasures—and that higher pleasures are to be preferred over lower pleasures. Bentham's and Mills views are often described as "classical utilitarianism." Utilitarianism has evolved significantly since Mill and Bentham. Philosophers have debated whether pleasure represents a rich enough conception of happiness. Some have argued that "enjoyment" or "satisfaction" or "preferences" are better descriptions. Others have argued that happiness cannot be limited to psychological states and that it consists in having certain character traits (e.g., virtues), developing distinctive human capacities (e.g., intellectual, aesthetic, and athletic excellences), and having certain types of relations with others (friendship and love).6 "Well-being" or "welfare" has often replaced happiness as the end of utilitarian action. It is not clear that it is necessary to take sides in these debates for the purposes of addressing questions in biomedical ethics. Utilitarians incorporate all these ideas of what contributes to happiness or well-being into their analysis of right actions by according them intrinsic or instrumental value. These classes of value refer, respectively, to what is valued for itself (i.e., as an end) or for the sake of something else (i.e., as a means). From a biomedical ethical perspective, it makes sense to consider all such values in trying to assess the well-being of patients and the consequences of actions. It is often common in biomedical ethics to simplify the assessment of good consequences by thinking in terms of preference satisfaction. This is imperfect, but it is a good rough guide in ethical decision-making in health care.

Utilitarianism and Biomedical Ethics non-maleficence do not cause harm.

beneficence prevent harm, reduce harm, and do good.

harm principle interference in the conduct of competent persons is only justified to prevent harm to others.

It is easy to draw from utilitarianism a number of consequentialist principles that are relevant to biomedical ethics, in particular principles of non-maleficence and beneficence. Non-maleficence is the idea, familiar in health care in the Hippocratic Oath, "to do no harm." Beneficence is the idea that we should go beyond not harming and do positive things to prevent harm, reduce harm, and do good. Harm prevention is the idea that it is always a reason for acting if it prevents harm. Doing good is, of course, the idea that it is a reason to act if the action brings about net positive good consequences overall. Each of these consequentialist/utilitarian principles is reflected in the Rasouli case. The family argued that withdrawing Mr Rasouli life support would cause him harm, thus violating a consequentialist principle of non-maleficence. They argued that maintaining his life .support would preserve his well-being and so would be doing good. By contrast, the doctors expressed concern that if Mr Rasouli were in a "locked in" state (i.e., is conscious but cannot communicate), he may be suffering greatly, so withdrawing treatment might prevent harm. We have already seen that settling the disagreement between these consequentialist principles crucially requires that we make some judgments about Mr Rasouli psychological state. This disagreement is thus not about the relevance of the principles but whether the psychological evidence favours one over the other in assessing Mr Rasoulfs well-being. A further utilitarian principle defended by Mill (in On Liberty) is his famous harm principle: "The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others."7 Mill's principle is widely accepted by utilitarians and others as a principle that protects individual liberty and autonomy, since it limits the power of government and society over individual choice to actions that cause harm to others. Actions that affect only the person performing them should not be interfered with, even if they appear contrary to their well-being, assuming the person is competent to make their own decisions. It was Mill's view that by balancing individual liberty and societal interference, individuals will be secure and yet free to behave

1.4 Normative Ethical Theories ®

and lay out their existence as they see fit. Thus, happiness in society will be maximized. We will return to a discussion of the harm principle below when we address the topic of paternalism in biomedical ethics.

Objections to Utilitarianism and Responses It will perhaps be evident now why utilitarianism is a prominent normative ethical theory. It reflects principles that are widely accepted and useful in thinking about moral problems. More generally, utilitarians think that their theory aptly reflects ideas of impartiality that are fundamental to ethical decision-making and proposes a useful, intuitively attractive way of resolving moral disputes—that is, by weighing the consequences of actions and choosing those that promote the most overall happiness or well-being of those affected. However, many theorists do not find utilitarianism convincing. Perhaps the most common and general objection to utilitarianism is that it does not adequately respect the inherent dignity of individuals. If it would maximize happiness to enslave the talented to work on our behalf, or have vast inequalities in wealth, or mistreat some minorities, or find innocent persons guilty in a court of law, or harvest the organs of one unsuspecting patient to save five, utilitarianism seems committed to saying it is our moral obligation to do so. Another objection is that utilitarianism is too impartial and personally demanding, requiring more than what morality demands. For example, does it really mean that people cannot give any preference to promoting the well-being of intimates, family members and friends, if happiness could be better promoted by helping others? Does utilitarianism demand that I give up one of my kidneys if that would help others? Must I not donate the kidney to my son if giving it to my neighbour would more effectively promote happiness? Another objection is that morality is often not forward-looking in the way that utilitarianism implies. Often, we have moral obligations because of agreements and relationships that we entered into in the past. It might do more to improve happiness if I break my contract with a business partner, but arguably I have a duty to keep the contract because of my earlier agreement. Indeed, the point of a contract is to bind the parties to their previous agreement in such circumstances. These sorts of objections show something of the complexity of our attitudes toward morality and the difficulty utilitarianism has in explaining that complexity. However, utilitarians have responses. The most common response is that a more careful analysis of each of the objections shows it is not plausible that utilitarianism is committed to these sorts of morally problematic positions. It is argued that a more careful analysis shows that enslaving some, accepting great disparities in wealth, and so on would not maximize happiness. More generally, utilitarians can say that these objections confuse the utilitarian principle (what is right is what maximizes overall happiness) with method for making moral decisions. Utilitarians argue that actions are right to the extent that they promote happiness, but this standard does not require that we consciously deliberate over and aim to maximize happiness with each of our choices. Rather, they argue that the most effective method for promoting happiness, given human cognitive and emotional limitations, is to follow a set of easy to understand and follow general rules and principles of morality including commitments to basic rights. These should be put aside only in circumstances when it is evident that they will not contribute to overall well-being. We should then try to incorporate these precedents back into our understanding of a set of basic welfaresupporting rules and principles. A second response is for utilitarians to try to brazen it out. Rather than argue (as the above does) that there is no unacceptable gap between our ordinary moral intuitions and the dictates of utilitarianism, some utilitarians admit that there can be but then argue that it is our intuitions that should give way. After all (so the argument goes), what people ordinarily think about moral matters undeniably contains elements of prejudice, superstition, misinformation, stupidity, and bad logic. Thus, if we are sympathetic and benevolent

paternalism interfering with

someone for their own good.

uii I

10

LI

II

al

uu

1 Morality and Moral Decision-Making

persons, we will adopt the Principle of Utility and (at least sometimes) revise our ordinary moral intuitions in the light of that rather than reject that Principle because it conflicts with those intuitions. For example, we ordinarily do think it is wrong to frame and execute an innocent man to save more lives (by preventing a riot, say). But a utilitarian might respond by saying that further reflection should make us reconsider. There is no morally relevant difference in the bare distinction between doing something (framing) and letting something happen (a disastrous riot). The only thing that counts is consequences, and in this regard, since all the parties in question are innocent, the choice is between bad consequences for one or for many. Is it thus so clear (the utilitarian might now ask) that utilitarianism gives improper moral advice? And just as we might revise our first judgment here, a utilitarian might argue that we should not be too quick to reject the theory because it conflicts with common moral intuitions. A third option is for utilitarians to retreat to what is called Negative Utilitarianism. On this view, there is no obligation to maximize happiness. Our sole duty is not to harm others or, if harm is unavoidable, not to do more harm to others than would be done by some alternative that we could have chosen instead. Negative Utilitarianism immediately removes the objections that utilitarianism places unacceptably heavy burdens on individuals and forces them to do repugnant things to promote the general good. Some utilitarians have embraced this view, and some have argued that it is Mill's view. Others, such as Peter Singer—who is perhaps the most influential practical utilitarian of our time—have held a view that is intermediate between maximizing and Negative Utilitarianism. Singer's position is that if we can prevent something very bad from happening without thereby sacrificing anything of comparable moral significance, we are wrong not to do it.8 Negative Utilitarianism may seem not to impose strong enough obligations to help others. Maximizing utilitarianism may seem to impose excessively onerous obligations to do so. Singer attempts to steer a middle course between these two views. Whether utilitarianism can be defended in the above sorts of ways against the stock objections to it is still a matter of controversy. But it must be admitted that these utilitarian responses have plausibility, and the theory is very much alive today. Nevertheless, opponents of utilitarianism find the responses ultimately too indirect, question-begging, ad hoc, and filled with problems of their own, including whether the responses and modifications just considered end up collapsing into a modified deontological outlook. We will not settle this dispute here. Suffice it to say that opponents of utilitarianism think that there are better ways to account for the nature of morality. It is to these theories that we now turn. Kantian Ethics

categorical imperative or the supreme principle of the categorical imperative the central concept in Kant's ethics; given several formulations.

Immanuel Kant (1724-1804) was an eighteenth-century philosopher who made important contributions to all spheres of philosophy and is generally regarded as the most important philosopher of the Enlightenment. Kant's views about morality have been particularly influential and are arguably reflected all around us in modern systems of constitutional democracy and human rights codes and laws. They also appear to play a prominent role in health care and biomedical ethics. For Kant, morality is deontological because it is not about promoting good consequences. Consequentialist moralities overlook the key aspect of morality which Kant argues is a matter of acting with a proper moral motive or good will. Acting with a moral motive means acting with respect toward others by recognizing their inherent dignity. Thus, Kant's ethics are often described as an ethic of respect for the dignity of persons. According to Kant,to act with a moral motive is to be motivated to act according to a principle that he calls the "categorical imperative." Kant calls moral duties "categorical imperatives" and contrasts them with "hypothetical imperatives." Understanding the latter helps us to understand the former. Hypothetical imperatives are things that we ought to do if we want or desire certain ends, as when one says, "If you want to lose weight, eat less." You can get rid of the imperative "eat less" by

1.4 Normative Ethical Theories

simply changing your desire to lose weight. By contrast, the imperative force of categorical imperatives cannot be removed in that way. They are directives grounded in reason referring to things that we must do whatever our desires may be. We must not exploit others and must show gratitude to those who have helped us, for example, whether or not we want or do not want to. This binding force is what characterizes a moral duty. They are things that we must do on pain of acting wrongly, whatever our desires happen to be. The categorical imperative occupies the same place in Kant's ethics as the Principle of Utility does in Bentham's and Mill's. Kant gives several formulations of what he means by this. Each formulation of the categorical imperative is supposed to be equivalent to the others, though this is controversial. The basic ideas behind his notion of a categorical imperative are reflected in two formulations: 1. The Formula of the Universal Law: Act only in accordance with that maxim through which you can at the same time will that it become a universal law9 2. The Formula of Humanity. So act that you use humanity, whether in your own person or that of another, always at the same time as an end, never merely as a means.10 Those new to Kant often find these formulas frustratingly difficult to understand, but they reflect simple, reasonably compelling ideas that can be readily explained. The first formula is designed to recognize that moral actions must be internally consistent and follow principles that are consistently applied in relevantly similar contexts. Thus, morality is not merely a matter of mutable emotional choices or whims or mere conventions but is a matter of acting on reasons that guide actions in all relevantly similar circumstances. If we make a moral choice, it must be one that is coherently expressed (for example, not self-contradictory) and consistently observed. The first formula is thus formal, covering all choices, and separates morality from mere emotion-driven preference or social convention. However, the first formula gives limited specific guidance about what our moral maxims should be. Here is where the second formula enters. It is the most distinctive and compelling aspect of Kant's theory and explains why it has been attractive to many who reject consequentialist moralities. The second formula is generally recognized as articulating a theory of respect for persons. We do this by respecting their ultimate value and thus their inherent dignity. This is what Kant means when he says we must always treat ourselves and others as ends. Kant argues this is reflected in the recognition that each of us has of our own ultimate value or worth. We have a conception of ourselves as having worth "beyond price," as he puts it, because of our capacity for reasoned choice. Of course, if we recognize this of ourselves, it is a simple requirement of consistency that we must recognize it of others, who have a similar conception of their worth and dignity. Thus, a universal morality emerges as an extension of an agent's recognition of her own absolute worth. But this still does not explain how we protect and preserve the dignity of ourselves and others. Here is where the notion of not treating persons as mere means plays a key role. According to Kant, humans must unavoidably use each other as means to get what we want. The key, however, is to do so in a way that respects them and preserves their dignity. Kant says we do this by not using them as mere means—that is, as mere objects or tools. Humans are not mere objects or tools. It is their capacity for choice or autonomy that so distinguishes them and, according to Kant, gives them a sense of their dignity. We avoid treating persons as mere means and respect them as ends by respecting their capacity for choice or autonomy. Kantian Ethics and Biomedical Ethics

Kant's ethical theory may still seem fairly abstract, but the ideas are intuitive and practical nevertheless. The simple upshot of all this is that when we enter into relations with others, we preserve their dignity by obtaining their consent before we ask them to do things. We have seen this idea already in the priority the Supreme Court put on recognizing that

®

111

8

H i1

11111

1 Morality and Moral Decision-Making

rights moral claims with special normative, apparently non-consequentialist, force.

Mr Rasouli had a right to consent to withdrawal of treatment. We can see it also in his family's position. They believed that Mr Rasouli was consenting to continued treatment with his "thumbs-up" gestures. They also believed that he is the sort of person who would consent to remain in his current circumstances. The Supreme Court's recognition that even if Mr Rasouli's preferences were unknown, its ruling that a decision should be based on what was in his best interests reflects this idea. A Kantian would say that this should be understood, so far as possible, as what a reasonable person in Mr Rasouli's circumstances would accept as being in his best interests. Thus, the doctors' decision to remove treatment might have been motivated partly by the idea that being conscious in a "locked in" state would be a profoundly bleak circumstance that few, if any, would find tolerable, and so we can be reasonably confident that Mr Rasouli would consent to withdrawal of treatment if he were in such a circumstance. Kantian ethics has been attractive to many. One key advantage is that it appears to avoid the objections levelled at utilitarianism. (As a good exercise to see if you understand Kantian ethics, take the description of it just given and see how it motivates many of the above objections to utilitarianism.) It is grounded in the idea that humans are not mere objects but are fundamentally self-determining beings. As such, the worst thing that can happen to them—the loss of their dignity—means that they have become, or are being used as, mere objects. It nicely explains why the loss of our mental faculties is perhaps our most profound fear, something that patients and their health care providers will readily appreciate. It explains why assaults, fraud, lying, emotional manipulation, exploitation, and so on are immoral (they treat individuals as mere objects by ignoring their right to free and informed consent) and even why we cannot simply interfere with others for their own good without their consent as consequentialism can seem to condone. Kantian ethics' relevance to biomedical ethics is clear, then, given the central role that consent to treatment plays in health care decisions. Kantian ethics even seems to explain and support the idea of democracy as government by consent of the people. Moreover, Kantian ethics does not seem as demanding as consequentialist ethics. Society is not required to adopt policies to promote maximum happiness overall but to establish the circumstances whereby individuals can exercise meaningful autonomy and so preserve human dignity. A Kantian social safety net, then, would be demanding fiscally but arguably not as demanding as a utilitarian one. And it would not be so demanding that a father could not choose to give his kidney to his son, if that is what he wishes to do. Finally, Kantian ethics also gives substance to the idea of rights, which are claims with special normative, non-consequentialist force. For example, rights to autonomy and to consent to medical treatment, rights to freedom of expression, security of the person, and not be discriminated against on the basis of race or sex (or other grounds) all protect us from being used as mere means and would seem to be justified whether or not respecting them would maximize overall happiness in every case or generally. In this sense, rights are sometimes said to "trump" conflicting consequentialist claims based in promoting overall well-being or happiness. Kantian ethics help to make sense of these ideas (though consequentialists will respond with their own account of rights as, in effect, important goods or benefits that are to be widely upheld). Objections to Kantian Ethics and Responses

However, important objections to Kantian ethics have been raised. Often, Kant's theory is criticized for its absolutism and lack of flexibility. Kant contributes to this impression in some o f his remarks—for example, about lying and suicide. But it is fair to say that critics of Kant often fail to capture the complexity of his views on these subjects. For example, Kant's published views about lying are quite nuanced and do not clearly reflect the position that is often attributed to him that one must never lie, including to "the murderer at the door" who asks for the whereabouts of his intended yid im.il And it is important to distinguish Kant's application of his own theory (and potential misapplication of it) from the theory

1.4 Normative Ethical Theories

itself, which is often considerably more flexible than Kant and some of his opponents recognize. For example, it seems to follow from Kant's theory that we may be able to lie to the murderer at the door if it is the only reasonable option to prevent a murder. This could be made into a universal law without contradiction under Kant's first formula. It directly prevents another person from being treated as a mere means as per the second formula. A Kantian might also say that the murderer cannot object to being lied to and being treated as a mere means, since the plan to murder someone presupposes the belief and motive that it is acceptable to use others as a mere means. More generally, it would make a sham of the whole Kantian system of respect for the dignity of persons if people who plan to use others as mere means were protected, and even facilitated, in carrying out their actions in ways that undermine the very protections the theory seeks to provide.12 Another objection to Kantian ethics is the concern that its anti-consequentialism goes too far. Surely consequences do matter in moral deliberations, but Kantian ethics sometimes appears to reject their relevance completely. Reading Kant as thinking this is also a misunderstanding. Kant objects only to the idea that the goal of morality is consequentialist. He claims, rather, that it is acting with a correct motive. But for Kant, far from being irrelevant, consequences are crucial in assessing whether a motive is morally correct or not. A Kantian must consider whether the consequences of an agent's behaviour treat the individuals it affects as ends and not as mere means—for example, whether consent is needed. So a Kantian can admit that a certain type and range of consequences are relevant. What a Kantian denies is only that the ultimate goal of morality is to produce good consequences—happiness or the greatest good overall. Although it is beyond this introduction to explain, Kant also argues that we have some duties to promote well-being, since the promotion of well-being is necessary to establishing the conditions for autonomous choice. A Kantian would say that to make the production of good consequences into an absolute duty (as utilitarians do) is inappropriate for a variety of reasons, including that it is too demanding and may be used to override the dignity and autonomy of particular individuals. However, there are other, arguably more serious, objections to Kantian ethics, in particular with respect to its attitudes toward non-rational beings and to suffering in general. Kantian ethics seems to allow that we can use non-rationals--for example, incompetent children, the mentally ill, the senile, and animals—as mere means, perhaps even using them in cruel medical experiments if it suited us to do so. This is a frightening prospect. However, Kantians can reasonably respond that we have responsibilities to children and others to bring them, or return them, to autonomy, so the objection does not work for such beings. Kantians can also say that if there are rational non-human animals, they are entitled to respect too. But there are still the cases of the permanently mentally ill, the senile, and non-rational animals. Their dignity and suffering do not seem to matter at all in Kantian ethics. To be fair, Kant does say that we should treat these categories of beings humanely. We should do this in order not to harden ourselves as a way of practising respectful behaviours that are necessary for our relations with autonomous persons. But to many, this is an inadequate response. They argue that these beings also deserve respect as ends and that causing suffering is always bad and a moral reason not to do it. These are reasonable concerns about Kantian ethics. However, the objections do not undermine the Kantian insight of the importance of respecting dignity and not using others as mere means. Rather than undermining Kant's theory, the objections suggest a need to extend the circle of respect further than Kant recognized. After all, these other classes are conscious beings and so are not mere objects. Suffering is something bad that happens to them and that cannot happen to mere objects. It might be appropriate, then, to extend an ethic of respect and recognize that these beings also have legitimate claims based on a recognition of their dignity. As with utilitarianism, there do not seem to be any clear, knock-down objections to Kantian ethics. Moreover, its appeal seems to lie in attractive ideas that are not so clearly or directly captured by utilitarian theory. We turn now to consider some other theoretical

8

on

8

1

I I :111

111111

1 Morality and Moral Decision-Making

approaches to normative ethics. These approaches are not as prominent as either utilitarianism or Kantianism, but it will be useful to describe and examine them to see what they might add to our understanding of morality.

Virtue Ethics virtue ethics morality Is principally a matter of realizing in action character traits (virtues) that express being a good person.

Virtue ethics reflects a long tradition in the West and elsewhere that emphasizes aspects of character that go into being a good person. An implication of virtue theory is that morality is not so much, and perhaps not at all, a matter of observing or following the sorts of rules and principles that are reflected in consequentialist and deontological moralities. When we face a moral problem, virtue ethicists say that we should look to the example set by good people and ask questions like "what would so-and-so do?" or "what would a good person do?" Virtue ethics reflects another familiar idea, then—namely, that morality is concerned, at least in part, with what makes a good person. Virtue ethicists often claim that the complexity of moral questions cannot be resolved by recourse to moral rules. When we face novel or difficult moral situations, we are often better off asking ourselves what certain moral paragons or exemplars with whom we are familiar would do. Virtue ethics was given its classic formulation in Western philosophy by Aristotle (384-322 BCE), but virtue ethics theories are present in other cultures as well—for example, in Confucian and Buddhist thought. Contemporary virtue theorists give formulations that differ at least somewhat from Aristotle's. However, we focus here on Aristotle's account, since it remains the point of departure for virtue ethics and contrasts effectively with utilitarianism and Kantian ethics. According to Aristotle, virtues are psychological states expressed in action that are necessary to functioning well as a human being and, therefore, represent the dominant, necessary goods in a happy (or eudaimon) There are two classes of virtues: intellectual and moral (or character) virtues. Both are necessary to making correct choices. Today, we might say that intellectual virtues are those required for applying the canons of evidence-based inquiry including virtues involved in empirical observation and deductive and non-deductive reasoning. They would also include virtues like fair-mindedness in the evaluation of evidence, willingness to consider opposing ideas, and ability to be self-critical and change one's position in light of new evidence or better understanding of existing evidence. Moral virtues, by contrast, are expressed in actions that reflect the training and habituation of emotions in ways that conform with and support correct action. As such, character virtues reflect a mean between excess and deficiency of feeling. The mean is not arithmetical. The appropriate mean is always contextual, but there can be excess or deficiency of feeling in any context. The mean reflects the idea that correct action is threatened by having too much or too little of a relevant feeling. For example, courage cannot reflect too much or too little fear of danger. Too little fear will lead to foolhardy actions, and too much fear will lead to cowardice. Thus, character virtues exist as a mean between vices reflecting excess and deficiency of feeling. Training of character virtues by practice develops habits so that we are able to do the right thing at the right time with the correct motive. Courage, temperance, friendliness, truthfulness, wit, generosity are a few examples of character virtues discussed by Aristotle, but it is easy to think of many others (e.g., resilience, patience, persistence, compassion, fidelity, tolerance). It is pretty obvious that all of the character virtues just described, and others besides, play a central role in making good doctors, nurses, and other health care providers and should be cultivated as part of their training. It is less clear how useful they are as decision-making tools in many circumstances, however. For example, it is difficult to see how asking "what would a good doctor do?" could resolve the dispute in the Rasouli case, for there is little reason to think that the doctors or any of the parties in this case lacked virtue in any significant measure. Remember that the Supreme Court did not say that the doctors' wish to withdraw treatment was wrong but that it needed to be justified before

1.4 Normative Ethical Theories 8

a legal tribunal that had been set up to adjudicate disagreements between doctors and patients. We saw, also, that the dispute might be resolved if we could answer some key disputed questions about non-moral facts. It is thus not clear how asking what a good person would do could resolve this question or determine what should be done in the absence of an answer to what the relevant non-moral facts were. Moreover, asking what a good doctor would do seems to evolve quickly into a consideration of principles of the sort already discussed: good doctors act to prevent harm, promote well-being, and respect autonomy and dignity. Finally, we often want to know in difficult situations what it is that makes a right action right in order to understand our choices and to help guide us in similar situations. Simply asking what a good person would do without relying on such principles does not give us that clarity. Many virtue theorists accept these sorts of criticisms. They argue that Aristotle is, nevertheless, correct that there is practical value in training moral and intellectual virtues. These help us to "tune" our psychological capacities in ways that can assist us in recognizing moral problems and relevant information and therefore in making better moral decisions. And Aristotle may also be right that the virtuous person has the dominant necessary components for leading a happy life, assuming (along with Aristotle) that certain other goods are also present, including material means, friends, a good education, and a prosperous, stable social environment. A final note about virtue ethics and the Rasouli case: Aristotle thought that virtues do not consist strictly in possessing certain psychological dispositions but are states expressed in action. Thus, he says, we would not consider someone who slept his whole life virtuous or happy, even if he had the requisite virtuous dispositions.'4 If the virtues are the dominant necessary parts of a happy life, we might imagine Aristotle arguing that Mr Rasouli could not be said to be a virtuous person or to have a meaningful, let alone happy, life if his consciousness was compromised and he was unable to act. Mr Rasouli's family appears to disagree with Aristotle, as would most utilitarians and Kantians. They would say that if Mr Rasouli is content in his circumstances, or if he chooses to live this way without violating the dignity of others, his life has meaningful value.

Care Ethics Care ethics is a recent development in normative ethics. It emerged initially from the empirical work of Harvard psychologist Carol Gilligan in the 1980s." Gilligan's examination of adolescent boys' and girls' responses to moral dilemmas suggested that the boys' solutions tended to reflect traditional rule or principle-oriented approaches to ethical theory, such as those found in utilitarian and, especially, Kantian ethics. The girls, by contrast, seemed to approach the dilemmas "in a different voice." Gilligan found that they took a more contextual approach that prized the preservation of relationships and caring. Gilligan claimed that the boys' behaviour reflected traditional psychological views of moral development, which give priority to principles and logical reasoning and tend to systematically overlook the different, caring ethical attitudes reflected in the girls' approaches to moral decision-making. It is fair to say that Gilligan's empirical findings have been controversial. Nevertheless, they led to attempts to develop a normative ethic of care, either as a stand-alone normative ethical theory or as a supplement to existing normative theories. The main features of the theory are evolving and so difficult to summarize. We can do no better than give the succinct account offered by Johnna Fisher, the late author of this chapter in previous editions of this text, herself a care ethicist, who described care ethics as committed to two basic principles (given in order of importance): 1. To minimize or avoid harm 2. To create, maintain, and protect positive relations

care ethics morality is fundamentally about caring, fostering positive relationships, and related concepts.

I I III

G

111111

1 Morality and Moral Decision-Making

Also reflected in these principles is care ethicists' scepticism about the commitments to impartiality that are elements of utilitarian and Kantian ethics. They argue that moral decision-making is more contextual than these theories allow and that they contain inadequate provision for the protection of the vulnerable and for the fostering of positive relations. We have seen already that there appears to be some truth to these criticisms of Kantian ethics, although a path forward for addressing them may be possible. It is difficult to see how utilitarian ethics systematically overlooks either of the principles of care, in light of the evident utility involved in minimizing and avoiding harm and fostering positive, caring relations. And both theories are more contextually sensitive than their critics often admit. But there is, nevertheless, reason to think that the importance of relations of caring have not received their due in normative ethical theory generally, and care ethics has at least helped us to develop a more nuanced understanding of caring in moral deliberations. Care ethics is also sometimes seen as an extension of virtue ethics. Caring is pretty obviously a virtue, but again it has not been given any notice in traditional virtue theory. Care ethics is also often taken as a critique of patriarchal biases found in traditional normative ethics. It emphasizes the protection of the vulnerable and repairing the status of those who are exploited or discriminated against (e.g., women and some minorities). It is understandably taken as a foundation for feminist ethics, although many care ethicists have resisted the label of a feminist ethic. Care ethics has also been attractive to those working in health care. Nurses in particular have taken an interest in the theory. Critics of care ethics are generally concerned that its focus as a normative ethical theory is too narrow. The protection of the vulnerable and fostering positive caring relations does not tell us how to deal with many ethical conflicts. For example, it is not clear what assistance care ethics can offer in the Rasouli case. Everyone involved agrees that a caring approach to Mr Rasouli is necessary, including preservation of the positive relations he has with his family and health care providers. But what underlies taking a caring approach to Mr Rasouli appears to be concern about his well-being and respect for his autonomy, his dignity, and his family's preferences. There is also an obvious issue about just allocation of scarce resources. We have a duty to care for Mr Rasouli, but we also have a duty to care for others who are vulnerable and who need the scarce resources Mr Rasouli is using. Principles such as preserving caring relations, minimizing suffering, and protecting the vulnerable are of little help in resolving these conflicts because they apply to all parties who need the resources. If so, we may need to look outside care ethics to find solutions to such problems. Moreover, it is hardly clear that utilitarian and Kantian ethics are as bloodlessly impartial as care ethicists sometimes suppose. A utilitarian can defend a mother who chooses to rescue her own drowning child over rescuing another child. There is obviously substantial utility in preserving caring maternal relationships. It is essential to promoting the well-being of children and establishing important familial relations of trust that dependent children need to rely on. As well, both families would be emotionally devastated by the loss of a child, but a mother who does not choose to rescue her own child would have to live with the added psychological disutility of choosing to let her own child die, a disutility she would not experience by letting an unrelated child die. A Kantian could say that the mother was free to act on her preference for saving her own child, since making this choice does not use anyone as a mere means or violate anyone else's rights. Perhaps a Kantian would also say that the choice to engage in acts of procreation brings with it duties others do not have to nurture offspring to autonomous adult personhood. Thus, a Kantian could say that the mother has a special prior, autonomously chosen obligation to care for her child, an obligation that she does not have toward other nearby children who are also in danger. Care ethicists have responded to criticisms by recognizing the importance of principles of justice, equality, impartiality, and autonomy. This presses the issue as to whether the theory is distinctive or only shows how other theories have overlooked important elements of morality that should be incorporated within them. But even if care ethics must

1.4 Normative Ethical Theories e

rely on principles imported from outside, it emphasizes something standard normative ethics does not—namely, the importance of trust and relationships. To be sure, Mill and Kant envisioned ethics as a practical discipline designed to prescribe or prevent certain actions, but they did not have the particular problems of clinical ethics at the centre of their thought. Care ethicists do—the theory was invented to better handle the kinds of conflicts that routinely occur in these contexts—and one of their enduring contributions is to emphasize the importance of forming and fostering relationships of trust between the parties in question. Once this is done, conversation and communication will be better, and negotiation over irreconcilable goals can take place with greater prospects of success rather than deteriorating into the ultimatums we find in the Rasouli case. Caring can, then, be crucial to productive moral decision-making and inquiry. Of course, the best efforts to produce the appropriate relationships may fail—this may indeed have happened in the Rasouli case—but the importance of these relationships in clinical ethics is emphasized in care ethics as it is not in any other theory. Again, we do not have to settle disagreements about the status of care ethics to do work in the area of biomedical ethics. As with the other normative theories we have considered, we may draw on the insights offered by care ethics as they appear relevant to moral decision-making.

1.5 Moral Reasoning in Biomedical Ethics The Principles of Biomedical Ethics A now-standard approach in biomedical ethics is to begin moral deliberation by recognizing the role of the following four principles of biomedical ethics'° in identifying and deliberating about ethical problems in health care: 1. 2. 3. 4.

Non-maleficence (do not cause harm) Beneficence (do good, prevent harm) Autonomy (respect preferences) Justice (treat like cases alike; be fair)

These principles are all suggested in the normative ethical theories we have reviewed. Non-maleficence and beneficence are often said to be consequentialist or utilitarian principles, whereas autonomy and justice are Kantian principles, but in fact all normative ethical theories sign on to these ideas in some way or another. They are prima facie principles. That is, they are not absolute requirements but represent reasons for action that hold in the absence of countervailing reasons. Thus, for example, to say that we ought prima facie to respect the autonomy of patients does not mean that we must always do what patients want but rather that we must act in accordance with their wishes unless we have a good reason not to (e.g., that doing so will put others at risk or treat them unfairly). Prima facie duties thus shift the onus onto those who do not want to carry them out to justify overriding them. This is consistent with a recognition that there is no agreed-upon theory of normative ethics but that there are, nevertheless, principles that can be drawn from these theories, are reflected in everyday moral deliberations, and are generally agreed upon in biomedical ethics. These so-called "four principles" may not be the only principles relevant to biomedical ethics, but their generality and the fact that they are recognized in theory and practice make them a useful starting point for thinking through moral problems. We have seen that these principles must be supplemented by recognizing the fundamental role that nonmoral facts play in ethical decision-making. In health care, these facts include clinical facts (diagnoses, prognoses, treatments, risks, and benefits); quality of life/death with and without treatment; the preferences of patients, their families, and health care providers; and any other relevant contextual facts (such as costs, what the law requires, and who else's interests are affected)."

principles of biomedical ethics non-maleficence, beneficence, autonomy, and justice.

YY I

18

YI

I I .111

our

1 Morality and Moral Decision-Making

This suggests a simple methodology for thinking about moral issues and problems in health care: 1. There is no moral problem if the non-moral facts are clear and the principles point to the same course of action. 2. There is a moral problem only i f a) there are conflicts between the principles or within them, or b) there is uncertainty about the meaning or relevance of the principles, or c) the non-moral facts are unclear. The four-principles approach to moral problems sketched above provides a way of finding the best evidence, and so the weightiest reasons, for resolving a dilemma or problem in clinical or health care contexts generally. Undoubtedly, moral deliberation is more complicated than this description suggests. Indeed, the principles of biomedical ethics are often criticized as being abstract and incomplete. But their relevance and power to assist with moral decision-making cannot be readily denied, and we have seen that their role in normative ethics is generally accepted or implied in every theory we have examined.

Ranking Principles

weak paternalism it is not permissible to interfere with a competent person's informed decision to accept or refuse an offered treatment.

strong paternalism it is permissible to interfere with a competent or incompetent person's decision about treatment if it is in their best interest.

As a way of enhancing the use of the above methodology, a limited ranking of these principles is generally accepted in health care ethics and law. As we saw earlier, paternalism is the concept of interfering with someone's liberty for their own good. It thus recognizes a possible conflict between beneficence and autonomy. In ethics and law, weak paternalism is a recognized principle that gives important priority to autonomy over beneficence. It is the view that individuals who have the autonomous capacity to consent (i.e., can understand the nature and consequences of their decisions, are fully informed, and consent voluntarily) have an absolute right to accept or refuse an offered treatment. That is to say no treatment can be imposed on them against their will. Autonomy trumps beneficence in these situations. We have seen this position defended in Kantian ethics and in utilitarian ethics in the form of Mill's harm principle. Kantian ethics says to treat an autonomous person against their will is to use them as a mere means, which we must never do. Mill's harm principle offers a different justification. Mill's view is that autonomous people generally know their own interests best. He also believed that people are by nature meddlesome and will interfere gratuitously in the lives of others if given the opportunity. From these considerations, Mill defends weak paternalism and (like Kant) opposed strong paternalism or the view that we can interfere with either non-autonomous or autonomous persons against their will any time it is in their best interests. Weak paternalism is usually held to have further implications for interfering with nonautonomous persons against their will. In fact, it is only with these implications that the notion of paternalism, or interfering with someone for their own good, really applies to weak paternalism. Specifically, interfering with non-autonomous or incompetent persons is permissible only if: 1. 2. 3. 4.

The preferences of the individual while autonomous are not known. The individual is at risk of harming themselves. We can intervene without causing more harm than is prevented. We choose the least intrusive measures available, including those involving the least restrictions on the autonomy and liberty of the individual. 5. Our action is not discriminatory. 6. All reasonable efforts are taken to bring the person interfered with to see that the interference is justified. (While acknowledging that interference can take place without such agreement, Mill makes the important point that it should not unless it is a case of utmost necessity because to impose it always "partakes of the degradation of slavery.")

1.5 Moral Reasoning in Biomedical Ethics

19

The conditions weak paternalism establishes on interfering with non-autonomous individuals are designed to prevent meddlesome behaviour that is not in their interests and to recognize that even non-autonomous persons are entitled to some respect for the choices they may make. It is, however, often a hard principle to live by, for autonomous individuals may refuse potentially life-saving interventions. But weak paternalism is generally accepted in law and ethics today. By contrast, strong paternalism is often tempting. It is difficult to stand by when preventable tragedies are occurring as weak paternalism requires (for example, when a competent Jehovah's Witness refuses a life-saving blood transfusion). Strong paternalist interventions are sometimes justified on the grounds that patients will later thank the health care providers for their actions. It is important to see, however, that this defence of strong paternalism makes it ambiguously distinct from weak paternalism, or a watered-down version of it. It implies that patients do not truly appreciate their current choice and that they would hypothetically consent to an intervention if they understood their situation better, since they would later thank their health care providers. In effect, then, this version of strong paternalism challenges the patienth capacity to make an autonomous decision. The problem with such justification is that it is very difficult to know what people will accept later. Kantians and utilitarians would say that it is a violation of dignity and simply presumptuous to believe that others know an individual's best interests better than they do. It is not unreasonable, then, to believe that thinking about respect for autonomy in this way—as hypothetical consent to what others conscientiously believe is in a person's best interests—rationalizes all manner of meddling in others' lives, which Kantian and utilitarian ethics guard against. Arguably, it is for these reasons that weak paternalism has won out in mainstream biomedical ethics—it is enshrined in every professional code of health care ethics in Canada—and in Canadian law, though debates continue in philosophy journals and in hospital hallways.

1.6 Professional Codes of Ethics and Law Students of biomedical ethics should be aware of and study the codes of ethics of the health professions. These codes can further understanding of the ethical issues involved in health care and can supplement the methodology just described. Health care professions have also developed policy statements on specific health care issues that are ethically informative and need to be consulted where applicable. For example, the Canadian Medical Association has policies on end-of-life decision-making, abortion, physicians' obligations during pandemics, protection of patient confidentiality, conflicts of interest with the pharmaceutical industry allocation of scarce resources, and research ethics, to name a few.'8 But it is important to recognize the substantial limitations of ethics codes and even of carefully elaborated issue-specific policies for moral decision-making. To begin with, it is a misconception to suppose that ethics codes and policies can supply ready answers to all or even most genuine ethical problems.19 This is already implied in the discussion so far. For example, codes of ethics do not tell us how to resolve conflicts among professional obligations. How does a doctor reconcile a patient like Mr Rasouli's right of access to health care with the competing claims of other patients who need access to the resources he is using? What should be done when rights to confidentiality conflict with duties to care for other patients—for example, when a genetic test discloses information that may affect other family members? Also, codes of ethics do not tell us what the relevant non-moral facts are in a case or how to resolve disputes about them. Nor do codes of ethics answer conceptual questions that may be at the core of a disagreement—for example, about what constitutes death or a futile treatment or confidentiality or harm. Ethics codes and policies can help to guide decision-making to the extent that they emphasize recognized standards of professional conduct that must be accounted for in health care decision-making. But generally, the ethical heavy lifting happens outside of

conflict of interest a situation in which a person has an obligation to a person or institution that is in opposition to another interest or obligation, which could corrupt the decision-making of that person.

I I HI

I

MIA

1 Morality and Moral Decision-Making

primary goods general all-purpose means, such as liberty, opportunities, education, and health, necessary for the pursuit of life goals.

these codes when principles or policies conflict, conceptual problems need to be addressed, or there is dispute or uncertainty about the contextual features of a situation. These issues arise to some degree or other whenever a genuine ethical problem is identified that must be worked through. Like morality law is intended to guide human action. In any civilized community, law is designed to follow morality. As a result, courts' reasons for judgments can often read like the moral arguments they are, and they are often a rich source of ethical guidance. For example, we will see in later chapters that the courts have offered helpful moral guidance on issues regarding consent to treatment and end-of-life decision-making, in some cases articulating novel moral arguments (e.g., the Supreme Court of Canada's recent decision to legalize physician-assisted dying, discussed in Chapter 6). But law has its limitations as an ethical tool. To begin with, if we are making moral decisions, we should generally consider the relevant moral features of a case first and what the law requires later. This is because moral decision-making is about making moral decisions, not legal ones. Sometimes law and morality conflict, and we have to be aware of this possibility and ask what is to be done. Can the disagreement be managed in some way? Should a test case be tried? Should efforts be made to change legislation? Should the law be ignored? Questions like these often pose their own ethical problems and dilemmas. They may be obscured if deliberation starts with a consideration of what is legally required. More generally, we cannot suppose that what is legally correct is necessarily morally correct. Nor can we suppose that what is morally correct requires a law to enforce it or that what is morally wrong requires a law to prohibit it. These are all reasons to be careful not to confuse moral and legal deliberations. It is instructive to note that the law in modern democracies is restricted with respect to imposing on individuals religious or other conceptions of what gives life meaning or what life goals should be pursued. This is, once more, a reflection of respect for autonomy and also the product of a hard-won recognition that social stability is best achieved when people and governments are not permitted to impose on others their views about how to live well. These considerations, which can be drawn from the ethical theories we have reviewed, support and accommodate a multiculturalism that is often taken to be a hallmark of Canadian democracy and was reflected in the Rasouli case. Arguably, the role of government in modern, pluralist, multicultural societies is to ensure that all-purpose means, sometimes called primary goods, like basic democratic rights, freedoms, and opportunities, are fairly distributed to people so that they can make their own choices by their own lights about how to live, as long as those choices do not interfere with the rights of others to these general all-purpose means or primary goods. A similar point applies to decisions in biomedical ethics. Health is one of those all-purpose means that modern democracies have recognized a duty to protect and fairly promote. The goal of health care, then, should be its provision. It should not be used as an occasion to impose on patients religious or other conceptions about what choices or ends make a life worth living.

1.7 Looking Ahead We hope that by now it is evident that biomedical ethics is an important, challenging, and, indeed, exciting sphere of inquiry. It demands the thoughtful attention of anyone with a serious interest in health care. Most of the ideas expressed in this introductory chapter will be reflected on and developed in later chapters. In the chapter introductions dealing with specific topics in biomedical ethics, we draw connections to the ideas and theories outlined in this chapter to further clarify their relevance and to deepen understanding of the topics covered. The text is designed in such a way that it is possible for a reader to examine the chapter topics in no particular order. Each chapter contains a set of study questions, a selection of actual cases that illustrate the ideas of the chapter in their natural habitat, and

1.7 Looking Ahead ® a list of suggested readings. Although we have not attempted to cover every topical issue in

biomedical ethics, we hope that we have provided a solid general introduction to the field that will encourage further interest, investigation, and study.

Notes 1. Cuthbertson v. Rasouli, 2013 5CC 53, https://scc-csc.lexum.com/scc-csc/scc-csden/item/13290/index. do (accessed 10 May 2016). 2. The case study draws on material in Philip C. Hebert, Good Medicine: The Art of Ethical Care in Canada (Toronto: Doubleday Canada, 2016), ch. vi. 3. James Rachels and Stuart Rachels, The Elements of Moral Philosophy, 8th edn (McGraw-Hill Education, 2014). 4. Simon Blackburn, Essays in Quasi-Realism (Oxford University Press, 1993). 5. J.S. Mill, Utilitarianism, 1863, http://wwwutilitarianism.com/millhtm (accessed 10 May 2016). 6. David 0. Brink, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University Press, 1989), ch. 8. 7. J.S. Mill, On Liberty, 1859, http://www.utilitarianism.com/ol/one.html (accessed 10 May 2016). 8. Peter Singer, "Famine, Affluence, and Morality," in Steven M. Cahn and Peter Markie, eds, Ethics: History, Theory, and Contemporary Issues (New York and Oxford: Oxford University Press, 1998), 800. 9. Immanuel Kant, "Groundwork of the Metaphysics of Morals," translated by Mary J. Gregor, in The Cambridge Edition of the Works of Immanuel Kant: Practical Philosophy (Cambridge: Cambridge University Press, 1999), 73. 10. Ibid., 80. 11. Allen W Wood, Kantian Ethics (Cambridge: Cambridge University Press, 2008), ch. 14. 12. Tamar Schapiro, "Compliance, Complicity, and the Nature of Nonideal Conditions," The Journal of Philosophy 100, no. 7 (2003): 329-55. 13. Aristotle, Nicomachean Ethics, 2nd edn, translated by Terence Irwin (Indianapolis: Hackett, 1999). 14. Ibid., 1098b35-1099a7. 15. Carol Gilligan, In a Different Voice: Psychological Theory and Women's Development, revised edn (Cambridge, MA: Harvard University Press, 1993). 16. Tom L. Beauchamp and James F Childress, Principles of Biomedical Ethics, 7th edn (Oxford University Press, 2012). 17. Albert Jonsen, Mark Siegler, and William J. Winsdale, Clinical Ethics: A Practical Approach to Ethical Decisions in Clinical Medicine, 8th edn (McGraw-Hill, 2010). 18. Canadian Medical Association, Issue-Specific Policies, https://www.cma.ca/En/Pages/issue-specificpolicies.aspx (accessed 9 May 2016). 19. Dale Beyerstein, "The Functions and Limitations of Professional Codes of Ethics," in Earl R. Winkler and Jerrold R. Coombs, eds, Applied Ethics: A Reader (Cambridge, MA: Blackwells, 1992), 416-25.

1.8 Study Questions 1. How does the material in this chapter support the claim that morality is a rational sphere of inquiry? 2. Explain why there is less fundamental disagreement about moral values than it would appear from observing moral debates. 3. To what extent d o the normative ethical theories discussed overlap, and to what extent do they diverge? 4. Take the methodology from section 1.5, and use it to conduct a formal analysis of the Hassan Rasouli case in section 1.2. Give an argument (i.e., present the evidence) for how you would resolve the disagreement between the doctors and Mr Rasouli's family based on what you know about the case. What are the limits, if any, to this methodology? 5. Look at an ethics code from one of the health care professions (readily available online). Identify a list of concepts that require clarification in order to apply the code. Try to clarify the meaning of those concepts.

11 111 i

e

m

II A

111111

1 Morality and Moral Decision-Making

6. Make a list of moral controversies in biomedical ethics and elsewhere, and identify what the disputed non-moral facts are in each case. How many of these controversies depend most crucially on the resolution of these disputes? Which ones do not? 7. How do the arguments given in section 1.3 show that respect for multiculturalism is not a moral relativist position? How is such respect for multiculturalism evident in the practice of health care in Canada?

1.9 Suggested Readings and Resources General Introductions to Moral Theory Pojman, Louis, and James Fieser, Ethics: Discovering Right and Wrong, 7th edn (Wadsworth Publishing, 2012). Rachels, James, and Stuart Rachels, The Elements of Moral Philosophy, 8th edn (McGraw-Hill Education, 2014). Schafer-Landau, Russ. Whatever Happened to Good and Evil? (Oxford University Press, 2004). Classic Texts in Normative Ethics Aristotle, Nicomachean Ethics, 2nd edn, translated by Terence Irwin (Indianapolis Hackett, 1999). Held, Virginia, The Ethics of Care: Personal, Political, Global (Oxford University Press, 2007). Kant, Immanuel. "Groundwork of the Metaphysics of Morals," translated by Mary J. Gregor, in The Cambridge Edition of the Works of Immanuel Kant: Practical Philosophy (Cambridge University Press, 1999). Mill, J. S . On Liberty, 1859, http://www.utilitarianism.com/ol/one.html. Utilitarianism, 1863, http://www.utilitarianism.com/mill1.htm. Singer, Peter, Practical Ethics, 3rd edn (Cambridge University Press, 2011). Canadian Biomedical Ethics Books and Resources Bioethics for Clinicians, Canadian Medical Association, http://www.cmaj.ca/site/misc/ bioethics_e.xhtml. Canadian Bioethics Society, https://www.bioethics.ca. Hebert, Philip C. Good Medicine: The Art of Ethical Care in Canada (Doubleday Canada, 2016). University o f Toronto Joint Centre for Bioethics, http://jcb.utoronto.ca. Yeo, Michael, Anne Moorehouse, Pamela Kahn, and Patricia Rodney, eds, Concepts and Cases in Nursing Ethics, 3rd edn (Broadview Press, 2010). Leading Biomedical Ethics Journals Bioethics Clinical Ethics Cambridge Quarterly of Healthcare Ethics Hastings Centre Report IRB: Ethics and Human Research Journal of Clinical Ethics Journal of Medical Ethics Nursing Ethics

Medical Decision-Making Self-Determination and Deciding for Others 2.1 Introduction For more than 2000 years, health care was dominated by the Hippocratic tradition. According to this tradition, health care providers had an obligation to act in the best interest of patients. This meant that if what the patient wanted conflicted with what was good for the patient, patient preferences were overridden. Health care providers knew best, and health care was paternalistic. All this has now changed in Canada and the rest of the Western world. The Hippocratic tradition has been replaced by an autonomy-based ethic known as "patient/family-centred care" whereby, as much as possible, patients and families are the decision-makers. This transformation reflects the triumph of weak paternalism over strong paternalism for the reasons given in Chapter 1 (section 1.5, page 18), and the triumph is complete. The philosophy of patient/family-centred care is accepted by the law in every province and territory in Canada and in every Canadian professional health care code of ethics The acceptance of the philosophy of patient/family-centred care has momentous consequences for medical decision-making. In this view, patients are entitled to make decisions for themselves insofar as they have decision-making capacity and their decisions are voluntary and informed. (We will return shortly to examine these three key conditions.) They can accept or ref use any treatment that is offered. They can live at risk, with neglect, self-neglect, and abuse. Or they can choose not to live at all. Their lives and choices are their own, and all that others can do is explain and argue. On the other hand, if the patient lacks one or more of these decision-making requirements to qualify as the decision-maker and the lack or lacks cannot be remedied before the decision must be made, the decision will be made by a substitute decision-maker. This substitute decision-maker will typically be the person who is most knowledgeable and concerned about the patient, but when such a person cannot be found it may be the physician or a court-appointed individual. The substitute decision-maker can make decisions on the basis of two possible kinds of judgments. The first is a substituted judgment. Substituted judgments are judgments that the decision-maker thinks the patient would have made. In making decisions on the basis of substituted judgments, the substitute decision-maker looks at the world with all the patient's hopes, fears, beliefs, desires, and eccentricities and makes the decision he or she reckons the patient would have made. The second is a best-interest judgment. Bestinterest judgments are judgments the reasonable person in the patient's position would make. Here the substitute decision-maker puts himself or herself in the shoes of the

substitute decisionmaker a person who makes decisions for a patient when the patient lacks decision-making capacity. A substitute decision-maker is sometimes also called a surrogate decision-maker or (when the person has been legally designated as the decision-maker by the patient when capable) a proxy decision-maker.

substituted judgment the judgment that a patient would make for himself or herself.

best-interest judgment the judgment that the reasonable person in the patient's position would make.

nu I

II

111

111111

® 2 Medical Decision-Making Patient

N Decision-making capacity

Lacks the characteristics necessary to make decisions for him/herself

Voluntary decision

Full information

Patient decides

Family + Physicians decide

N Substituted judgment

Best-interest judgment

Figure 2.1 Decision Tree reasonable person and makes the decision on the basis of more objective societally determined criteria, such as reduction of suffering and preservation of quantity and quality of life. When substituted judgments and best-interest judgments conflict, the autonomydriven philosophy of patient/family-centred care tells us to give systematic precedence to substituted judgments. The philosophy of patient/family-centered care enables us to construct a decision tree for medical decision-making. The best circumstance is when the patient has decisionmaking capacity, can act voluntarily and is fully informed, and tells us what he or she wants. The next best is when the patient fails to have one or more of these characteristics but for whom there is a firm substituted judgment. The worst is when we do not know what the patient would want and have to fall back on a best-interest judgment. This tree can be set out as in Figure 2.1. The decision tree in Figure 2.1 is widely accepted in both ethics and law But applying it raises difficulties at every point. It is sometimes hard to determine whether the patient has decision-making capacity, is acting voluntarily, or is fully informed. If the patient fails to have one or more of these features, it can be hard to identify the appropriate substitute decision-maker. Once the substitute decision-maker has been identified, there can be daunting epistemological problems in arriving at what the patient would want or what the reasonable person in the patient's position would want. The rest of the chapter is concerned with exploring these matters. The section headings follow the order of topics in the decision tree reading from left to right.

Decision-Making by Patient decision-making capacity individuals can understand the nature and consequences of their decisions and are not subject to distorting mental illness or mental states. Individuals who have these abilities are also referred to as capable or competent.

Decision-Making Capacity Three conditions must be satisfied for patients to have decision-making capacity. (These conditions must also be satisfied by a substitute decision-maker.) 1.

The patient must be able to understand the nature and consequences of his or her decisions. Understanding is a cognitive capacity, and it is a matter of degree. The clinical test

Decision-Making by Patient S

for determining the degree of understanding a patient has is to explain the situation in ordinary language and ask the patient to repeat that back in his or her own words and manner. 2. The patient must be free from distorting mental illness. Mental illness also comes in degrees and can range from almost normal states of neuroses to significantly impairing states such as manias, clinical depression, neurotic compulsions, and paranoia. Persons may be able to understand everything but still have their decision-making capacity compromised by mental illness. A patient with clinical depression, for example, may have no cognitive deficit but may simply not care about his or her well-being. 3. The patient must be free from other distorting mental states. One may be capable of understanding everything and not be impaired by mental illness but still be subject to distortion from mental states such as nervousness, passion, agitation, excitement, pain, shock, and fatigue. These states can be induced by such factors as disease, injury, alcohol, narcotics, or social circumstances and can be as impairing as an inability to understand or think clearly by reason of mental illness. Since decision-making capacity is analyzed in terms of the above conditions and they are matters of degree, there are different degrees of decision-making capacity. The question thus arises as to what the appropriate degree of capacity is. Allen E. Buchanan and Dan W. Brock take up this question in "Standards of Competence" (see page 32). They argue that the appropriate standard varies depending on the degree of risk involved in the decision. If the risk of saying yes or no to offered treatment is great, the standard must be high. If there is no significant risk, the standard may be low And if the risk is in between, the standard must be proportionate to that risk. They thus arrive at what they call a "process" standard of competence that tries to steer a middle course between respecting the choices of the patient and protecting his or her well-being. Brock and Buchanan cite some medical examples of such decision-making, and Cases 1 and 5 (see pages 91 and 95) provide a couple more.

Voluntary Decision Once patients are identified as having decision-making capacity, two further conditions must be satisfied before they can make any decision. Their decisions must be voluntary and informed, and both conditions call for further discussion. To make a voluntary decision, the patient does not have to be free of all influences. In medical contexts, the patient must take into account what the physician and health care providers say about diagnosis and prognosis, how the decision will affect the patient, what the family thinks should be done, how the family and others will be affected, and so on. These considerations will certainly influence the patient's decision, but the line between voluntary and non-voluntary decisions is not drawn in terms of whether the decisions are influenced or not but in terms of whether the influences are undue. Some influences are compatible with voluntariness, but undue influences are not. Voluntary decisions can thus be defined negatively. To say that a decision is voluntary is to say that it is not a result of influences such as coercion, duress, manipulation, or adverse social or medical conditions, and so on. Coercion and duress include things such as physical violence or threats of it, as well as threats to interfere improperly with rights, such as rights to treatment or food. Manipulation may take the form of withdrawing or threatening to withdraw one's help, company, or affection. It is also often difficult to distinguish manipulation from forms of persuasion that involve coaxing and imploring or pressure from physicians and other health care providers. Adverse social conditions such as poverty, lack of social supports, chronic pain, or dependency may likewise exert influences so that decisions are no longer significantly free.

voluntary decision

a decision made when a person is not subject to undue influences.

II All

111111

2 Medical Decision-Making

Evaluating whether any of these conditions is present, and if so to what degree, often calls for subtle judgment on the part of physicians, nurses, and social workers. This is especially so because health care providers spend a good part of their time trying to talk patients into accepting things patients initially do not want. This is an important part of the job and may involve attempts to persuade, but it cannot involve coercing or manipulating. In the end, the decision must still be the patient's, and distinguishing between caring concern and coercion calls for delicate judgment. Susan Sherwin, in "A Relational Approach to Autonomy in Health Care" (see page 35), raises a more general problem about voluntariness from a feminist perspective. On her relational analysis of autonomy, all of an individual's choices are embedded in social structures, and some of them can be (and in the case of women often are) oppressive. Thus, when women make health care decisions, these decisions may not be truly autonomous even though no undue influence such as coercion, manipulation, or duress can be identified. Sherwin does not for this reason reject the current emphasis on autonomy in health care or recommend that others make decisions for the oppressed. The ideal solution is to remove the conditions of oppression. Until then, it may be necessary to provide the vulnerable with greater support in decision-making. Informed Decision informed decision a decision made when a person knows everything he or she would want to know before making the decision.

We now turn to the informed condition. For a patient to make an informed decision, he or she must know everything that would make a difference to that decision. In "The Nuts and Bolts of Obtaining Consent to Treatment" (see page 45), L.E. Rozovsky and F.A. Rozovsky tell us that this level of disclosure requires (among other things that do not concern us here) that physicians tell patients: 1. The diagnosis and probable prognosis; 2. The reasonable treatment options available, including that of non-treatment; 3. The nature of those treatments, i.e., whether they are surgery, medication, electroshock, and so on; 4. The potential risks and benefits of treatment and non-treatment. Of special importance and interest is the disclosure of risk. In deciding what risks to disdose, the physician must tell the patient enough so that the patient will not regret his or her decision. The physician's aim is to avoid the circumstance in which patients can say that if they had only known of such and such a risk, they would not have made the decision they did. On this the law and ethics agree, but the law and ethics can diverge on what must be disclosed to avoid such regret. According to Rozovsky and Rozovsky, the law does not require that the physician tell the patient about those risks that can be taken to be common knowledge or are minor and unlikely. But if the risk is serious (e.g., death, deafness, or paralysis), the law requires that it be disclosed even if there is only a vanishingly small chance of it occurring, This requirement, however, can cause problems, and the four readings that follow are all concerned in different ways with challenging that it is ethically required. Ruth R. Faden and Tom L. Beauchamp, in "The Concept of Informed Consent" (see page 47), are concerned that obsession with the level of disclosure required by law may impair autonomous action. They begin by distinguishing two senses of informed consent. Sense 1 amounts to "autonomous authorization." In this sense, the patient or subject has given consent if the patient has a substantial understanding of an intervention, is substantially free of the control of others, and intentionally authorizes a professional to perform that intervention. By contrast, Sense 2 amounts to "effective authorization." In this sense, an individual has given consent if the rules and regulations that define the institutional practice of informed consent are satisfied. Thus, if the institution defines informed consent

Decision-Making by Patient ®

as the satisfaction of a set of rules Rl-Rn, as soon as those rules are satisfied informed consent has been obtained. They then argue that an informed consent in Sense 1 ("autonomous authorization") can fail to be an informed consent in Sense 2 ("effective authorization") because it does not conform to certain defining rules (e.g., witnessing or wait-period requirements). And conversely, "informed consent" in Sense 2 need not constitute "authorization." Giving decisionmakers a full list of alternatives, risks, and benefits is no guarantee that they understand what they are consenting to, and indeed it may be more confusing than helpful. Sense 1 and Sense 2 informed consent are not necessarily inconsistent. The ideal circumstance, in fact, is when the conditions that define Sense 2 informed consent also satisfy Sense 1 informed consent. But these senses can diverge, and the worry is that the demand for institutional efficiency and fear of legal liability, which encourage effective consent, will overshadow the importance of obtaining informed consent in the sense of autonomous authorization. And to the extent that it does this, the primary moral rationale for informed consent will be lost. Howard Brody, in "Transparency: Informed Consent in Primary Care" (see page 51), likewise questions the requirement that informed consent demands the disclosure of all serious risks, however improbable. Brody identifies four circumstances in which physicians commonly make treatment recommendations: 1. 2. 3. 4.

The intervention is low-risk and routine, and the patient almost always chooses it. The intervention i s not routine but seems t o offer a clear benefit with minimal risk. The intervention offers substantial chances for benefit but also very substantial risks. The intervention offers substantial risks and extremely questionable benefits, but all alternatives do the same.

On the one hand, Brody contends that if the physician's decision to recommend an intervention involves a careful consideration of risks and benefits (as it presumably would in circumstances 3 and 4 above), those risks must be disclosed. On the other hand, Brody claims that most interventions in primary care are not arrived at in such circumstances. Far more commonly, treatment recommendations are made in circumstances like 1 and 2 above. Virtually all interventions have some remote and serious risks, but Brody argues that if those risks do not figure in the physician's recommendation, they need not be disclosed. The task of informing the patient is complete as soon as the physicians reasoning for the recommendation has been made transparent to the patient, the patient is invited to ask questions, and those questions are answered. The article by Joseph Kaufert and John O'Neil, "Culture, Power, and Informed Consent: The Impact of Aboriginal Health Interpreters on Decision-Making" (see page 56), also challenges the legal standard b y identifying a circumstance i n which i t seems impossible to meet and desirable to ignore. Their article details the difficulties of communicating with persons who have a radically different culture, language, and educational level than the physician and hence of getting what we would recognize as informed consent. In the case Kaufert and O'Neil report, the translator did much of the communicating, and the patient was not told of the nature of the recommended procedures in any detail, the risks and benefits, the available alternatives, or the possibility of non-treatment. The patient was also largely silent, with the result that the physician proceeded more on the basis of "assent," whereby the patient accepts treatment in the sense of not refusing it, rather than of "consent," whereby the patient has a good understanding of what she is getting into. One may lament this, and perhaps regard it as demeaning, but what is the alternative? Onora O'Neill, in "Informed Consent and Public Health" (see page 60), distinguishes between two different contexts in which consent can be sought: therapeutic and public health. After identifying a host of problems for therapeutic consent (which in her view

II

WlI

1114W

2 Medical Decision-Making

are intractable), she contends that in public health contexts where consent is sought for (say) immunization, the aim is radically different. In clinical contexts, the only interests at stake are those of the patient. But in public health contexts, the public has a keen interest in compliance with the proposed treatment because a public good (herd immunity in the case of vaccination) is achieved only if many participate and set back if they do not. O'Neill then argues, relying on Mill's harm principle, which allows interference with the liberty of individuals to prevent harm to others (see Chapter 1, page 18), that when public health risks are sufficiently grave, informed consent is not necessary for treatment. But even if withholding information (e.g., about the risks of vaccines) will achieve a public health good, there is a question of whether that is appropriate in an open democratic society as opposed to disclosing all and relying on public education to secure compliance, supplemented as necessary by quarantines or penalties for non-compliance.

Decision-Making by Substitute Decision-Makers We now turn to the problem of making decisions for patients when they cannot make them for themselves. Such decision-making will be done by substitute decision-makers and involve making substituted judgments or best-interest judgments. Substituted Judgments

advance directive a legal document made by persons with decision-making capacity to guide medical decisions when they lack decisionmaking capacity.

proxy directive an advance directive in which persons designate who is to make decisions for them when they lack decision-making capacity.

instructional advance directive an advance directive in which persons designate what medical decisions they would or would not want made when they lack decision-making capacity.

Substituted judgments are the judgments that decision-makers think the patient would have made and are the best basis for decision-making when patients lack decision-making capability. To arrive at a substituted judgment, the decision-maker may rely on an informal „ values inventory" of the person. By spending time with a person, discussing the news, watching medical shows together, and so on, a substitute decision-maker can often get a good idea of the patient's basic values and beliefs, which can be used to determine what the patient would want in certain circumstances. Another way of arriving at what the patient would want comes from an advance directive made by the person when capable. Advance directives are documents that allow individuals to project their health care preferences into a time when they cannot make decisions for themselves. As such, they promise to make decision-making better for patients, lessen the burden on health care providers and substitute decision-makers, and simplify and (sometimes) by limiting treatment reduce costs to the system. For these reasons, advance directives have become enormously popular. Many health authorities urge individuals to make them, and they are endorsed by the Canadian Medical Association in "Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures" (see page 76). Advance directives come in two varieties. The first is a proxy directive, which allows individuals to say who they would like to be their substitute decision-maker. In the absence of a proxy directive, the decision-maker will be selected from a legal line that begins with spouse, adult children, and parents and carries on until a single decision-maker is identified. Proxy directives are thus a good idea if one wants someone other than the first (or any) person on the legal line to make decisions for him or her. The second is an instructional advance directive, which allows individuals to say what they would like done in what circumstances. When the instructional advance directive is legally valid—i.e., made in accordance with the laws of the jurisdiction and clearly identifies what the individual wants done in the circumstances at hand—the directive has all the force of the patient's in-person instruction. As such, decisions made on its basis are not substituted judgments but the patient's judgments. However, instructional advance directives are not always or

Decision-Making by Substitute Decision-Makers

even typically legally valid or perfectly clear, and in such cases substitute decision-makers have to try to understand what the patient would have wanted and so make a genuine substituted judgment. Advance directives work best when patients unambiguously identify what care they want and are unlikely to change their minds. They can thus be recommended if patients have ideological views about certain procedures (e.g., no blood products because they are a Jehovah's Witness) or have had experiences with specific interventions (e.g., ventilation) that they do not want to repeat, or are visibly disabled and fear that they may not get maximum care if they do not ask for it, and so on. But in the absence of such special views, it is difficult for individuals to say what they would or would not want in what health care conditions. This difficulty may stem from not knowing what it is like to be in possible medical conditions; what conditions they can adjust to; what their prognosis is with and without treatment; and the nature, degree of invasiveness, and risks of available treatments. It is thus hard for individuals to confidently say that they do not want life-saving treatment if they become demented or quadriplegic or want electro-convulsive therapy if they suffer from serious depression. There is also the difficulty that an individual's preferences may change, so what they want now may not be what they will want or appreciate receiving in the future. Advance directives thus come with perils that individuals considering them must weigh against their promised benefits. These issues are discussed in the reading by Angela Fagerlin and Carl E. Schneider, titled "Enough: The Failure of the Living Will" (see page 63). The reading is also notable for reviewing empirical evidence about the usefulness of living wills, much of which raises doubts about their value. While many who have studied advance directives and seen medical decision-making up close enthusiastically support them, many others think that patients will do better by leaving the decision to be made at the time by a loved one they trust.

Best-Interest Judgments Best-interest judgments come into play when the patient does not have decision-making capability and a substituted judgment is not available. Candidates for best-interest decisionmaking include patients in all age groups from infants to the old and with varying degrees of decision-making capability from the comatose to the almost-capable. The patient may be a stranger or someone the substitute decision-maker knows very well but does not know what he or she would want in the circumstances at hand. The aim of best-interest judgments (as of substituted judgments) is to give patients the health care they would have chosen for themselves if they were able to. But since what they would have chosen is unknown, decision-makers have to gamble, and the standard used to determine what treatment is appropriate is that of the "reasonable person," meaning "most people in the patient's circumstances." The idea is that providing the care that most people with the patient's medical condition, known religious or cultural beliefs, and social situation would want has the best chance of giving the patient the care he or she would have chosen. Children and Infants

Some of the most poignant and difficult cases of best-interest judgments involve children, and in this section we consider two sorts of hard cases. The first is presented by Christine Harrison, Nuala P Kenny, Mona Sidarous, and Mary Rowell and addresses some of the problems encountered in involving children in the decision-making. ("Involving Children in Medical Decisions"; see page 76). They consider the case of an 11-year-old child who has had experience with severe illness and the burdens of treatment and does not want further treatment. The authors favour a family-centred approach to the decision-making

YY 1

dl

II hill

1111.1.1

2 Medical Decision-Making

whereby the interests of everyone are taken into account. But any preferences that the child, family, or treating team may have must be measured against the best interest of the child. The view that we must always act in the best interest of the child is sometimes challenged,' but it is the standard view, and the Bioethics Committee of the Canadian Paediatric Society (CPS), in its position statement 'Treatment Decisions Regarding Infants, Children, and Adolescents" (see page 80), is clear and firm that the interest of the child must be paramount. Our second sort of hard case concerns two of the most common and worrisome problems facing physicians in the treatment of children. The first is when the family requests more life-sustaining treatment than the physician thinks appropriate. The second is when the family requests less than the physician wants to provide and asks the physician to withdraw or withhold life-sustaining treatment (wLST). When can physicians say "no" to such requests? We will take the cases in turn. The CPS, in "Treatment Decisions Regarding Infants, Children, and Adolescents" cited above, addresses the first of these situations. It contends that physicians can WLST when there is consensus that there is a high degree of probability that: 1. There is irreversible progression to imminent death; 2. Treatment is clearly ineffective or harmful; 3. Life will be severely shortened regardless of treatment, and the limitation or withdrawal of interventions will allow greater palliative and comfort care; 4. Lives will be filled with intolerable distress and suffering that cannot be prevented or alleviated. In saying this, the CPS sets a limit on how much treatment families can request and receive. But this view conflicts with that of the Judicial Council of the American Medical Association (AMA) when it wrote: "In desperate situations involving newborns, the advice and judgment of the physician should be readily available, but the decision whether to exert maximal efforts to sustain life should be the choice of the parents" (see page 82). Given that decision-makers must always act in the child's best interest, which of these should be followed depends on which will best promote that interest. But it is hard to say where that interest lies. The AMA can reasonably say that life is so valuable that it is in the best interest of the child to take great risks to preserve it. On the other hand, the CPS can say that continued life-sustaining treatment will likely do the child no good and cause it great suffering and that, thus, providing maximal treatment is not a gamble that is in the child's best interest. Most pediatricians accept some version of the CPS's view, but the AMA's position is unrefuted, and the issue continues unresolved. We now turn to the opposite problem of when the family wants the physician to WLST and the physician disagrees. Since physicians must always act in the child's best interest, they can clearly say "no" to family requests to WLST when they think life-sustaining treatment will benefit the child. No one can disagree that beneficial treatment must be provided, but disagreement can occur over what treatment is beneficial, and examples of situations in which it is appropriate to WLST would be helpful. The Royal College of Paediatrics and Child Health identifies five such situations. These are when (1) the child is brain dead, (2) the child is in a permanent vegetative state, (3) treatment will only delay death and not alleviate suffering, (4) the child will survive with unacceptable impairments, and (5) "the child and/or family feel that in the face of progressive and irreversible illness further treatment is more than can be borne," and they may request to have "a particular treatment withdrawn or to refuse further treatment irrespective of the medical opinion that it may be of some benefit" (see page 81). Situations 1-4 are uncontroversial and generally accepted. Situation 5, however, introduces a new and novel possibility and calls for further discussion. Situation 5 as stated

Decision-Making by Substitute Decision-Makers 31

above is taken from the "Summary" that the Royal College provides of its guideline (as is the selection in the text). But in the body of the guideline, the phrase "in the face of progressive and irreversible illness" is omitted. This is a substantial difference. If the phrase is included, treatment can be removed because of intolerable suffering only when the child is already in a seriously compromised condition and a healthy child would not emerge whatever treatment is provided. On the other hand, if the phrase is omitted, treatment can be removed if suffering is intolerable even if a healthy child could emerge. On this view, just as substitute decision-makers for incompetent adults can refuse very burdensome treatments (e.g., radical chemotherapy) even if a good result is possible, so could families do the same for their children. The college never removes this ambiguity, and readers should carefully consider which view is better. If intolerable suffering despite the possibility of a good outcome is sometimes allowed as a reason for the family to say "no" to life-sustaining treatment, the sphere of family autonomy is expanded and that of physician autonomy is shrunk in the way described in Chapter 4 (see "Professional Autonomy and Professional Ethics"). Religion and Culture

Other cases calling for a best-interest determination are those in which religious or cultural views conflict with the prevailing views of Western society. One common such case concerns Jehovah's Witnesses. It is now standard to allow competent adults to make a voluntary and informed choice to forego potentially life-saving blood transfusions but not to allow them to make such decisions for their children. Jehovah's Witness parents, however, often disagree with this stance, arguing that it deprives them of the most important decisions they could make for their child—namely about their spiritual welfare and the possibility of everlasting life. Every court that has addressed the issue has rejected this parental claim, but the rejection is typically not accompanied with a philosophically satisfying explanation. If the Jehovah's Witness belief is that to receive a blood transfusion is to lose eternal life, one may ask why it is in the best interest of the child to receive the transfusion. It does not seem a good gamble to risk infinite loss to make a finite gain, whatever the probabilities of each may be. As long as it is possible that giving blood to a child will deprive that child of eternal life, it is not clear why withholding blood should not be the best strategic choice for the child. Thus, unless one takes the plunge and says that there is no possibility whatsoever that the Jehovah's Witness belief is true and provides reasons for thinking that, the argument for forcing blood is not complete. Cultural practices are also often in conflict with the norms of Western society. In "Ethical Relativism in a Multicultural Society" (see page 83), Ruth Macklin discusses three particularly difficult cases of this sort. One concerns what to do when the physician has bad news about a father and the family says, "Don't tell dad," explaining that in their culture such information is not shared. A second is when individuals in certain cultures (Macklin cites the Navaho) do not want to hear about risks and possible bad outcomes and the physician wants to secure consent for a serious surgery. Still another involves rituals (such as female circumcision) or healing practices (such as inflicting burns) that are socially mandatory or common in a family culture but child abuse in ours. These sorts of matters are often dealt with briskly by appealing to hospital rules or the law, but the philosophic and ethical rationales for whatever decisions are made are not always perfectly clear or satisfying for a society that celebrates multiculturalism. Note 1.

By, for example, John Hardwig in "What about the Family?" The Hastings Center Report 20, no. 2 (March/April 1990): 5-10.

ii

di]

wu

C

I 2 Medical Decision-Making

2.2 Determining Decision-Making Capacity

Standards of Competence Allen E. Buchanan and Dan W. Brock

I. Different Standards of Competence

A number of different standards of competence have been identified and supported in the literature, although statutory and case law provide little help in articulating precise standards.' It is neither feasible nor necessary to discuss here all the alternatives that have been proposed. Instead, the range of alternatives will be delineated, and the difficulties of the main standards will be examined in order to clarify and defend [our decision-relative analysis]. More or less stringent standards of competence in effect strike different balances between the values of patient well-being and self-determination. A. A Minimal Standard of Competence

An example of a minimal standard of competence is that the patient merely be able to express a preference. This standard respects every expressed choice of a patient, and so is not in fact a criterion of competent choice at all.' It entirely disregards whether defects or mistakes are present in the reasoning process leading to the choice, whether the choice is in accord with the patientb own conception of his or her good, and whether the choice would be harmful to the patient. It thus fails to provide any protection for patient well-being, and it is insensitive to the way the value of self-determination itself varies both with the nature of the decision to be made and with differences in people's capacities to choose in accordance with their conceptions of their own good. B. An Outcome Standard of Competence

At the other extreme are standards that look solely to the content or outcome of the decision—for example, the standard that the choice be a reasonable one, or be what other reasonable or rational persons would choose. On this view, failure of the patient's choice to match some such allegedly objective outcome standard of choice entails that it is an incompetent choice. Such a standard maximally protects patient well-being—although only according to the standard's

conception of well-being—but fails adequately to respect patient self-determination. At bottom, a person's interest in self-determination is his or her interest in defining, revising over time, and pursuing his or her own particular conception of the good life. [With so-called ideal or objective] theories of the good for persons, there are serious practical or fallibilist risks associated with any purportedly objective standard for the correct decision—the standard may ignore the patient's own distinctive conception of the good and may constitute enforcement of unjustified ideals or unjustifiably substitute another's conception of what is best for the patient. Moreover, even such a standard's theoretical claim to protect maximally a patient's well-being is only as strong as the objective account of a person's well-being on which the standard rests. Many proponents of ideal theories only assert the ideals and fail even to recognize the need for justifying them, much less proceed to do so. Although ascertaining the correct or best theory of individual well-being or the good for persons is a complex and controversial task, . . . any standard of individual well-being that does not ultimately rest on an individual's own underlying and enduring aims and values is both problematic in theory and subject to intolerable abuse in practice. There may be room in some broad policy decisions or overall theories of justice for more "objective" and interpersonal measures of wellbeing that fail fully to reflect differences in individuals' own views of their well-being,3 but we believe there is much less room for such purportedly objective measures in the kind of judgments of concern here—judgments about appropriate treatment for an individual patient. Thus, a standard that judges competence by comparing the content of a patient's decision to some objective standard for the correct decision may fail even to protect appropriately a patient's well-being. C. A Process Standard of Decision-Making Competence

An adequate standard of competence will focus primarily not on the content of the patient's decision but on the process of the reasoning that leads up to that decision. There are two central questions for any process standard of competence. First, a process standard must set a level of reasoning required for the patient to be

Source: From Deciding for Others: The Ethics of Surrogate Decision Malting (Cambridge University Press: Cambridge, 1990), 48-57. Copyright © 1990 Cambridge University Press. Reprinted with the permission of Cambridge University Press.

Buchanan/Brock: Standards of Competence

competent. In other words, how well must the patient understand and reason to be competent? How much can understanding be limited or reasoning be defective and still be compatible with competence? The second question often passes without explicit notice by those evaluating competence. How certain must those persons evaluating competence be about how well the patient has understood and reasoned in coming to a decision? This second question is important because it is common in cases of marginal or questionable competence for there to be a significant degree of uncertainty about the patient's reasoning and decision-making process that can never be eliminated. II. Relation of the Process Standard of Competence to Expected Harms and Benefits

Because the competence evaluation requires striking a balance between the two values of respecting patients' rights to decide for themselves and protecting them from the harmful consequences of their own choices, it should be clear that no single standard of competence— no single answer to the questions above—can be adequate for all decisions. This is true because (1) the degree of expected harm from choices made at a given level of understanding and reasoning can vary from none to the most serious, including major disability or death, and because (2) the importance or value to the patient of self-determination can vary depending on the choice being made. There is an important implication of this view that the standard of competence ought to vary in part with the expected harms or benefits to the patient of acting in accordance with the patient's choice—namely, that just because a patient is competent to consent to a treatment, it does not follow that the patient is competent to refuse it, and vice versa. For example, consent to a low-risk lifesaving procedure by an otherwise healthy individual should require only a minimal level of competence, but refusal of that same procedure by such an individual should require the highest level of competence. Because the appropriate level of competence properly required for a particular decision must be adjusted to the consequences of acting on that decision, no single standard of decision-making competence is adequate. Instead, the level of competence appropriately required for decision-making varies along a full range from low/minimal to high/maximal. Table 1 illustrates this variation, with the treatment choices listed used only as examples of any treatment choice with that relative risk benefit assessment. The net balance of expected benefits and risks of the patient's choice in comparison with other

alternatives will usually be determined by the physician. This assessment should focus on the expected effects of a particular treatment option in forwarding the patient's underlying and enduring aims and values, to the extent that these are known. When the patient's aims and values are not known, the risk/benefit assessment will balance the expected effects of a particular treatment option in achieving the general goals of health care in prolonging life, preventing injury and disability, and relieving suffering as against its risks of harm. The table indicates that the relevant comparison is with other available alternatives, and the degree to which the net benefit/risk balance of the alternative chosen is better or worse than that for optimal alternative treatment options. It should be noted that a choice might properly require only low/minimal competence, even though its expected risks exceed its expected benefits or it is more generally a high-risk treatment, because all other available alternatives have substantially worse risk/benefit ratios. Table 1 also indicates, for each level of competence, the relative importance of different grounds for believing that a patient's own choice best promotes his or her well-being. This brings out an important point. For all patient choices, other people responsible for deciding whether those choices should be respected should have grounds for believing that the choice, if it is to be honoured, is reasonably in accord with the patient's well-being (although the choice need not, of course, maximally promote the patient's interests). When the patient's level of decision-making competence need be only at the low/minimal level, as in the agreement to a lumbar puncture for presumed meningitis, these grounds derive only minimally from the fact that the patient has chosen the option in question; they principally stem from others' positive assessment of the choice's expected effects on the patient's well-being. At the other extreme, when the expected effects of the patient's choice for his or her well-being appear to be substantially worse than available alternatives, as in the refusal of a simple appendectomy, the requirement of a high/maximal level of competence provides grounds for relying on the patient's decision as itself establishing that the choice best fits the patient's good (his or her own underlying and enduring aims and values). The highest level of competence should assure that no significant mistakes in the patient's reasoning and decision-making are present, and is required to rebut the presumption that the choice is not in fact reasonably related to the patient's interests. When the expected effects for the patient's wellbeing of his or her choice are approximately comparable

33

SW 1

s

II di

91

1111:u

2 Medical Decision-Making

Table 1 Decision-Making Competence and Patient Well-Being

The patient's treatment choice

Others' risk/benefit assessment of that choice in comparison with other alternatives

Patient consents to lumbar puncture for presumed meningitis

Level of decision-making competence required

Grounds for believing patient's choice best promotes/protects own well-being

Net balance substantially better than for possible alternatives

Low/minimal

Principally the benefit/risk assessment made by others

Patient chooses lumpectomy for breast cancer

Net balance roughly comparable to that of other alternatives

Moderate/median

Roughly equally from the benefit/risk assessment made by others and from the patient's decision that the chosen alternative best fits own conception of own good

Patient refuses surgery for simple appendectomy

Net balance substantially worse than for another alternative or alternatives

High/maximal

Principally from patient's decision that the chosen alternative best fits own conception of own good

to those of alternatives, as in the choice of a lumpectomy for treatment of breast cancer, a moderate/median level of competence is sufficient to provide reasonable grounds that the choice promotes the patient's good and that her well-being is adequately protected. It is also reasonable to assume that as the level of competence required increases (from minimal to maximal), the instrumental value or importance of respecting the patient's self-determination increases as well, specifically the part of the value of self-determination that rests on the assumption that persons will secure their good when they choose for themselves. As competence increases, other things being equal, the likelihood of this happening increases. Thus, according to the concept of competence endorsed here, a particular individual's decisionmaking capacity at a given time may be sufficient for making a decision to refuse a diagnostic procedure when forgoing the procedure does not carry a significant risk, although it would not necessarily be sufficient for refusing a surgical procedure that would correct a life-threatening condition. The greater the risk relative to other alternatives—where risk is a function of the severity of the expected harm and the probability of its occurrence—the greater the level of communication, understanding, and reasoning skills required for competence to make that decision. It is not always true, however, that if a person is competent to make one decision, then he or she is competent to make another decision so long as it involves equal risk. Even if the risk is the same, one decision may be more complex, and hence require a higher level of capacity for understanding options and reasoning about consequences. In the previous section, we rejected a standard of competence that looks to the content or outcome of

the decision in favour of a standard that focuses on the process of the patient's reasoning. This may appear inconsistent with our insistence here that the appropriate level of decision-making capacity required for competence should depend in significant part on the effects for the patient's well-being of accepting his or her choice, since what those effects are clearly depends on the content or outcome of the patient's choice. However, there is no inconsistency. The competence evaluation addresses the process of the patient's reasoning, whereas the degree of defectiveness and limitation of, and uncertainty about, that process that is compatible with competence depends in significant part on the likely harm to the patient's well-being of accepting his or her choice. To the extent that they are known, the effects on the patient's well-being should be evaluated in terms of his or her own underlying and enduring aims and values, or, where these are not known, in terms of the effects on life and health. Thus in our approach there is no use of an "objective" standard for the best or correct decision that is known to be in conflict with the patient's own underlying and enduring aims and values, which was the objectionable feature of a content or outcome standard of competence. The evaluation of the patient's decision-making will seek to assess how well the patient has understood the nature of the proposed treatment and any significant alternatives, the expected benefits and risks and the likelihood of each, the reason for the recommendation, and then whether the patient has made a choice that reasonably conforms to his or her underlying and enduring aims and values. Two broad kinds of defect are then possible: first, "factual" misunderstanding about the nature and likelihood of an outcome, for example from limitations in cognitive understanding resulting from stroke or from impairment of short-term memory

Buchanan/Brock: Standards of Competence C

resulting from dementia; second, failure of the patient's choice to be based on his or her underlying and enduring aims and values, for example because depression has temporarily distorted them so that the patient "no longer cares" about restoration of the function he or she had valued before becoming depressed.4 A crude but perhaps helpful way of characterizing the proper aim of the evaluator of the competence of a seemingly harmful or "bad" patient choice is to think of him or her addressing the patient in this fashion: "Help me try to understand and make sense of your choice. Help me to see whether your choice is reasonable, not in the sense that it is what I or most people would choose, but that it is reasonable for you in light of your underlying and enduring aims and values." This is the proper focus of a process standard of competence. Some may object that misguided paternalists will always be ready to assert that their interference with the patient's choice is "deep down" in accord with what we have called the patient's "underlying and

enduring aims and values," or at least with what these would be except for unfortunate distortions. If there is no objective way to determine a person's underlying and enduring aims and values then the worry is that our view will lead to excessive paternalism. We acknowledge that this determination will often be difficult and uncertain, for example in cases like severe chronic depression, leading to genuine and justified uncertainty about the patient's "true" aims and values. But any claims that the aims and values actually expressed by the patient are not his or her underlying and enduring aims and values should be based on evidence of the distortion of the actual aims and values independent of their mere difference with some other, "better" aims and values. Just as the process standard of competence focuses on the process of the patient's reasoning, so also it requires evidence of a process of distortion of the patient's aims and values to justify evaluating choices by a standard other than the patient's actually expressed aims and values. . . .

Notes 1.

2. 3.

See especially Roth, L.H., Meisel, A., and Lidz, C.W., (1977), "Tests of Competency to Consent to Treatment," in American Journal of Psychiatry 134: 279-84; what they call "tests" are what we call "standards." An excellent discussion of competence generally, and of Roth et tests for competence in particular, is Freedman, B., (1981), "Competence, Marginal and Otherwise," in International Journal of Law and Psychiatry 4: 53-72. Cf. Freedman, op. cit. For example, John Rawls makes such claims for an objective and interpersonal account of "primary goods" to

A Relational Approach to Autonomy in Health Care Susan Sherwin

Respect for patient autonomy (or self-direction) is broadly understood as recognition that patients have the authority to make decisions about their own health care. . . . I propose a feminist analysis of autonomy, making vivid both our attraction to and distrust of the dominant interpretation of this concept. I begin by reviewing

4.

be used in evaluating persons' well-being within a theory of justice; cf. Rawls, J. (1971), A Theory of Justice (Harvard University Press: Cambridge, MA). This second kind of decision-making defect illustrates the inadequacy of the tests that Roth, Meisel, and Lidz call "the ability to understand" and "actual understanding" tests (cf. Roth et al., op. cit., 281-2). The clinically depressed patient may evidence no failure to understand the harmful consequences of his choice, but instead evidence indifference to those consequences as result of his depression.

. . . some difficulties I find with the usual interpretations of the concept, focusing especially on difficulties that arise from a specifically feminist perspective. In response to these problems, I propose an alternative conception of autonomy that I label "relational" though the terms socially situated or contextualized would describe it equally well. To avoid confusion, I explicitly distinguish my use of the term relational from that of some other feminist authors, such as Carol Gilligan (1982), who reserve it to refer only to the narrower set of interpersonal relations. I apply the term to the full range of influential human relations, personal and public. Oppression permeates both personal and public

Source: From The Politics of Women's Health: Exploring Agency and Autonomy, by Susan Sherwin et al. Used by permission of Temple University Press. Copyright © 1998 by Temple University. All rights reserved.

II SI

ww

2 Medical Decision-Making

relationships; hence, I prefer to politicize the understanding of the term relational as a way of emphasizing the political dimensions of the multiple relationships that structure an individual's selfhood, rather than to reserve the term to protect a sphere of purely private relationships that may appear to be free of political influence.' I explain why I think the relational alternative is more successful than the familiar individualistic interpretation at addressing the concerns identified. Finally, I briefly indicate some of the implications of adopting a relational interpretation of autonomy with respect to some of the issues discussed elsewhere in this book [The Politics of Women's Health: Exploring Agency and Autonomy, ed. Susan Sherwin (Temple University Press: 1988)], and I identify some of the changes that this notion of relational autonomy suggests for the delivery of health services. . . . Problems with the Autonomy Ideal . . . Despite this broad consensus about the value of a principle of respect for patient autonomy in health care, there are many problems with the principle as it is usually interpreted and applied in health care ethics. As many health critics have observed, we need to question how much control individual patients really have over the determination of their treatment within the stressful world of health care services. Even a casual encounter with most modern hospitals reveals that wide agreement about the moral importance of respect for patient autonomy does not always translate into a set of practices that actually respect and foster patient autonomy in any meaningful sense. Ensuring that patients meet some measure of informed choice—or, more commonly, informed consent2—bef ore receiving or declining treatment has become accepted as the most promising mechanism for ensuring patient autonomy in health care settings, but, in practice, the effectiveness of the actual procedures used to obtain informed consent usually falls short of fully protecting patient autonomy. This gap is easy to understand: attention to patient autonomy can be a time-consuming business and the demands of identifying patient values and preferences are often sacrificed in the face of heavy patient loads and staff shortages. In addition, health care providers are often constrained from promoting and responding to patients' autonomy in health care because of pressures they experience to contain health care costs and to avoid making themselves liable to lawsuits. Moreover, most health care providers are generally not well trained in the communication skills necessary to ensure that patients have the requisite understanding to provide genuine informed consent.

This problem is compounded within our increasingly diverse urban communities, where differences in language and culture between health care providers and the patients they serve may create enormous practical barriers to informed choice. There are yet deeper problems with the ideal of autonomy invoked in most bioethical discussions. The paradigm offered for informed consent is built on a model of articulate, intelligent patients who are accustomed to making decisions about the course of their lives and who possess the resources necessary to allow them a range of options to choose among. Decisions are constructed as a product of objective calculation on the basis of near perfect information. Clearly, not all patients meet these ideal conditions (perhaps none does), yet there are no satisfactory guidelines available about how to proceed when dealing with patients who do not fit the paradigm. Feminist analysis reveals several problems inherent in the very construction of the concept of autonomy that is at the heart of most bioethics discussions.3 One problem is that autonomy provisions are sometimes interpreted as functioning independently of and outweighing all other moral values. More specifically, autonomy is often understood to exist in conflict with the demands of justice because the requirements of the latter may have to be imposed on unwilling citizens. Autonomy is frequently interpreted to mean freedom from interference; this analysis can be invoked (as it frequently is) to oppose taxation as coercive and, hence, a violation of personal autonomy. But coercive measures like taxation are essential if a society wants to reduce inequity and provide the disadvantaged with access to the means (e.g., basic necessities, social respect, education, and health care) that are necessary for meaningful exercise of their autonomy. In contrast to traditional accounts of autonomy that accept and indeed presume some sort of tension between autonomy and justice, feminism encourages us to see the connections between these two central moral ideals. In fact, autonomy language is often used to hide the workings of privilege and to mask the barriers of oppression. For example, within North America it seems that people who were raised in an atmosphere of privilege and respect come rather easily to think of themselves as independent and self-governing; it feels natural to them to conceive of themselves as autonomous. Having been taught that they need only to apply themselves in order to take advantage of the opportunities available to them, most learn to think of their successes as self-created and deserved. Such thinking encourages them to be oblivious to the barriers that oppression and disadvantage create, and it allows

Sherwin: A Relational Approach to Autonomy in Health Care

them to see the failures of others as evidence of the latters' unwillingness to exercise their own presumed autonomy responsibly. This individualistic approach to autonomy makes it very easy for people of privilege to remain ignorant of the social arrangements that support their own sense of independence, such as the institutions that provide them with an exceptionally good education and a relatively high degree of personal safety. Encouraged to focus on their own sense of individual accomplishment, they are inclined to blame less well-situated people for their lack of comparable success rather than to appreciate the costs of oppression. This familiar sort of thinking tends to interfere with people's ability to see the importance of supportive social conditions for fostering autonomous action. By focusing instead on the injustice that is associated with oppression, feminism helps us to recognize that autonomy is best achieved where the social conditions that support it are in place. Hence, it provides us with an alternative perspective for understanding a socially grounded notion of autonomy. Further, the standard conception of autonomy, especially as it is invoked in bioethics, tends to place the focus of concern quite narrowly on particular decisions of individuals; that is, it is common to speak of specific health care decisions as autonomous, or, at least, of the patient as autonomous with respect to the decision at hand. Such analyses discourage attention to the context in which decisions are actually made. Patient decisions are considered to be autonomous if the patient is (1) deemed to be sufficiently competent (rational) to make the decision at issue, (2) makes a (reasonable) choice from a set of available options, (3) has adequate information and understanding about the available choices, and (4) is free from explicit coercion toward (or away from) one of those options. It is assumed that these criteria can be evaluated in any particular case, simply by looking at the state of the patient and her deliberations in isolation from the social conditions that structure her options. Yet, each of these conditions is more problematic than is generally recognized. The competency criterion threatens to exclude people who are oppressed from the scope of autonomy provisions altogether. This is because competency is often equated with being rational,4 yet the rationality of women and members of other oppressed groups is frequently denied. In fact, as Genevieve Lloyd (1984) has shown, the very concept of rationality has been constructed in opposition to the traits that are stereotypically assigned to women (e.g., by requiring that agents demonstrate objectivity and emotional distance),5 with the result that women are often seen as simply incapable of rationality.6 Similar problems arise

with respect to stereotypical assumptions about members of racial minorities, indigenous peoples, persons with disabilities, welfare recipients, people from developing countries, those who are non-literate, and so on. Minimally, then, health care providers must become sensitive to the ways in which oppressive stereotypes can undermine their ability to recognize some sorts of patients as being rational or competent. Consider, also, the second condition, which has to do with making a (reasonable) choice from the set of available options. Here, the difficulty is that the set of available options is constructed in ways that may already seriously limit the patient's autonomy by prematurely excluding options the patient might have preferred. There is a whole series of complex decisions that together shape the set of options that health care providers are able to offer their patients: these can involve such factors as the forces that structure research programs, the types of results that journals are willing to publish, curriculum priorities in medical and other professional schools, and funding policies within the health care system.' While all patients will face limited choices by virtue of these sorts of institutional policy decisions, the consequences are especially significant for members of oppressed groups because they tend to be underrepresented on the bodies that make these earlier decisions, and therefore their interests are less likely to be reflected in each of the background decisions that are made. In general, the sorts of institutional decisions in question tend to reflect the biases of discriminatory values and practices. Hence, the outcomes of these multiple earlier decisions can have a significant impact on an oppressed patient's ultimate autonomy by disproportionately and unfairly restricting the choices available to her. Nevertheless, such background conditions are seldom visible within discussions of patient autonomy in bioethics. The third condition is also problematic in that the information made available to patients is, inevitably, the information that has been deemed worthy of study and that is considered relevant by the health care providers involved. Again, research, publication, and education policies largely determine what sorts of data are collected and, significantly, what questions are neglected; systemic bias unquestionably influences these policies. Further, the very large gap in life experience between physicians, who are, by virtue of their professional status, relatively privileged members of society, and some of their seriously disadvantaged patients makes the likelihood of the former anticipating the specific information needs of the latter questionable. While an open consent process will help reduce this gap by providing patients with the opportunity to raise

37

II w

I11111

2 Medical Decision-Making

questions, patients often feel too intimidated to ask or even formulate questions, especially when they feel socially and intellectually inferior to their physicians and when the physicians project an image of being busy with more important demands. Often, one needs some information in order to know what further questions to ask, and large gaps in perspective between patients and their health care providers may result in a breakdown in communication because of false assumptions by either participant. The fourth condition, the one that demands freedom from coercion in exercising choice, is extremely difficult to evaluate when the individual in question is oppressed. The task becomes even trickier if the choice is in a sphere that is tied to her oppression. The condition of being oppressed can be so fundamentally restrictive that it is distorting to describe as autonomous some specific choices made under such conditions. For example, many women believe they have no real choice but to seek expensive, risky cosmetic surgery because they accurately perceive that their opportunities for success in work or love depend on their more closely approximating some externally defined standard of beauty. Similar sorts of questions arise with respect to some women's choice of dangerous, unproven experiments in new reproductive technologies because continued childlessness can be expected to have devastating consequences for their lives. In other cases, women sometimes choose to have abortions because they fear that giving birth will involve them in unwanted and lifelong relationships with abusive partners. Some women have little access to contraceptives and find themselves choosing sterilization as the most effective way of resisting immediate demands of their partners even if they might want more children in the future. Or, some women seek out prenatal diagnosis and selective abortion of cherished fetuses because they realize that they cannot afford to raise a child born with a serious disability, though they would value such a child themselves. Many middle-class Western women choose hormone replacement therapy at menopause because they recognize that their social and economic lives may be threatened if they appear to be aging too quickly. When a woman's sense of herself and her range of opportunities have been oppressively constructed in ways that (seem to) leave her little choice but to pursue all available options in the pursuit of beauty or childbearing or when she is raised in a culture that ties her own sense of herself to external norms of physical appearance or fulfillment associated with childbearing or, conversely, when having a(nother) child will impose unjust and intolerable costs on her, it does not seem sufficient to restrict our analysis to the degree

of autonomy associated with her immediate decision about a particular treatment offered. We need a way of acknowledging how oppressive circumstances can interfere with autonomy, but this is not easily captured in traditional accounts. Finally, there are good reasons to be wary of the ways in which the appearance of choice is used to mask the normalizing powers of medicine and other healthrelated institutions. As Michel Foucault (1979, 1980) suggests, in modern societies the illusion of choice can be part of the mechanism for controlling behaviour. Indeed, it is possible that bioethical efforts to guarantee the exercise of individual informed choice may actually make the exercise of medical authority even more powerful and effective than it would be under more traditionally paternalistic models. In practice, the ideal of informed choice amounts to assuring patients of the opportunity to consent to one of a limited list of relatively similar, medically encouraged procedures. Thus, informed consent procedures aimed simply at protecting autonomy in the narrow sense of specific choice among pre-selected options may ultimately serve to secure the compliance of docile patients who operate under the illusion of autonomy by virtue of being invited to consent to procedures they are socially encouraged to choose. Unless we find a way of identifying a deeper sense of autonomy than that associated with the expression of individual preference in selecting among a limited set of similar options, we run the risk of struggling to protect not patient autonomy but the very mechanisms that ensure compliant medical consumers, preoccupied with the task of selecting among a narrow range of treatments.

Focus on the Individual A striking feature o f most bioethical discussions about patient autonomy is their exclusive focus on individual patients; this pattern mirrors medicine's consistent tendency to approach illness as primarily a problem of particular patients.8 Similar problems are associated with each discipline. Within the medical tradition, suffering is located and addressed in the individuals who experience it rather than in the social arrangements that may be responsible for causing the problem. Instead of exploring the cultural context that tolerates and even supports practices such as war, pollution, sexual violence, and systemic unemployment—practices that contribute to much of the illness that occupies modern medicine— physicians generally respond to the symptoms troubling particular patients in isolation from the context that produces these conditions. Apart from populationbased epidemiological studies (which, typically, restrict

Sherwin: A Relational Approach to Autonomy in Health Care

their focus to a narrow range of patterns of illness and often exclude or distort important social dimensions), medicine is primarily oriented toward dealing with individuals who have become ill (or pregnant, [in] fertile, or menopausal). This orientation directs the vast majority of research money and expertise toward the things that can be done to change the individual, but it often ignores key elements at the source of the problems. For example, physicians tend to respond to infertility either by trivializing the problem and telling women to go home and "relax," or by prescribing hormonal and surgical treatment of particular women, rather than by demanding that research and public health efforts be aimed at preventing pelvic inflammatory disease, which causes many cases of infertility, or by encouraging wide public debate (or private reflections) on the powerful social pressures to reproduce that are directed at women. In similar fashion, the mainstream scientific and medical communities respond to the growth of breast cancer rates by promoting individual responsibility for self-examination and by searching for the gene(s) that makes some women particularly susceptible to the disease; when it is found in a patient, the principal medical therapy available is to perform "prophylactic" double mastectomies. Few physicians demand examination of the potential contributory role played by the use of pesticides or chlorine, or the practice of feeding artificial hormones to agricultural animals. Or they deal with dramatically increased skin cancer rates by promoting the personal use of sunscreens while resigning themselves to the continued depletion of the ozone layer. In another area, health care professionals generally deal with the devastating effects of domestic violence by patching up its victims, providing them with medications to relieve depression and advice to move out of their homes, and devising pathological names for victims who stay in violent relationships ("battered woman syndrome" and "self-defeating personality disorder"), but few actively challenge the sexism that accepts male violence as a "natural" response to frustration and fears of abandonment. Some qualifications are in order. Clearly these are crude and imprecise generalizations. They describe a general orientation of current health practices, but they certainly do not capture the work of all those involved in medical research and practice. Fortunately, there are practitioners and researchers engaged in the very sorts of investigation I call for, but they are exceptional, not typical. Moreover, I do not want to imply that medicine should simply abandon its concern with treating disease in individuals. I understand that prevention strategies will not eliminate all illness and I believe that

personalized health care must continue to be made available to those who become ill. Further, I want to be clear that my critique does not imply that physicians or other direct care providers are necessarily the ones who ought to be assuming the task of identifying the social and environmental causes of disease. Health care training, and especially the training of physicians, is directed at developing the requisite skills for the extremely important work of caring for individuals who become ill. The responsibility for investigating the social causes of illness and for changing hazardous conditions is a social one that is probably best met by those who undertake different sorts of training and study. The problem is that medicine, despite the limits of its expertise and focus, is the primary agent of health care activity in our society, and physicians are granted significant social authority to be the arbiters of health policy. Hence, when medicine makes the treatment of individuals its primary focus, we must understand that important gaps are created in our society's ability to understand and promote good health. In parallel fashion, autonomy-focused bioethics concentrates its practitioners' attention on the preferences of particular patients, and it is, thereby, complicit in the individualistic orientation of medicine. It asks health care providers to ensure that individual patients have the information they need to make rational decisions about their health care, yet it does not ask the necessary questions about the circumstances in which such decisions are made. The emphasis most bioethicists place on traditional, individualistic understandings of autonomy reinforces the tendency of health care providers and ethicists to neglect exploration of the deep social causes and conditions that contribute to health and illness. Moreover, it encourages patients to see their own health care decisions in isolation from those of anyone else, thereby increasing their sense of vulnerability and dependence on medical authority. The narrow individual focus that characterizes the central traditions within both medicine and bioethics obscures our need to consider questions of power, dominance, and privilege in our interpretations and responses to illness and other health-related matters as well as in our interpretations of the ideal of autonomy. These ways of structuring thought and practice make it difficult to see the political dimensions of illness, and, in a parallel way, they obscure the political dimensions of the conventional criteria for autonomous deliberation. As a result, they interfere with our ability to identify and pursue more effective health practices while helping to foster a social environment that ignores and tolerates oppression. In both cases, a broader political perspective is necessary if we are to avoid the problems

39

II aIII

40

111111

2 Medical Decision-Making

created by restricting our focus to individuals apart from their location. Feminism offers just such a broader perspective. In contrast to the standard approaches in bioethics, feminism raises questions about the social basis for decisions about health and health care at all levels. Here, as elsewhere, feminists are inclined to ask whose interests are served and whose are harmed by the traditional ways of structuring thought and practice. By asking these questions, we are able to see how assumptions of individual-based medicine help to preserve the social and political status quo. For example, the current taxonomy in Canada designates certain sorts of conditions (e.g., infertility, cancer, heart disease, anxiety) as appropriate for medical intervention, and it provides grounds for ensuring that such needs are met. At the same time, it views other sorts of conditions (e.g., malnutrition, fear of assault, low self-esteem) as falling beyond the purview of the health care system and, therefore, as ineligible to draw on the considerable resources allocated to the delivery of health services.9 In this way, individualistic assumptions support a system that provides expert care for many of the health complaints of those with greatest financial privilege while dismissing as outside the scope of health care many of the sources of illness that primarily affect the disadvantaged. A more social vision of health would require us to investigate ways in which non-medical strategies, such as improving social and material conditions for disadvantaged groups, can affect the health status of different segments of the community.1° None of the concerns I have identified argues against maintaining a strong commitment to autonomy in bioethical deliberations. In fact, I have no wish to abandon this ideal (just as I have no desire to abandon patient-centred medical care). I still believe that a principle of respect for patient autonomy is an important element of good patient care. Moreover, I believe that appeal to a principle of respect for autonomy can be an important instrument in challenging oppression, and it can actually serve as the basis for many of the feminist criticisms I present with respect to our current health care system. What these criticisms do suggest, however, is that we must pursue a more careful and politically sensitive interpretation of the range of possible restrictions on autonomy than is found in most of the non-feminist bioethics literature. We need to be able to look at specific decisions as well as the context that influences and sometimes limits such decisions. Many of the troublesome examples I review above are entirely compatible with traditional conceptions of autonomy even though the patients in question may be facing

unjust barriers to care or may be acting in response to oppressive circumstances; traditional conceptions are inadequate to the extent that they make invisible the oppression that structures such decisions. By focusing only on the moment of medical decision-making, traditional views fail to examine how specific decisions are embedded within a complex set of relations and policies that constrain (or, ideally, promote) an individual's ability to exercise autonomy with respect to any particular choice. To understand this puzzle it is necessary to distinguish between agency and autonomy. To exercise agency, one need only exercise reasonable choice." The women who choose some of the controversial practices discussed (e.g., abortion to avoid contact with an abusive partner, cosmetic surgery to conform to artificial norms of beauty, use of dangerous forms of reproductive technology) are exercising agency, clearly they are making choices, and, often, those choices are rational under the circumstances. They also meet the demands of conventional notions of autonomy that ask only that anyone contemplating such procedures be competent, or capable of choosing (wisely), have available information current practice deems relevant, and be free of direct coercion. But insofar as their behaviour accepts and adapts to oppression, describing it as autonomous seems inadequate. Together, the habits of equating agency (the making of a choice) with autonomy (selfgovernance) and accepting as given the prevailing social arrangements have the effect of helping to perpetuate oppression: when we limit our analysis to the quality of an individual's choice under existing conditions (or when we fail to inquire why some people do not even seek health services), we ignore the significance of oppressive conditions. Minimally, autonomous persons should be able to resist oppression—not just act in compliance with it—and be able to refuse the choices oppression seems to make nearly irresistible. Ideally, they should be able to escape from the structures of oppression altogether and create new options that are not defined by these structures either positively or negatively. In order to ensure that we recognize and address the restrictions that oppression places on people's health choices, then, we need a wider notion of autonomy that will allow us to distinguish genuinely autonomous behaviour from acts of merely rational agency. This conception must provide room to challenge the quality of an agent's specific decision-making ability and the social norms that encourage agents to participate in practices that may be partially constitutive of their oppression.'2 A richer, more politically sensitive standard of autonomy should make visible the impact

Sherwin: A Relational Approach to Autonomy in Health Care 41

of oppression on a person's choices as well as on her very ability to exercise autonomy fully. Such a conception has the advantage of allowing us to avoid the trap of focusing on the supposed flaws of the individual who is choosing under oppressive circumstances (e.g., by dismissing her choices as "false consciousness"), for it is able to recognize that such choices can be reasonable for the agent. Instead, it directs our attention to the conditions that shape the agent's choice and it makes those conditions the basis of critical analysis. The problems that I identify with the conventional interpretation of patient autonomy reveal a need to expand our understanding of the types of forces that interfere with a patient's autonomy. On non-feminist accounts, these are irrationality, failure to recognize that a choice is called for, lack of necessary information, and coercion (including psychological compulsion). Since each of these conditions must be reinterpreted to allow for the ways in which oppression may be operating, we must add to this list recognition of the costs and effects of oppression and of the particular ways in which oppression is manifested. But we must do more than simply modify our interpretation of the four criteria reviewed above. We also need an understanding of the ways in which a person can be encouraged to develop (or discouraged from developing) the ability to exercise autonomy. For this task, we need to consider the presence or absence of meaningful opportunities to build the skills required to be able to exercise autonomy well (Meyers 1989), including the existence of appropriate material and social conditions. In addition, our account should reflect the fact that many decisionmakers, especially women, place the interests of others at the centre of their deliberations. Such an analysis will allow us to ensure that autonomy standards reflect not only the quality of reasoning displayed by a patient at the moment of medical decision-making but also the circumstances that surround this decision-making. A Relational Alternative A major reason for many of the problems identified with the autonomy ideal is that the term is commonly understood to represent freedom of action for agents who are paradigmatically regarded as independent, self-interested, and self-sufficient. As such, it is part of a larger North American cultural ideal of competitive individualism in which every citizen is to be left "free" to negotiate "his" way through the complex interactions of social, economic, and political life." The feminist literature is filled with criticism of such models of agency and autonomy. for example, many feminists object that this ideal appeals to a model of personhood that is

distorting because, in fact, no one is fully independent. As well, they observe that this model is exclusionary because those who are most obviously dependent on others (e.g., because of disability or financial need) seem to be disqualified from consideration in ways that others are not. Many feminists object that the view of individuals as isolated social units is not only false but impoverished: much of who we are and what we value is rooted in our relationships and affinities with others. Also, many feminists take issue with the common assumption that agents are single-mindedly self-interested, when so much of our experience is devoted to building or maintaining personal relationships and communities.14 If we are to effectively address these concerns, we need to move away from the familiar Western understanding of autonomy as self-defining, self-interested, and self-protecting, as if the self were simply some special kind of property to be preserved." Under most popular interpretations, the structure of the autonomy—heteronomy framework (governance by self or by others) is predicated on a certain view of persons and society in which the individual is thought to be somehow separate from and to exist independently of the larger society, each person's major concern is to be protected from the demands and encroachment of others. This sort of conception fails to account for the complexity of the relations that exist between persons and their culture. It idealizes decisions that are free from outside influence without acknowledging that all persons are, to a significant degree, socially constructed, that their identities, values, concepts, and perceptions are, in large measure, products of their social environment. Since notions of the self are at the heart of autonomy discussions, alternative interpretations of autonomy must begin with an alternative conception of the self. Curiously, despite its focus on individuals, standard interpretations of autonomy have tended to think of selves as generic rather than distinctive beings. In the traditional view, individuals tend to be treated as interchangeable in that no attention is paid to the details of personal experience. Hence, there is no space within standard conceptions to accommodate important differences among agents, especially the effects that oppression (or social privilege) has on a person's ability to exercise autonomy. In order to capture these kinds of social concerns, some feminists have proposed turning to a relational conception of personhood that recognizes the importance of social forces in shaping each person's identity, development, and aspirations.m Following this suggestion, I now explore a relational interpretation of autonomy that is built around a relational

II 111

e

11111I

2 Medical Decision-Making

conception of the self that is explicitly feminist in its conception. Under relational theory, selfhood is seen as an ongoing process, rather than as something static or fixed. Relational selves are inherently social beings that are significantly shaped and modified within a web of interconnected (and sometimes conflicting) relationships. Individuals engage in the activities that are constitutive of identity and autonomy (e.g., defining, questioning, revising, and pursuing projects) within a configuration of relationships, both interpersonal and political. By including attention to political relationships of power and powerlessness, this interpretation of relational theory provides room to recognize how the forces of oppression can interfere with an individual's ability to exercise autonomy by undermining her sense of herself as an autonomous agent and by depriving her of opportunities to exercise autonomy. Thus, it is able to provide us with insight into why it is that oppressed people often seem less autonomous than others even when offered a comparable range of choices. Under a relational view, autonomy is best understood to be a capacity or skill that is developed (and constrained) by social circumstances. It is exercised within relationships and social structures that jointly help to shape the individual while also affecting others' responses to her efforts at autonomy. Diana Meyers (1989) has developed one such theory of personal autonomy. She argues that autonomy involves a particular competency that requires the development of specific skills. As such, it can be either enhanced or diminished by the sort of socialization the agent experiences. Meyers shows how the specific gender socialization most (Western) women undergo trains them in social docility and rewards them for defining their interests in terms of others, thereby robbing them of the opportunity to develop the essential capacity of self-direction Such training relegates most women to a category she labels "minimally autonomous" (as distinct from her more desirable categories of medially autonomous and fully autonomous). Relational theory allows us to appreciate how each relationship a person participates in plays a role in fostering or inhibiting that individual's capacity for autonomous action by encouraging or restricting her opportunities to understand herself as an autonomous agent and to practice exercising the requisite skills. Such a conception makes clear the importance of discovering the ways in which oppression often reduces a person's ability to develop and exercise the skills that are necessary for achieving a reasonable degree of autonomy. For instance, relational theory allows us to see the damaging effects on autonomy of internalized

oppression. Feminists have long understood that one of the most insidious features of oppression is its tendency to become internalized in the minds of its victims. This is because internalized oppression diminishes the capacity of its victims to develop self-respect, and, as several feminists have argued, reduced (or compromised) self-respect undermines autonomy by undermining the individual's sense of herself as capable of making independent judgments (Meyers 1989; Dillon 1992; Benson 1991, 1994). Moreover, as Susan Babbitt (1993, 1996) has argued, these oppression-induced barriers to autonomy cannot necessarily be rectified simply by providing those affected with more information or by removing explicit coercive forces (as the traditional view assumes). When the messages of reduced selfworth are internalized, agents tend to lose the ability even to know their own objective interests. According to Babbitt, in such cases transformative experiences can be far more important to autonomy than access to alternative information. Feminist theory suggests, then, that women and members of other oppressed groups can be helped to increase their autonomy skills by being offered more opportunities to exercise those skills and a supportive climate for practising them (Meyers 1989), by being provided with the opportunity to develop stronger senses of self-esteem (Benson 1994; Dillon 1992; Meyers 1989), by having the opportunity for transformative experiences that make visible the forces of oppression (Babbitt 1993, 1996), and by having experiences of making choices that are not influenced by the wishes of those who dominate them (Babbitt 1993, 1996). Autonomy requires more than the effective exercise of personal resources and skills, however, generally, it also demands that appropriate structural conditions be met. Relational theory reminds us that material restrictions, including very restricted economic resources, ongoing fear of assault, and lack of educational opportunity (i.e., the sorts of circumstances that are often part of the condition of being oppressed), constitute real limitations on the options available to the agent. Moreover, it helps us to see how socially constructed stereotypes can reduce both society's and the agent's sense of that person's ability to act autonomously. Relational theory allows us to recognize how such diminished expectations readily become translated into diminished capacities. The relational interpretation I favour is feminist in that it takes into account the impact of social and political structures, especially sexism and other forms of oppression, on the lives and opportunities of individuals. It acknowledges that the presence or absence of a degree of autonomy is not just a matter of being

Sherwin: A Relational Approach to Autonomy in Health Care

offered a choice. It also requires that the person have had the opportunity to develop the skills necessary for making the type of choice in question, the experience of being respected in her decisions, and encouragement to reflect on her own values. The society, not just the agent, is subject to critical scrutiny under the rubric of relational autonomy. It is important, however, to avoid an account that denies any scope for autonomy on the part of those who are oppressed. Such a conclusion would be dangerous, since the widespread perception of limited autonomy can easily become a self-fulfilling prophecy. Moreover, such a conclusion would be false. Many members of oppressed groups do manage to develop autonomy skills and, thus, are able to act autonomously in a wide variety of situations, though the particular demands of acting autonomously under oppression are easily overlooked (Benson 1991). Some feminists, such as bell hooks (1990) and Sarah Hoagland (1992), have observed that the marginality associated with being oppressed can sometimes provide people with better opportunities than are available to more well-situated citizens for questioning social norms and devising their own patterns of resistance to social convention. Because those who are especially marginalized (e.g., those who are multiply oppressed or who are "deviant" with respect to important social norms) may have no significant social privilege to lose, they are, sometimes, freer than others to demand changes in the status quo. They may be far more likely to engage in resistance to the norms of oppression than are those who derive some personal benefits from oppressive structures (e.g., middle-class, able-bodied, married women). Still, we must not make the mistake of romanticizing the opportunities available to the oppressed. An adequate conception of autonomy should afford individuals more than the opportunity to resist oppression; it should also ensure that they have opportunities to actively shape their world. A relational conception of autonomy seems better suited than the traditional models to handle the complexities of such paradoxes because it encourages us to attend to the complex ways in which the detailed circumstances of an individual social and political circumstances can affect her ability to act in different kinds of contexts. When relational autonomy reveals the disadvantage associated with oppression in terms of autonomy, the response should not be that others are thereby licensed to make decisions for those who are oppressed; this response would only increase their powerlessness. Rather, it demands attention to ways in which oppressed people can be helped to develop

the requisite autonomy skills. The best way of course to help oppressed people to develop autonomy skills is to remove the conditions of their oppression. Short of that, long-term social projects can help to provide educational opportunities to counter the psychological burdens of oppression. In the short term, it may be necessary to spend more time than usual in supporting patients in the deliberative process of decision-making and providing them with access to relevant political as well as medical information when they contemplate controversial procedures (e.g., information about the social dimensions of hormone replacement therapy). Relational autonomy is not only about changing the individual, however. It also demands attention to ways in which the range of choices before those who belong to oppressed groups can be modified to include more non-oppressive options, that is, options that will not further entrench their existing oppression (as often happens, for example, when women choose cosmetic surgery or the use of many reproductive technologies). Whereas in traditional autonomy theory only the mode and quality of specific decisions are evaluated, feminist relational autonomy regards the range and nature of available and acceptable options as being at least as important as the quality of specific decision-making. Only when we understand the ways in which oppression can infect the background or baseline conditions under which choices are to be made will we be able to modify those conditions and work toward the possibility of greater autonomy by promoting non-oppressive alternatives. As in health matters, it is important in relational discussions not to lose sight of the need to continue to maintain some focus on the individual. Relational autonomy redefines autonomy as the social project it is, but it does not deny that autonomy ultimately resides in individuals. Our attention to social and political contexts helps deepen and enrich the narrow and impoverished view of autonomy available under individualistic conceptions, but it does not support wholesale neglect of the needs and interests of individuals in favour of broader social and political interests. Rather, it can be seen as democratizing access to autonomy by helping to identify and remove the effects of barriers to autonomy that are created by oppression. A relational approach can help to move autonomy from the largely exclusive preserve of the socially privileged and see that it is combined with a commitment to social justice in order to ensure that oppression is not allowed to continue simply because its victims have been deprived of the resources necessary to exercise the autonomy required to challenge it. . . .

II III

O

uuJJ

2 Medical Decision-Making

Notes 1.

Some Network members prefer the terms "contextual" or "situated" as a way of avoiding all confusion with those feminists who reserve the term "relational" to refer exclusively to interpersonal relations. I feel that this usage perpetuates the misleading sense that interpersonal relations are themselves "apolitical." I have, therefore, chosen to insist on a thoroughly political reading of the term "relational" that applies to both interpersonal and more public sorts of relations. 2. Informed choice suggests a wider scope for patient autonomy than informed consent in that it includes the possibility of patients' initiating treatment suggestions, where informed consent implies that the role of the patient is merely to consent to the treatment proposed by the physician; further, informed choice makes more explicit that patients ought also to be free to refuse recommended treatments as well as to accept them. 3. Many of these concerns are not exclusive to feminists; several have also been raised by other sorts of critics. I call them feminist because I came to these concerns through a feminist analysis that attends to the role in society of systems of dominance and oppression, especially those connected with gender. 4. This reduction may be a result of a tendency to collapse the ideal of personal autonomy central to bioethics discussions with the concept of moral autonomy developed by Immanuel Kant. 5. It is often taken as a truism in our culture that emotional involvement constitutes irrationality, that emotions are direct threats to rationality. It is hard to see, however, how decisions about important life decisions are improved if they are made without any emotional attachment to the outcomes. 6. Susan Babbitt (1996) argues that the traditional conception of rationality is defined in terms of propositional understanding in ways that obscure the experiences and needs of opprecced people. 7. For example, research priorities have led to the situation where birth control pills are available only for women, and this increases the pressure on women seeking temporary protection against pregnancy to take the pill even when it endangers their health.

8. I focus primarily on medicine since it is the dominant health profession and is responsible for organization of most health services in developed countries. Most health professions involve a similar bias toward treatment of individuals, though some (e.g., social work) pride themselves on attending to social structures as well as individual need, and most health professions, including medicine, include subspecialties concerned with matters of public health. 9. Because health care is a provincial responsibility, there are differences in the precise services offered from province to province and from one administration to the next within provinces The examples here are broad generalizations. 10. Such considerations do play a role in health care planning at a governmental level where the focus shifts from medical interventions to the idea of health determinants, but here, too, there is excessive attention paid to what the individual can and should be doing ("healthism") and insufficient concern about promoting egalitarian social conditions. 11. The language of agency and autonomy is quite varied within feminist (and other) discourse For example, the term agency is used throughout the collection Provoking Agents: Gender and Agency in Theory and Practice (Gardiner 1995) in ways that sometimes appear to overlap with my usage of relational autonomy. Susan Babbitt (1996), on the other hand, seems to use the two terms in ways analogous to the use here. 12. In addition, we need the conceptual space to be able to acknowledge that restrictive definitions of health sometimes preempt autonomy analysis by limiting the opportunity of some people even to enter the relatively well-funded health care system for assistance with problems (e.g., poverty) that affect their health. 13. The agent imagined in such cases is always stereotypically masculine 14. Feminist discussion of these and other critiques can be found in Gilligan 1982; Baier 1985; Code 1991; and Held 1993. 15. See Nedelsky 1989 for discussion of this view and its limitations. 16. For example, Baier 1985; Code 1991; and Held 1993.

References Babbitt, Susan. 1993. "Feminism and Objective Interests," in Feminist Epistemologies, ed. Linda Alcoff and Elizabeth Potter. Routledge: New York. ---. 1996. Impossible Dreams: Rationality, Integrity, and Moral Imagination. Westview Press: Boulder, CO. Baier, Annette. 1985. "What Do Women Want in a Moral Theory?," in Nous 19 (1): 53-63. Benson, Paul. 1991. "Autonomy and Oppressive Socialization," in Social Theory and Practice 17 (3): 385-408.

1994. "Free Agency and Self-Worth," in Journal of Philosophy 91 (12): 650-68. Code, Lorraine. 1991. What Can She Know? Feminist Theory and the Construction of Knowledge. Cornell University Press: Ithaca, NY. Dillon, Robin. 1992. "Toward a Feminist Conception of SelfRespect," in Hypatia 7 (1): 52-69. Foucault, Michel. 1979. Discipline and Punish. Vintage: New York.

Sherwin: A Relational Approach to Autonomy in Health Care 0

—. 1980. Power/Knowledge, ed. Colin Gordon. Harvester: Brighton, England. Gardiner, Judith Regan. 1995. Provoking Agents: Gender and Agency in Theory and Practice. University of Illinois Press: Chicago. Gilligan, Carol. 1982. In a Different Voice: Psychological Theory and Women's Moral Development. Harvard University Press: Cambridge, MA. Held, Virginia 1993. Feminist Morality: Transforming Culture, Society, and Politics. University of Chicago Press: Chicago. Hoagland, Sarah Lucia 1992. "Lesbian Ethics and Female Agency," in Explorations in Feminist Ethics: Theory and

Practice, ed. Susan Browning Cole and Susan CoultrapMcQuin. Indiana University Press: Bloomington. hooks, bell. 1990. Yearning: Race, Gender; and Cultural Politics. Between the Lines: Toronto. Lloyd, Genevieve. 1984. The Man of Reason: "Male" and "Female" in Western Philosophy. University of Minnesota Press: Minneapolis. Meyers, Diana T. 1989. Self, Society, and Personal Choice. Columbia University Press: New York. Nedelsky, Jennifer. 1989. "Reconceiving Autonomy," in Yale Journal of Law and Feminism 1 (1): 7-36.

2.3 Informed Consent

The Nuts and Bolts of Obtaining Consent to Treatment LE. and F.A. Rozovsky Introduction

Consent to treatment is one o f the most hotly discussed medicolegal issues in medicine. Physicians complain that courts and legislative assemblies have imposed undue burdens upon them in dealing with their patients. Many physicians are fearful of litigation for failing to meet appropriate legal standards in obtaining consent to treatment. The law of consent need not be looked upon as a burdensome task nor as an unwarranted invasion of the physician—patient relationship. Moreover, physicians need not be unduly alarmed about the prospect of consent lawsuits. Although there have been many court cases involving consent issues, it is becoming dear that plaintiffs must overcome significant hurdles to be successful. When consent is obtained properly physicians need not concern themselves with such litigation. Obtaining a proper authorization for treatment not only safeguards the rights of the patient, it forms the basis for a solid defence against consent lawsuits. The key is for doctors to know the necessary steps in obtaining consent. Consent Is a Process Not a Form

When a patient signs a consent form many assume that the document constitutes consent. This is not correct,

even in those provinces in which legislation or regulations focus on so-called signed, written consents. Consent is actually a process that culminates in the parties agreeing to the permissible scope of a diagnostic, medical, or surgical procedure. As noted in the accompanying illustration, there are several steps to the consent process. Some of these are "mechanical" in nature while others involve the legal requirements for obtaining consent. The consent process can be broken down into seven different categories of legal concern: 1. 2. 3. 4. 5. 6. 7.

Voluntary Atmosphere for Consent; Legal Capacity to Give Consent; Mental Competency to Give Consent; Adequate Disclosure o f Information; Candid Replies to Patient's Questions; Consent Must Be Specific to Procedure; and Documentation of Consent.

Getting to know the patient requires the physician to obtain a thorough personal and family history regarding such matters as prior illnesses, known or suspected allergies to foods or medications, current medications taken by the patient, etc. This enables the physician to tailor the consent process to the specific patient. The patient should not be viewed in a vacuum; rather, the totality of the patient's surroundings must be taken into account including his lifestyle, economic and financial status, the probability of success in the procedure, and whether it is an elective or an imperative procedure. What a reasonable person would want to know is measured in terms of what is considered significant or

Source From Econ Medical Professional Advisory Program, Bulletin No. 3, 1985, pp. 1-3. Reprinted with permission.

®

2 Medical Decision-Making

ESTABLISH

DETERMINE

Voluntary atmosphere

Assess mental

Establish relationship by getting to know patient

competency to determine if patient can give consent

Legal capacity

DISCLOSE

DEFINE

Risks, benefits, reasonable alternatives Adequate disclosure of information by answering patient's questions

With the patient

-70

permissible scope of the procedure to be performed

AUTHORIZE & DOCUMENT Express, implied; verbally or in writing Consistent style of documentation

VALID CONSENT

Figure 1 The Consent Process

material information in his decision-making process. This would include: 1. The nature and purpose o f the proposed procedure; 2. Probable risks and benefits associated with the procedure, including the probability of serious injury or death; 3. Reasonable alternative medical or surgical procedures and risks associated with them; 4. The risks of foregoing all diagnostic procedures or treatment; 5. The length of time, if any, the patient can postpone the procedure and the probable risks involved in doing so; 6. The probable length of time that the patient will be hospitalized and unable to resume his employment and regular activities; 7. The probable need for follow-up care and additional surgery, radiation treatment, or chemotherapy; and 8. The right of the patient to withdraw consent. It is clear that in the treatment context the courts do not expect clinicians to disclose ALL known risks and benefits. The prospect of remote risks need not be revealed. The focus is on what the average, reasonable, and prudent person would want to know in the circumstances. To be sure, the circumstances can and do change the scope of the disclosure. If the procedure is novel or experimental there should be a greater degree of information imparted to the patient. Furthermore, clinicians need not disclose information about facts that are readily known to the lay public. Taking aspirin to relieve pain, a high temperature is indicative of infection, or the like is the type of information the clinician may generally assume the public understands. There are exceptions, however, as in the case of patients from less developed nations or different cultures who cannot be expected to appreciate what is "common knowledge."

Candid Responses to Patient Inquiries

Doctors should respond in a candid manner to patients' questions. Inquiries made by patients may be indicative of a potential lapse in the consent process. It may mean that a patient has not completely understood some aspect of the information, or it may signify a new concern. Whatever the reason, questions posed by patients should be addressed in a straightforward manner. Authorization for the Procedure to Be Forfeited

The consent of the patient must be specific to the procedure to be performed. An authorization for treatment is not an open-ended invitation to doctors to help themselves. The authorization obtained at the conclusion of the consent process should be highly specific, outlining what the doctor and staff are and are not permitted to do. Some flexibility, however, can be built into the authorization in terms of the permissible source of the procedure. For example, in an exploratory operation of the abdomen, it may be difficult to define the extent of the procedure. The surgeon can reach an agreement with the patient regarding the taking of tissue specimens for immediate pathology assessment, the removal of organs, the resection of bowel, the performance of a colostomy or an ileostomy. Under no circumstances should the physician extend the scope of the procedure beyond its permissible limits. The only exception to this general rule is if an unanticipated problem occurs requiring an immediate response. Even at this stage, however, the measures undertaken must be reasonable in the context of the problem. If, for example, in the course of unrelated surgery the surgeon discovered a bleeding ulcer that requires prompt surgical attention, he could lawfully extend the surgery to correct the newly discovered problem. However, if the problem that is detected does not require an

Rozovsky: The Nuts and Bolts of Obtaining Consent to Treatment

immediate response, the physician cannot forge ahead and treat it. He must first obtain either an authorization from the patient or his legal representative to extend the procedure which was consented to originally. Documentation of Consent

Unless provincial law specifically requires it, the patient's consent to treatment need not be given in writing. Consent may be expressed or implied, verbally or in writing. However, some form of documentation is necessary for purposes of legal defence. The documentation used may be a consent form filed in the patient's record or a detailed note written in the patient's chart outlining the dialogue between the doctor and the patient. Consent should be documented when the patient gives his authorization for treatment. To do otherwise

The Concept of Informed Consent Ruth R. Faden and Tom L Beauchamp

What is an informed consent? Answering this question is complicated because there are two common, entrenched, and starkly different meanings of "informed consent." That is, the term is analyzable in two profoundly different ways—not because of mere subtle differences of connotation that appear in different contexts, but because two different conceptions of informed consent have emerged from its history and are still at work, however unnoticed, in literature on the subject. In one sense, which we label senses, "informed consent" is analyzable as a particular kind of action by individual patients and subjects: an autonomous authorization. In the second sense, sense2, informed consent is analyzable in terms of the web of cultural and policy rules and requirements of consent that collectively form the social practice of informed consent in institutional contexts where groups of patients and subjects must be treated in accordance with rules, policies, and standard practices. Here, informed consents are not always autonomous acts, nor are they always in any meaningful respect authorizations. Senses: Informed Consent as Autonomous Authorization

The idea of an informed consent suggests that a patient or subject does more than express agreement with,

jeopardizes the value of the documentation in legal defence. It should never be deferred until sometime after the procedure has been performed. The best practice is to record consent in a consistent fashion, keeping in mind provincial and institutional standards for the type of documentation to be employed. Conclusion

The process of consent has several purposes, the most important of which is to educate the patient about impending diagnostic, medical, or surgical interventions and to provide a firm basis for a legal defence to allegations of inadequate authorization for care. The best protection from consent litigation is adherence to the rules for consent, including adequate documentation of an authorization for treatment.

acquiesce in, yield to, or comply with an arrangement or a proposal. He or she actively authorizes the proposal in the act of consent. John may assent to a treatment plan without authorizing it. The assent may be a mere submission to the doctor's authoritative order, in which case John does not call on his own authority in order to give permission, and thus does not authorize the plan. Instead, he acts like a child who submits, yields, or assents to the school principal's spanking and in no way gives permission for or authorizes the spanking. Just as the child merely submits to an authority in a system where the lines of authority are quite clear, so often do patients Accordingly, an informed consent in senses should be defined as follows: an informed consent is an autonomous action by a subject or a patient that authorizes a professional either to involve the subject in research or to initiate a medical plan for the patient (or both). We can whittle down this definition by saying that an informed consent in sense, is given if a patient or subject with (1) substantial understanding and (2) in substantial absence of control by others (3) intentionally (4) authorizes a professional (to do intervention I). All substantially autonomous acts satisfy conditions 1-3; but it does not follow from that analysis alone that all such acts satisfy 4. The fourth condition is what distinguishes informed consent as one hind of autonomous action. (Note also that the definition restricts the kinds of authorization to medical and research contexts.) A person whose act satisfies conditions 1-3 but who refuses an intervention gives an informed refusal.

Source: From A History and Theory of Informed Consent (Oxford: Oxford University Press: 1986), 145-9. © 1986 by Oxford University Press. By permission of Oxford University Press, USA.

47

II

rill

111111

2 Medical Decision-Making

The Problem of Shared Decision-Making the exclusive province of the physician, and thus some This analysis of informed consent in sense, is delib- decisions are likely always to remain subject excluerately silent on the question of how the authorizer sively to the physician's authorization. Moreover, in the and agent(s) being authorized arrive at an agreement uncommon situation, a patient could autonomously about the performance of "I." Recent commentators authorize the physician to make all decisions about on informed consent in clinical medicine, notably Jay medical treatment, thus giving his or her informed Katz and the Presidenth Commission, have tended to consent to an arrangement that scarcely resembles the equate the idea of informed consent with a model of sharing of decision-making between doctor and patient. "shared decision-making" between doctor and patient. The Presidents Commission titles the first chapter of its Authorization report on informed consent in the patient—practitioner In authorizing, one both assumes responsibility for relationship "Informed Consent as Active, Shared what one has authorized and transfers to another Decision Making," while in Katz's work "the idea of one's authority to implement it. There is no informed informed consent" and "mutual decision-making" are consent unless one understands these features of the treated as virtually synonymous terms.' act and intends to perform that act. That is, one must There is of course an historical relationship in clin- understand that one is assuming responsibility and ical medicine between medical decision-making and warranting another to proceed. To say that one assumes responsibility does not informed consent. The emergence of the legal doctrine of informed consent was instrumental in drawing atten- quite locate the essence of the matter, however, because tion to issues of decision-making as well as authority a transfer of responsibility as well as of authority also in the doctor—patient relationship. Nevertheless, it occurs. The crucial element in an authorization is that is a confusion to treat informed consent and shared the person who authorizes uses whatever right, power, decision-making as anything like synonymous. For one or control he or she possesses in the situation to endow thing, informed consent is not restricted to clinical another with the right to act. In so doing, the authormedicine. It is a term that applies equally to biomed- izer assumes some responsibility for the actions taken ical and behavioural research contexts where a model by the other person. Here one could either authorize of shared decision-making is frequently inappropriate. broadly so that a person can act in accordance with Even in clinical contexts, the social and psychological general guidelines, or narrowly so as to authorize only a dynamics involved in selecting medical interventions particular, carefully circumscribed procedure. should be distinguished from the patient's authorization. We endorse Katzs view that effective communica- Sense2: Informed Consent as tion between professional and patient or subject is often Effective Consent instrumental in obtaining informed consents (sense), but we resist his conviction that the idea o f informed By contrast to sensep sense2, or effective consent, is consent entails that the patient and physician "share a policy-oriented sense whose conditions are not decision-making," or "reason together," or reach a con- derivable solely from analyses of autonomy and ausensus about what is in the patient's best interest. This thorization, or even from broad notions of respect for is a manipulation of the concept from a too singular autonomy. "Informed consent" in this second sense and defined moral perspective on the practice of medi- does not refer to autonomous authorization, but to a cine that is in effect a moral program for changing the legally or institutionally effective (sometimes misleadpractice. Although the patient and physician may reach ingly called valid) authorization from a patient or a a decision together, they need not. It is the essence of subject. Such an authorization is "effective" because it informed consent in sense, only that the patient or sub- has been obtained through procedures that satisfy the ject authorizes autonomously; it is a matter of indifference rules and requirements defining a specific institutional where or how the proposal being authorized originates. practice in health care or in research. The social and legal practice of requiring profesFor example, one might advocate a model of shared decision-making for the doctor—patient relationship sionals to obtain informed consent emerged in instiwithout simultaneously advocating that every medical tutional contexts, where conformity to operative rules procedure requires the consent of patients. Even re- was and still is the sole necessary and sufficient condilationships characterized by an ample slice of shared tion of informed consent. Any consent is an informed decision-making, mutual trust, and respect would and consent in sense, if it satisfies whatever operative rules should permit many decisions about routine and low- apply to the practice of informed consent. Sense, rerisk aspects of the patient's medical treatment to remain quirements for informed consent typically do not focus

Faden/Beauchamp: The Concept of Informed Consent

on the autonomy of the act of giving consent (as sense, does), but rather on regulating the behaviour of the consent-seeker and on establishing procedures and rules for the context of consent. Such requirements of professional behaviour and procedure are obviously more readily monitored and enforced by institutions. However, because formal institutional rules such as federal regulations and hospital policies govern whether an act of authorizing is effective, a patient or subject can autonomously authorize an intervention, and so give an informed consent in sense,, and yet not effectively authorize that intervention in sense2. Consider the following example. Carol and Martie are 19-year-old, identical twins attending the same university. Martie was born with multiple birth defects, and has only one kidney. When both sisters are involved in an automobile accident, Carol is not badly hurt, but her sister is seriously injured. It is quickly determined that Martie desperately needs a kidney transplant. After detailed discussions with the transplant team and with friends, Carol consents to be the donor. There is no question that Carol's authorization of the transplant surgery is substantially autonomous. She is well informed and has long anticipated being in just such a circumstance. She has had ample opportunity over the years to consider what she would do were she faced with such a decision Unfortunately, Carol's parents, who were in Nepal at the time of the accident, do not approve of her decision. Furious that they were not consulted, they decide to sue the transplant team and the hospital for having performed an unauthorized surgery on their minor daughter. (In this state the legal age to consent to surgical procedures is 21.) According to our analysis, Carol gave her informed consent in sense, to the surgery, but she did not give her informed consent in sense2. That is, she autonomously authorized the transplant and thereby gave an informed consent in sense, but did not give a consent that was effective under the operative legal and institutional policy, which in this case required that the person consenting be a legally authorized agent. Examples of other policies that can define sense2 informed consent (but not sense) include rules that consent be witnessed by an auditor or that there be a one-day waiting period between solicitation of consent and implementation of the intervention in order for the person's authorization to be effective. Such rules can and do vary, both within the United States by jurisdiction and institution, and across the countries of the world. Medical and research codes, as well as case law and federal regulations, have developed models of informed consent that are delineated entirely in a sense2 format, although they have sometimes attempted to justify the

rules by appeal to something like sense,. For example, disclosure conditions for informed consent are central to the history of "informed consent" in sense2, because disclosure has traditionally been a necessary condition of effective informed consent (and sometimes a sufficient condition!). The legal doctrine of informed consent is primarily a law of disclosure; satisfaction of disclosure rules virtually consumes "informed consent" in law This should come as no surprise, because the legal system needs a generally applicable informed consent mechanism by which injury and responsibility can be readily and fairly assessed in court. These disclosure requirements in the legal and regulatory contexts are not conditions of "informed consent" in sensei; indeed disclosure may be entirely irrelevant to giving an informed consent in sense,. If a person has an adequate understanding of relevant information without benefit of a disclosure, then it makes no difference whether someone discloses that information. Other sense2 rules besides those of disclosure have been enforced. These include rules requiring evidence of adequate comprehension of information and the aforementioned rules requiring the presence of auditor witnesses and mandatory waiting periods. Sense2 informed consent requirements generally take the form of rules focusing on disclosure, comprehension, the minimization of potentially controlling influences, and competence. These requirements express the present-day mainstream conception in the federal government of the United States. They are also typical of international documents and state regulations, which all reflect a sense2 orientation The Relationship Between Sense, and Sense2 A sense, "informed consent" can fail to be an informed consent in sense2 by a lack of conformity to applicable rules and requirements. Similarly, an informed consent in sense2 may not be an informed consent in sense,. The rules and requirements that determine sense2 consents need not result in autonomous authorizations at all in order to qualify as informed consents. Such peculiarities in informed consent law have led Jay Katz to argue that the legal doctrine of "informed consent" bears a "name" that "promises much more than its construction in case law has delivered." He has argued insightfully that the courts have, in effect, imposed a mere duty to warn on physicians, an obligation confined to risk disclosures and statements of proposed interventions. He maintains that "This judicially imposed obligation must be distinguished from the idea of informed consent, namely, that patients have a decisive role to play in the medical decision-making process. The idea of informed consent, though alluded to also in case law, cannot be implemented, as courts

49

■ II

I

I I Ell

1.11

111111

2 Medical Decision-Making

have attempted, by only expanding the disclosure requirements." By their actions and declarations, Katz believes, the courts have made informed consent a "cruel hoax" and have allowed "the idea of informed consent . . . to wither on the vine."' The most plausible interpretation of Katz's contentions is through the sense,/sense2 distinction. If a physician obtains a consent under the courts' criteria, then an informed consent (sense2) has been obtained. But it does not follow that the courts are using the right standards, or sufficiently rigorous standards in light of a stricter autonomy-based model—or Idea" as Katz puts it—of informed consent (sense).3 If Katz is correct that the courts have made a mockery of informed consent and of its moral justification in respect for autonomy then of course his criticisms are thoroughly justified. At the same time, it should be recognized that people can proffer legally or institutionally effective authorizations under prevailing rules even if they fall far short of the standards implicit in sense,. Despite the differences between sense, and sense2, a definition of informed consent need not fall into one or the other class of definitions. It may conform to both. Many definitions of informed consent in policy contexts reflect at least a strong and definite reliance on informed consent in sense,. Although the conditions of sense, are not logically necessary conditions for sense2, we take it as morally axiomatic that they ought to serve—and in fact have served—as the benchmark or model against which the moral adequacy of a definition framed for sense2 purposes is to be evaluated. This position is, roughly speaking, Katz's position. A defence of the moral viewpoint that policies governing informed consent in sense2 should be formulated to conform to the standards of informed consent in sense, is not hard to express. The goal of informed consent in medical care and in research—that is, the purpose behind the obligation to obtain informed consent—is to enable potential subjects and patients to make autonomous decisions about whether to grant or refuse authorization for medical and research interventions. Accordingly, embedded in the reason for having the social institution of informed consent is the idea that

institutional requirements for informed consent in sense2 should be intended to maximize the likelihood that the conditions of informed consent in sense, will be satisfied. A major problem at the policy level, where rules and requirements must be developed and applied in the aggregate, is the following: the obligations imposed to enable patients and subjects to make authorization decisions must be evaluated not only in terms of the demands of a set of abstract conditions of "true" or sense, informed consent, but also in terms of the impact of imposing such obligations or requirements on various institutions with their concrete concerns and priorities. One must take account of what is fair and reasonable to require of health care professionals and researchers, the effect of alternative consent requirements on efficiency and effectiveness in the delivery of health care and the advancement of science, and—particularly in medical care—the effect of requirements on the welfare of patients. Also relevant are considerations peculiar to the particular social context, such as proof, precedent, or liability theory in case law, or regulatory authority and due process in the development of federal regulations and IRB consent policies. Moreover, at the sense2 level, one must resolve not only which requirements will define effective consent; one must also settle on the rules stipulating the conditions under which effective consent must be obtained. In some cases, hard decisions must be made about whether requirements of informed consent (in sense2) should be imposed at all, even though informed consent (in sense2) could realistically and meaningfully be obtained in the circumstances and could serve as a model for institutional rules. For example, should there be any consent requirements in the cases of minimalrisk medical procedures and research activities? This need to balance is not a problem for informed consent in sensel, which is not policy-oriented. Thus, it is possible to have a morally acceptable set of requirements for informed consent in sense2 that deviates considerably from the conditions of informed consent in sense,. However, the burden of moral proof rests with those who defend such deviations since the primary moral justification of the obligation to obtain informed consent is respect for autonomous action.

Notes 1.

President's Commission. Making Health Care Decisions, vol. 1, 15; Katz, Jay. 1984. The Silent World of Doctor and Patient. The Free Press: New York, 87; and Katz, Jay. 1973. "The Regulation of Human Research—Reflections and Proposals," in Clinical Research 21: 758-91. Katz does not provide a sustained analysis of joint or shared decisionmaking, and it is unclear precisely how he would relate this notion to informed consent.

2.

3.

Katz, Jay. 1980. "Disclosure and Consent," in Genetics and the Law iI, ed. A. Milunsky and G. Annas. Plenum Press: New York, 122, 128. We have already noted that Katz's "idea" of informed consent as the active involvement of patients in the medical decision-making process is different from our senses.

Brody: Transparency: Informed Consent in Primary Care

Transparency: Informed Consent in Primary Care Howard Brody While the patient's right to give informed consent to medical treatment is now well established both in US law and in biomedical ethics, evidence continues to suggest that the concept has been poorly integrated into American medical practice, and that in many instances the needs and desires of patients are not being well met by current policies.' It appears that the theory and the practice of informed consent are out of joint in some crucial ways. This is particularly true for primary care settings, a context typically ignored by medical ethics literature, but where the majority of doctor—patient encounters occur. Indeed, some have suggested that the concept of informed consent is virtually foreign to primary care medicine where benign paternalism appropriately reigns and where respect for patient autonomy is almost completely absent.2 It is worth asking whether current legal standards for informed consent tend to resolve the problem or to exacerbate it. I will maintain that accepted legal standards, at least in the form commonly employed by courts, send physicians the wrong message about what is expected of them. An alternative standard that would send physicians the correct message, a conversation standard, is probably unworkable legally. As an alternative, I will propose a transparency standard as a compromise that gives physicians a doable task and allows courts to review appropriately. I must begin, however, by briefly identifying some assumptions crucial to the development of this position even though space precludes complete argumentation and documentation. Crucial Assumptions Informed consent is a meaningful ethical concept only to the extent that it can be realized and promoted within the ongoing practice of good medicine. This need not imply diminished respect for patient autonomy, for there are excellent reasons to regard respect for patient autonomy as a central feature of good medical care. Informed consent, properly understood, must be considered an essential ingredient of good patient care, and a physician who lacks the skills to inform patients appropriately and obtain proper consent as a non-medical, legalistic exercise designed to promote

patient autonomy, one that interrupts the process of medical care. However, available empirical evidence strongly suggests that this is precisely how physicians currently view informed consent practices. Informed consent is still seen as bureaucratic legalism rather than as part of patient care. Physicians often deny the existence of realistic treatment alternatives, thereby attenuating the perceived need to inform the patient of meaningful options. While patients may be informed, efforts are seldom made to assess accurately the patient's actual need or desire for information, or what the patient then proceeds to do with the information provided. Physicians typically underestimate patients' desire to be informed and overestimate their desire to be involved in decision-making. Physicians may also view informed consent as an empty charade, since they are confident in their abilities to manipulate consent by how they discuss or divulge in formation.3 A third assumption is that there are important differences between the practice of primary care medicine and the tertiary care settings that have been most frequently discussed in the literature on informed consent. The models of informed consent discussed below typically take as the paradigm case something like surgery for breast cancer or the performance of an invasive and risky radiologic procedure. It is assumed that the risks to the patient are significant, and the values placed on alternative forms of treatment are quite weighty. Moreover, it is assumed that the specialist physician performing the procedure probably does a fairly limited number of procedures and thus could be expected to know exhaustively the precise risks, benefits, and alternatives for each. Primary care medicine, however, fails to fit this model. The primary care physician, instead of performing five or six complicated and risky procedures frequently, may engage in several hundred treatment modalities during an average week of practice. In many cases, risks to the patient are negligible and conflicts over patient values and the goals of treatment or nontreatment are of little consequence. Moreover, in contrast to the tertiary care patient, the typical ambulatory patient is much better able to exercise freedom of choice and somewhat less likely to be intimidated by either the severity of the disease or the expertise of the physician; the opportunities for changing one's mind once treatment has begun are also much greater. Indeed, in primary care, it is much more likely for the full process of informed consent to treatment (such as the beginning

Source: From The Hastings Center Report 19, no. 5 (1989), 5-9. Reprinted by permission of The Hastings Center and the author.

8

II

41I1

1111;k1

C

I 2 Medical Decision-Making

and the dose adjustment of an antihypertensive medication) to occur over several office visits rather than at one single point in time. It might be argued that for all these reasons, the stakes are so low in primary care that it is fully appropriate for informed consent to be interpreted only with regard to the specialized or tertiary care setting. I believe that this is quite incorrect for three reasons. First, good primary care medicine ought to embrace respect for patient autonomy, and if patient autonomy is operationalized in informed consent, properly understood, then it ought to be part and parcel of good primary care. Second, the claim that the primary care physician cannot be expected to obtain the patient's informed consent seems to undermine the idea that informed consent could or ought to be part of the daily practice of medicine. Third, primary care encounters are statistically more common than the highly specialized encounters previously used as models for the concept of informed consent.' Accepted Legal Standards Most of the literature on legal approaches to informed consent addresses the tension between the community practice standard and the reasonable patient standard, with the latter seen as the more satisfactory, emerging legal standard.' However, neither standard sends the proper message to the physician about what is expected of her to promote patient autonomy effectively and to serve the informational needs of patients in daily practice. The community practice standard sends the wrong message because it leaves the door open too wide for physician paternalism. The physician is instructed to behave as other physicians in that specialty behave, regardless of how well or how poorly that behaviour serves patients' needs. Certainly, behaving the way other physicians behave is a task we might expect physicians to readily accomplish; unfortunately the standard fails to inform them of the end toward which the task is aimed. The reasonable patient standard does a much better job of indicating the centrality of respect for patient autonomy and the desired outcome of the informed consent process, which is revealing the information that a reasonable person would need to make an informed and rational decision. This standard is particularly valuable when modified to include the specific informational and decisional needs of a particular patient. If certain things were true about the relationship between medicine and law in today's society, the reasonable patient standard would provide acceptable

guidance to physicians. One feature would be that physicians esteem the law as a positive force in guiding their practice, rather than as a threat to their well-being that must be handled defensively. Another element would be a prospective consideration by the law of what the physician could reasonably have been expected to do in practice, rather than a retrospective review armed with the foreknowledge that some significant patient harm has already occurred. Unfortunately, given the present legal climate, the physician is much more likely to get a mixed or an undesirable message from the reasonable patient standard. The message the physician hears from the reasonable patient standard is that one must exhaustively lay out all possible risks as well as benefits and alternatives of the proposed procedure. If one remembers to discuss fifty possible risks, and the patient in a particular case suffers the fifty-first, the physician might subsequently be found liable for incomplete disclosure. Since lawsuits are triggered when patients suffer harm, disclosure of risk becomes relatively more important than disclosure of benefits. Moreover, disclosure of information becomes much more critical than effective patient participation in decision-making. Physicians consider it more important to document what they said to the patient than to document how the patient used or thought about that information subsequently. In specialty practice, many of these concerns can be nicely met by detailed written or videotaped consent documents, which can provide the depth of information required while still putting the benefits and alternatives in proper context. This is workable when one engages in a limited number of procedures and can have a complete document or videotape for each.° However, this approach is not feasible for primary care, when the number of procedures may be much more numerous and the time available with each patient may be considerably less. Moreover, it is simply not realistic to expect even the best educated of primary care physicians to rattle off at a moment's notice a detailed list of significant risks attached to any of the many drugs and therapeutic modalities they recommend. This sets informed consent apart from all other aspects of medical practice in a way that I believe is widely perceived by non-paternalistic primary care physicians, but which is almost never commented upon in the medical ethics literature. To the physician obtaining informed consent, you never know when you are finished. When a primary care physician is told to treat a patient for strep throat or to counsel a person suffering a normal grief reaction from the recent death of a relative, the physician has a good sense of what it means to complete the task at hand. When a physician

Brody: Transparency: Informed Consent in Primary Care e

is told to obtain the patient's informed consent for a medical intervention, the impression is quite different. A list of as many possible risks as can be thought of may still omit some significant ones. A list of all the risks that actually have occurred may still not have dealt with the patient's need to know risks in relation to benefits and alternatives. A description of all benefits, risks, and alternatives may not establish whether the patient has understood the information. If the patient says he understands, the physician has to wonder whether he really understands or whether he is simply saying this to be accommodating. As the law currently appears to operate (in the perception of the defensively minded physician), there never comes a point at which you can be certain that you have adequately completed your legal as well as your ethical task. The point is not simply that physicians are paranoid about the law; more fundamentally, physicians are getting a message that informed consent is very different from any other task they are asked to perform in medicine. If physicians conclude that informed consent is therefore not properly part of medicine at all, but is rather a legalistic and bureaucratic hurdle they must overcome at their own peril, blame cannot be attributed to paternalistic attitudes or lack of respect for patient autonomy. The Conversation Model

A metaphor employed by Jay Katz, informed consent as conversation, provides an approach to respect for patient autonomy that can be readily integrated within primary care practice.' Just as the specific needs of an individual patient for information, or the meaning that patient will attach to the information as it is presented, cannot be known in advance, one cannot always tell in advance how a conversation is going to turn out. One must follow the process along and take one's cues from the unfolding conversation itself. Despite the absence of any formal rules for carrying out or completing a conversation on a specific subject, most people have a good intuitive grasp of what it means for a conversation to be finished, what it means to change the subject in the middle of a conversation, and what it means to later reopen a conversation one had thought was completed when something new has just arisen. Thus, the metaphor suggests that informed consent consists not in a formal process carried out strictly by protocol but in a conversation designed to encourage patient participation in all medical decisions to the extent that the patient wishes to be included. The idea of informed consent as physician—patient conversation could, when properly developed, be a useful analytic tool for ethical

issues in informed consent, and could also be a powerful educational tool for highlighting the skills and attitudes that a physician needs to successfully integrate this process within patient care. If primary care physicians understand informed consent as this sort of conversation process, the idea that exact rules cannot be given for its successful management could cease to be a mystery. Physicians would instead be guided to rely on their own intuitions and communication skills, with careful attention to information received from the patient, to determine when an adequate job had been done in the informed consent process. Moreover, physicians would be encouraged to see informed consent as a genuinely mutual and participatory process, instead of being reduced to the one-way disclosure of information. In effect, informed consent could be demystified, and located within the context of the everyday relationships between physician and patient, albeit with a renewed emphasis on patient participation.8 Unfortunately, the conversation metaphor does not lend itself to ready translation into a legal standard for determining whether or not the physician has satisfied her basic responsibilities to the patient. There seems to be an inherently subjective element to conversation that makes it ill-suited as a legal standard for review of controversial cases. A conversation in which one participates is by its nature a very different thing from the same conversation described to an outsider. It is hard to imagine how a jury could be instructed to determine in retrospect whether or not a particular conversation was adequate for its purposes. However, without the possibility for legal review, the message that patient autonomy is an important value and that patients have important rights within primary care would seem to be severely undermined. The question then is whether some of the important strengths of the conversation model can be retained in another model that does allow better guidance. The Transparency Standard

I propose the transparency standard as a means to operationalize the best features of the conversation model in medical practice. According to this standard, adequate informed consent is obtained when a reasonably informed patient is allowed to participate in the medical decision to the extent that patient wishes. In turn, "reasonably informed" consists of two features: (1) the physician discloses the basis on which the proposed treatment, or alternative possible treatments, has been chosen; and (2) the patient is allowed to ask questions suggested by the

I I Li

54

.111111

2 Medical Decision-Making

disdosure of the physician's reasoning, and those questions are answered to the patient's satisfaction. According to the transparency model, the key to reasonable disclosure is not adherence to existing standards of other practitioners, nor is it adherence to a list of risks that a hypothetical reasonable patient would want to know. Instead, disclosure is adequate when the physician's basic thinking has been rendered transparent to the patient. If the physician arrives at a recommended therapeutic or diagnostic intervention only after carefully examining a list of risks and benefits, then rendering the physician's thinking transparent requires that those risks and benefits be detailed for the patient. If the physician's thinking has not followed that route but has reached its conclusion by other considerations, then what needs to be disclosed to the patient is accordingly different. Essentially, the transparency standard requires the physician to engage in the typical patient management thought process, only to do it out loud in language understandable to the patient.9 To see how this might work in practice, consider the following as possible general decision-making strategies that might be used by a primary physician: 1. The intervention, in addition to being presumably low-risk, is also routine and automatic. The physician, faced with a case like that presented by the patient, almost always chooses this treatment. 2. The decision is not routine but seems to offer clear benefit with minimal risk 3. The proposed procedure offers substantial chances for benefit, but also very substantial risks. 4. The proposed intervention offers substantial risks and extremely questionable benefits. Unfortunately, possible alternative courses of action also have high risk and uncertain benefit. The exact risks entailed by treatment loom much larger in the physician's own thinking in cases 3 and 4 than in cases 1 and 2. The transparency standard would require that physicians at least mention the various risks to patients in scenarios 3 and 4, but would not necessarily require physicians exhaustively to describe risks, unless the patient asked, in scenarios 1 and 2. The transparency standard seems to offer some considerable advantages for informing physicians what can legitimately be expected of them in the promotion of patient autonomy while carrying out the activities of primary care medicine. We would hope that the well-trained primary care physician generally thinks before acting. On that assumption, the physician can be told exactly when she is finished obtaining informed consent—first, she has to share her thinking

with the patient; secondly, she has to encourage and answer questions; and third, she has to discover how participatory he wishes to be and facilitate that level of participation. This seems a much more reasonable task within primary care than an exhaustive listing of often irrelevant risk factors. There are also considerable advantages for the patient in this approach. The patient retains the right to ask for an exhaustive recital of risks and alternatives. However, the vast majority of patients, in a primary care setting particularly, would wish to supplement a standardized recital of risks and benefits of treatment with some questions like, "Yes, doctor, but what does this really mean for me? What meaning am I supposed to attach to the information that you've just given?" For example, in scenarios 1 and 2, the precise and specific risk probabilities and possibilities are very small considerations in the thinking of the physician, and reciting an exhaustive list of risks would seriously misstate just what the physician was thinking. If the physician did detail a laundry list of risk factors, the patient might very well ask, "Well, doctor, just what should I think about what you have just told me?" and the thoughtful and concerned physician might well reply, "There's certainly a small possibility that one of these bad things will happen to you; but I think the chance is extremely remote and in my own practice I have never seen anything like that occur." The patient is very likely to give much more weight to that statement, putting the risks in perspective, than he is to the listing of risks. And that emphasis corresponds with an understanding of how the physician herself has reached the decision. The transparency standard should further facilitate and encourage useful questions from patients. If a patient is given a routine list of risks and benefits and then is asked "Do you have any questions?" the response may well be perfunctory and automatic. If the patient is told precisely the grounds on which the physician has made her recommendation, and then asked the same question, the response is much more likely to be individualized and meaningful. There certainly would be problems in applying the transparency standard in the courtroom, but these do not appear to be materially more difficult than those encountered in applying other standards; moreover, this standard could call attention to more important features in the ethical relationship between physician and patient. Consider the fairly typical case, in which a patient suffers harm from the occurrence of a rare but predictable complication of a procedure, and then claims that he would not have consented had he known about that risk Under the present "enlightened" court standards, the jury would examine whether a

Brody: Transparency: Informed Consent in Primary Care S

reasonable patient would have needed to know about that risk factor prior to making a decision on the proposed intervention. Under the transparency standard, the question would instead be whether the physician thought about that risk factor as a relevant consideration prior to recommending the course of action to the patient. If the physician did seriously consider that risk factor, but failed to reveal that to the patient, he was in effect making up the patient's mind in advance about what risks were worth accepting. In that situation, the physician could easily be held liable. If, on the other hand, that risk was considered too insignificant to play a role in determining which intervention ought to be performed, the physician may still have rendered his thinking completely transparent to the patient even though that specific risk factor was not mentioned. In this circumstance, the physician would be held to have done an adequate job of disclosing information.w A question would still exist as to whether a competent physician ought to have known about that risk factor and ought to have considered it more carefully prior to doing the procedure. But that question raises the issue of negligence, which is where such considerations properly belong, and removes the problem from the context of informed consent. Obviously, the standard of informed consent is misapplied if it is intended by itself to prevent the practice of negligent medicine. Transparency in Medical Practice Will adopting a legal standard like transparency change medical practice for the better? Ultimately only empirical research will answer this question. We know almost nothing about the sorts of conversations primary care physicians now have with their patients, or what would happen if these physicians routinely tried harder to share their basic thinking about therapeutic choices. In this setting it is possible to argue that the transparency standard will have deleterious effects. Perhaps the physician's basic thinking will fail to include risk issues that patients, from their perspective, would regard as substantial. Perhaps how physicians

think about therapeutic choice will prove to be too idiosyncratic and variable to serve as any sort of standard. Perhaps disclosing basic thinking processes will impede rather than promote optimal patient participation in decisions. But the transparency standard must be judged, not only against ideal medical practice, but also against the present-day standard and the message it sends to practitioners. I have argued that that message is, "You can protect yourself legally only by guessing all bad outcomes that might occur and warning each patient explicitly that he might suffer any of them." The transparency standard is an attempt to send the message, "You can protect yourself legally by conversing with your patients in a way that promotes their participation in medical decisions, and more specifically by making sure that they see the basic reasoning you used to arrive at the recommended treatment." It seems at least plausible to me that the attempt is worth making. The reasonable person standard may still be the best way to view informed consent in highly specialized settings where a relatively small number of discrete and potentially risky procedures are the daily order of business. In primary care settings, the best ethical advice we can give physicians is to view informed consent as an ongoing process of conversation designed to maximize patient participation after adequately revealing the key facts. Because the conversation metaphor does not by itself suggest measures for later judicial review, a transparency standard, or something like it, may be a reasonable way to operationalize that concept in primary care practice. Some positive side-effects of this might be more focus on good diagnostic and therapeutic decision-making on the physician's part, since it will be understood that the patient will be made aware of what the physician's reasoning process has been like, and better documentation of management decisions in the patient record. If these occur, then it will be clearer that the standard of informed consent has promoted rather than impeded high quality patient care.

Notes 1. 2.

3.

Charles W Lidz et al., (1983), "Barriers to Informed Consent," in Annals of Internal Medicine 99 (4): 539-43. Tom L. Beauchamp and Laurence McCullough, (1984), Medical Ethics: The Moral Responsibilities of Physicians (Prentice-Hall: Englewood Cliffs, NJ). For a concise overview of empirical data about contemporary informed consent practices see Ruth R. Faden and Tom L. Beauchamp, (1986), A History and

4.

Theory of Informed Consent (Oxford University Press: New York), 98-9 and associated footnotes. For efforts to address ethical aspects of primary care practice, see Ronald J. Christie and Barry Hoffmaster, (1986), Ethical Issues in Family Medicine (Oxford University Press: New York); and Harmon L. Smith and Larry R. Churchill, (1986), Professional Ethics and Primary Care Medicine (Duke University Press Durham, NC).

II III

G

I

IIan

411

1111.1.1

2 Medical Decision-Making 5.

6.

7.

Faden and Beauchamp, A History and Theory of Informed Consent, 23-49 and 114-50. I have also greatly benefited from an unpublished paper by Margaret Wallace. For a specialty opinion to the contrary, see W.H. Coles, et al., (1987), "Teaching Informed Consent," in Further Developments in Assessing Clinical Competence, ed. Ian R Hart and Ronald M. Harden (Can-Heal Publications: Montreal), 241-70. This paper is interesting in applying to specialty care a model very much like the one I propose for primary care. Jay Katz, (1984), The Silent World of Doctor and Patient (Free Press: New York).

Culture, Power, and Informed Consent: The Impact of Aboriginal Health Interpreters on Decision-Making Joseph Kaufert and John O'Neil Introduction

The signing of a consent agreement prior to surgery or invasive diagnostic or treatment procedures is a pivotal event in the negotiation of trust in the doctor—patient relationship. Most analysts have focused on the legal, ethical or procedural aspects of consent. However, there is growing recognition of the importance of considering political and cultural factors which lie outside the immediate context of the medical encounter and beyond the control of either physician or patient. This [reading] will examine the processes through which consent is negotiated when the patient is a Native from one of the remote areas of northern Canada. It will explore the application of ethnomedical approaches emphasizing explanatory models and an interactionist framework to understanding the impact of intermediaries in cross-cultural negotiation of consent. In our research on cross-cultural communication in urban hospitals, it was apparent that negotiations around the signing of a consent form provided the clearest illustration of the unequal knowledge and power of the clinician and the patient. The clinician's approach to obtaining consent was based primarily on a biomedical understanding of a particular disease and associated treatment procedures; the approach of the Native patient to giving consent was based on experiential and cultural knowledge of past and present illnesses, interpretations of the social meaning of hospital regulations and health professional behaviour, and general attitudes defining intergroup relations in the wider society . . .

8 Howard Brody, (1987), Stories of Sickness (Yale University Press: New Haven), 171-81. 9. For an interesting study of physicians' practices on this point, see William C. Wu and Robert A. Pearlman, (1988), "Consent in Medical Decision-making: The Role of Communication," inJournal of General Internal Medicine 3 (1): 9-14. 10. A court case that might point the way toward this line of reasoning is Precourt v. Frederick, 395 Mass. 689 (1985). See William J. Curran, (1986), "Informed Consent in Malpractice Cases: A Turn Toward Reality," in New England Journal of Medicine 314 (7): 429-31

Mediating Client and Physician Explanations of Invasive Diagnostic Procedures: A Case Example

The case documents the communication with an Aboriginal patient who was asked to consent to gastroscopic and colonoscopic examinations. The 46-year-old Cree-speaking woman was referred from a northern nursing station for further investigation of anemia by a gastroenterologist in Winnipeg, A series of encounters were videotaped at each stage of the diagnostic workup. Several encounters involved the signing of formal consent agreements. In each encounter the physician worked with a Cree-speaking medical interpreter to explain diagnostic and treatment options and to negotiate patient consent for examinations of the stomach, small and large intestine. In the initial encounter, the physician attempted to evaluate the patient's understanding of her own problem and to explain his diagnostic model of the probable cause of anemia. Specifically he attempted to move from discussing the client's understanding of anemia (conceptualized by the patient in terms of weakness) to a more complex model linking the loss of blood to the presence of lesions caused by anti-inflammatory medication. Following a cursory explanation of the general diagnosis, the physician moved to a series of diagnostic questions about presence of blood in the patient's stool. DOCTOR: She's anemic and pale, which means she

must be losing blood. INTERPRETER (Cree): This is what he says about you. You are pale, you have no blood. (Cree term for anemia connotes bloodless state.) DOCTOR: Has she had any bleeding from the bowel when she's had a bowel movement? INTERPRETER (Cree): When you have a bowel movement, do you notice any blood?

Source: From Health and Canadian Society: Sociological Perspectives, 3rd edn., D. Coburn and C. D'Arcy (Toronto: University of Toronto Press, 1998), 131-46. Reprinted by permission of the publisher.

Kaufert/O'Neil: Culture, Power, and Informed Consent C) PATIENT

(Cree): I'm not sure.

INTERPRETER (Cree): Is your stool ever black, or very light? What does it look like? PATIENT

(Cree): Sometimes dark.

At this point the patient told the interpreter that she did not understand how her "weakness" (anemia) was related to questions about gastrointestinal symptomatology in the physician's reference to dark stools. Without asking for additional explanation from the physician, the interpreter attempted to link the patient's understanding of her anemia with the concept of blood loss. INTERPRETER (Cree): We want to know, he says, why it is that you are lacking blood. Thath why he asked you what your stool looks like. Sometimes you lose your blood from there, when your stool is black.

In discussing the probable etiology of the woman's anemia, the physician introduced a complex explanatory model which explained gastric or intestinal bleeding in terms of the possible side effects of antiinflammatory medication for rheumatism. The patient again indicated that she did not understand why the questions about her experience with medication for rheumatism were relevant to the current diagnosis of problems with weakness and blood loss. The interpreter provided an unprompted explanation linking the line of questioning about the side effects of antiinflammatory drugs with the concept of blood loss. INTERPRETER (Cree): He says that those pills you are taking for rheumatism, sometimes they cause you to bleed inside, or you will spit up blood. Not everyone has these effects. This is why he wants to know about your medication.

from where you swallow. Nothing wrong that can be seen. PATIENT:

(Nods, but makes no verbal response.)

We're going to put a small tube in from the colon. It's only this big. To have a look, to see if there's any abnormality. It won't take too long and it will be very quick and you shouldn't be uncomfortable with it at all. Okay? . . . So we'll go ahead and do that now while we can. DOCTOR:

INTERPRETER (Cree): He wants to see you over here from where your bowel movements come from. Something will be put there, like the first one (the tube you swallowed), but smaller. So you can be examined "down there." Maybe somewhere "down there," it'll be seen that you are losing blood from there. The reason why you are lacking blood. That's what h8 looking for. The cause for your blood loss.

During the exchange the patient's willingness to extend the initial consent agreement to cover the investigation of the lower bowel is inferred from a nod and no real alternative was suggested by the physician. The patient was asked to initial the addition to the consent agreement, without formal translation of the English text. Colonoscopic and radiological examination of the intestine revealed a benign polyp. The physician recommended that the polyp should be cauterized through a second colonoscopy and asked the patient to sign a consent form for the additional procedure. Although risks and benefits were not formally discussed the interpreter elaborated on the basic diagnostic information provided by the physician. The interpreter also introduced a more formal decision point at which the patient was asked to give her formal consent. DOCTOR: We

Following gastroscopy, the physician attempted to explain the results of the gastric studies and at the same time to extend the initial consent agreement to permit colonoscopic examination of the lower bowel. DOCTOR: Everything looked good. There was no ulcer and no nasty disease in the stomach or the esophagus. No bleeding. You're still anemic so we still want to find out if there's any bleeding from the lower end.

(Cree): He says this about you: there's nothing visible in your stomach. Nothing, no sores, lumps, what they call "ulcers." Nothing

INTERPRETER

x-rayed the bowel.

INTERPRETER (Cree): And this is what they did this morning—when you were x-rayed. The pictures of the area you have bowel movements. PATIENT (Cree):

Yes.

DOCTOR: And

that shows a polyp, a small benign tumour. And I have to take that out.

INTERPRETER (Cree): This picture they took this morning. He saw it already. There's something growing there. About this size. And it has to be removed, because you might bleed from there.

II II I

II

111

111111

2 Medical Decision-Making PATIENT (Cree): Yes. DOCTOR: Now I can take it out without an operation, by putting a tube inside the bowel, and putting a wire around it and burning the polyp off. That stops the bleeding, no need for an operation. INTERPRETER (Cree): He says they can put in a tube like before and burn off the growth. PATIENT:

(Nods but makes no verbal response.)

If she wants to make the arrangements for the hospital admission she can come down and sign the consent form. DOCTOR:

INTERPRETER (English): Will

INTERPRETER (Cree): Do you want to have this procedure done? Will you consent to have this growth removed—burned? Do you consent to have it done? PATIENT

(Cree): I don't know.

INTERPRETER: You

know, if it's not removed it may bleed. It may cause problems.

(English directed at physician): Dr , isn't it true that if it's not removed, it can bleed and she can become anemic? INTERPRETER

That's correct, we feel that your anemia may result from the bleeding of the polyp.

DOCTOR:

there be complications?

DOCTOR: There

are a few complications but I think it would be difficult to explain them all. INTERPRETER (Cree): After this procedure has been done you won't be staying here at the hospital. You'll be able to go home on Saturday. It will be done on Friday, then youll already be able to go home on Saturday. It won't be long. But its entirely up to you.

In this exchange, the interpreter is providing more than a simple elaboration of the risks, benefits, and rationale for the procedure. The physician assumes that his explanation of the reasons for doing the procedure will be sufficient to obtain the patient's consent. In addition the physician closes the exchange by asserting that it would be too difficult to explain all the possible complications. However, in both instances, the interpreter does not provide a literal translation of the physician's side comments, but attempts to assure the patient and justify the procedure through explanations addressing the patient's concern with the length of her stay and desire to return to her home community. The interpreter assures the patient, "You'll be able to go home on Saturday," emphasizing the expectation that the operation will be minor and she won't be separated from family and home for long. The interpreter's statement linking approval of the consent to the patient's early return to her community occurs in Cree and therefore is not accessible to the physician. As the end of the encounter the physician included the consent agreement with the hospital admission protocol and assumed it would be signed with the other paper work. The interpreter provided a more direct opportunity for the patient to give or withhold consent.

INTERPRETER (Cree): If it's not removed, you may end up with cancer. You know? And you will not have an operation. Its harder when a person has an operation. You know? And [this procedure] that he's going to do will get it on time. Before it begins to bleed or starts to grow. You're lucky it's caught on time. And it will bother you when you have a bowel movement. This way there's no danger that this growth will bleed. PATIENT (Cree): I still

don't know.

INTERPRETER (Cree): Well if you want to come in for the procedure while you are here? It's all up to you to think about.

Again, the interpreter has assumed responsibility for providing a rationale for the procedure and explaining the potential benefits. Her explanations are also clearly based on her own medical knowledge, and her understanding of the patient's explanatory model. The fear of cancer in the Native community is linked to general understandings about the history of infectious disease epidemics that nearly destroyed Aboriginal society in North America. Cancer is increasingly viewed as the new "epidemic." The interpreter is using her knowledge of these fears to negotiate the patient's consent, but she is also using Cree models of negotiation emphasizing individual autonomy. Her final statement emphasizes her client's ultimate personal responsibility: "It's all up to you to think about." At this point, the patient accompanied the interpreter and physician to the appointment desk and scheduled the colonoscopy for the following day. After a brief summary of the text (which was printed in English) was provided by the interpreter, the patient

Kaufert/O'Neil: Culture, Power, and Informed Consent

signed the consent form. The formal act of signing the form was immediately subordinated to a discussion of specific arrangements for the client's discharge from hospital and travel arrangements for returning to the reserve community. In this case, consent was negotiated by drawing on expressions of trust by the patient about her relationship with the interpreter. The physician initially assumed that little explanation was required for consent, and indicated his unwillingness to negotiate. Information-sharing occurred gradually over the course of the encounter. The interpreter assumed the negotiator's role, based on shared cultural understanding of both biomedicine and Native culture. Throughout the sequence of interaction the interpreter worked to elicit and clarify both the client's and clinician's interpretation of the condition and the program of treatment. However, the interpreter's intervention introduced a third party into the clinician—client relationship. She directly influenced the course of the decision by independently introducing new information about illness and treatment options. She also imposed decision points where the patient could actually exercise her option to consent. The patient's willingness to allow the interpreter to negotiate consent is also evident in her reluctance to ask the physician direct questions. In response to the interpreter's questions about her understanding, she repeats at several junctures, "I don't know" This provides a cue for the interpreter to introduce further information or elaborate relative risks. The final consent is passive, in the sense that the patient signs the forms without further resistance. Power and Control in Consent Decisions among Native Canadians

The case study demonstrates the role of interpreteradvocates in redressing cultural and structural constraints for Native people in consent negotiations in urban hospitals (Kaufert and Koolage 1984). In some urban hospitals interpreters have expanded their role beyond narrow language translation functions to assume advocacy roles which empower the client through providing information about the structure

of the health care system and elaborating these treatment options. However, in consent negotiations involving both Indian and Inuit patients, the presence of a language interpreter or patient advocate as an intermediary raises a number of ethical and sociopolitical issues. For example, the involvement of interpreteradvocates in the doctor—patient relationship may shift responsibility and initiative for disclosure from the professional to the intermediary. The translator may exercise control through selective interpretation of information provided by the client. The interpreter may also priorize and filter information about treatment options and associated risks or benefits. The interpreters' function of mediating and priorizing information occurs within a linguistic and cultural "black box." Within this box the interpreter may actively intervene on the patient's behalf, or use his or her cultural knowledge and personal rapport with the patient to reinforce the clinician's definition of appropriate treatment choices. Consideration of this variable by clinicians in obtaining informed consent is important. In summary, sociocultural analysis of real interaction sequences in the negotiation of consent between the clinician and client differs fundamentally from legal or ethical analysis of the marker decisions. For Native clients, agreements may reflect the emergence of trust relationships achieved through an extended, incremental process of exchange rather than a formal, final contract. Interactionist and ethnomedical approaches more clearly reveal the communication processes and power relations which are part of the process of translating and prioritizing information. Our analysis of the role of Native medical interpreters in both case studies clearly indicates that dyadic clinician—client interaction is strongly influenced by intermediaries. Both confirm that translators, cultural brokers, and personal advocates negotiate shared meanings and influence the balance of power in cross-cultural, clinical communication. As well as demarcating formal legal and ethical decision points, cross-cultural consent agreements also function as integrative rituals through which participants reconcile power imbalance and negotiate clinical trust.

References

Kaufert, J., and W. Koolage. 1984. "Role Conflict among Culture Brokers: The Experience of Native Canadian Medical Interpreters," in Social Science and Medicine 18 (3): 283-6.

This manuscript contains sections of a previously published manuscript, "Biomedical Rituals and Informed Consent," originally printed in Social Science Perspectives on Medical Ethics, ed. G. Weisz, University of Pennsylvania Press, 1991.

I I UN

60

1111111

2 Medical Decision-Making

Informed Consent and Public Health Onora O'Neill 1. Informed Consent and Public Health

Medical ethics has been transformed during the past 30 years. One conspicuous change has been a steadily increasing focus on informed consent, which is now usually taken to be essential for any ethically acceptable medical practice. The literature on informed consent in medical ethics is vast and repetitive. However, it has significant limitations. Some of the difficulties are well known and recalcitrant, although this seldom dents the enthusiasm of those who think informed consent essential to (or even sufficient for) ethically acceptable medical practice. In this paper I shall mention the commonly discussed difficulties but shall concentrate on some less discussed but philosophically deeper difficulties that limit the use to which informed consent procedures can be put in public health provision. Some of the most frequent disagreements about informed consent are about the basic reasons for thinking that it is ethically important. Is informed consent required to respect persons or to respect the autonomy of persons? If the latter, which conception of autonomy is relevant? If some persons are more autonomous than others, will informed consent procedures be more important for them? Or will they, on the contrary, be more important for those with limited autonomy? Alternatively are informed consent procedures required because they provide a degree of assurance that patients are not deceived or coerced in the course of clinical practice? (Faden & Beauchamp 1986; Wolpe 1998; O'Neill 2002a). A second, even more frequently discussed, range of problems arises when patients cannot grasp the information that is essential to giving informed consent. If they cannot understand the proposition to which their consent is sought, they cannot give or refuse informed consent. The hard cases are numerous and intractable. Many patients cannot consent to medical intervention or treatment because they are too young, too ill, too disabled, or too demented to understand the information that they would have to grasp to make an informed choice. As they can hardly be denied medical treatment because of these difficulties, it must be given without their consent. Should it then be given on the basis of others' consent (e.g., that of parents, guardians, or relatives)? Or does the very idea of proxy consent

undermine the fundamental concerns that are taken to justify informed consent requirements or even show disrespect for individuals or their autonomy? Should proxy consent perhaps be set aside as mere pretence and replaced with greater reliance on professional judgement of each patient's best interests? Or would doing so revert to unacceptable medical paternalism and so fail to respect patients and their (faltering) autonomy? Responses to these questions often propose ways of revising or refining the procedures used for requesting and recording consent to make consenting easier for less competent patients. But even the most energetic and time-consuming presentation of user-friendly information, even the most elaborate and detailed consent forms and procedures, will not make informed consent possible for numerous patients with various types of incapacity. However, these are not the deepest difficulties. Informed consent procedures are problematic not merely because their philosophical rationale is disputed, and not merely because some individuals lack competence to consent, or lack it at some times. The most basic philosophical difficulties with informed consent arise because consent is a propositional attitude. Consenting—like other cognitive states or acts such as knowing, believing, understanding, hoping, wondering, thinking, desiring, or fearing—takes a proposition as its object. Hence consent is never directed at a medical intervention as such but rather at some proposition that describes an intended intervention. However, any intervention can be described in many different ways. Even in the easy cases, where competent patients consent to a procedure or treatment described in a certain way, they may not be aware of and may not consent to other true descriptions of the same intervention or of its more obvious effects. This can be the case even when the second description is entailed by the first or when it refers to an obvious consequence of the state of affairs described by the first. So propositional attitudes are opaque. A person may know or understand or hope that x but not know or understand or hope that y even where x entails y. A patient may consent to an intervention, and the intervention as described may entail or bring about certain conditions; however, the patient may not consent to those conditions because he or she may not grasp the entailment relation or the causal connection. For example, a parent may consent to the removal of tissues from their dead child, and the undifferentiated reference to tissues will cover entire organs; however,

Source: From Philosophical Transactions of the Royal Society B 359 (2004), 1133-6. Reprinted with permission.

O'Neill: Informed Consent and Public Health 61

the parent may not know that this is so and be upset to discover that entire organs were removed on the basis of general consent to the removal of tissues. Because propositional attitudes are always opaque, the basic difficulties of informed consent procedures cannot be reliably eliminated by making information more available or consent procedures easier to follow. For unless the information is actually understood by those whose informed consent is requested, genuine consent will not stretch to the relevant proposition. This difficulty is ubiquitous within the central debates of medical ethics, although barely discussed (O'Neill 2002b). 2. Explicit or Implied? Specific or Generic?

These difficulties have been heightened rather than resolved by recent attempts to improve informed consent procedures. Two supposed improvements are often advocated and raise particular difficulties. The first demands that all consent should be explicit rather than implied; the second that it be specific rather than generic. The distinction between explicit and implied consent contrasts ways of consenting. Explicit consent typically relies on documents, signatures, and formal statements; it may require witnesses who confirm that proper procedures for consenting have been followed. The formal procedures are typically designed to create enduring records, thereby reducing later uncertainty about the consent given and perhaps forestalling dissatisfaction, complaint, or litigation. Patients who consent explicitly to proposed interventions thereby accept that they cannot later claim that they were injured or wronged and accept that they will not have grounds for complaint or litigation. By contrast, implied consent is inferred from a patient's action. For example, agreement to blood being taken or to having an injection is standardly signified by extending one's arm for the doctor to take the blood or give the injection. No documentation of the consent is required. It would be possible—but laborious—to replace the implied consent that is currently seen as sufficient in these and similar cases with explicit consent procedures. It would be possible—but strenuous—to introduce explicit consent procedures for the most minor and routine of medical interventions. However much we introduce additional explicit procedures and consent forms for interventions that are now performed on the basis of implied consent, explicit consent always relies on background understandings that remain implicit. The longest and most complex consent form cannot include a complete description of everything that will be done. Much is taken as understood, and consent based on those understandings can only be

implied. No programme for replacing implied with explicit consent can be complete. The distinction between specific and generic consent applies to the propositions to which consent is given, rather than to processes of consenting. For example, consent may be given to the removal of tissues, or alternatively to the removal of a specific sort of tissue, or even to the removal of tissue for a specific use, such as diagnosis, or as part of a cancer treatment or postmortem to determine the cause of death. The descriptions to which consent is given are always incomplete. We can always add more detail. So those who believe that informed consent should be highly specific need to explain how specific it has to be to constitute ethically adequate informed consent. Answering this question may be no easier than answering the pseudo-question "how long is a piece of string?" These problems are not "merely theoretical." They surface and create practical problems wherever data protection issues arise, in secondary data analyses, indeed in the use of tissues for comparative study. If personal information is to be "processed" (obtained, recorded, processed, sorted, used) only in accordance with the consent of those to whom it pertains (the "data subjects"), then it will constantly turn out that if the consent obtained is sufficiently specific to permit a certain use, it will also be sufficiently specific to preclude other uses. Similarly, if tissues and information from past patients are to be studied for purposes that could not have been anticipated, they must be studied without specific consent, because such consent could not in principle have been requested or given at the relevant time. The problem is not merely that in the past consent procedures were too lax and that the relevant consents were not obtained or not recorded with sufficient clarity (although that was often the case). The problem is deeper, indeed irresolvable, because many valuable purposes could not have been anticipated at the time that tissues were removed and stored. For example, nobody could tell in advance when information and tissues obtained in the course of treating past patients will turn out to be useful for unanticipated research. Both clinical information and tissue samples pertaining to deceased patients may later turn out to be vital for reaching a better understanding of new diseases. When the first patients with VCJD died, the only way in which pathologists could determine whether this was a new disease was by comparing their brain tissue with samples taken from patients who had died of Creutzfeldt— Jakob disease across many decades in many countries. Tissue and information from those who died in the 1918 influenza epidemic may yet prove valuable in studying emerging diseases (Gamblin et al. 2004). It is

I I al

II1La

2 Medical Decision-Making

difficult to see how secondary data analysis of this sort can proceed if access to tissue samples from and data pertaining to past patients, their treatment and their clinical outcomes, cannot be consulted without specific prior consent to such studies. Although the incidence of disease could be monitored on the basis of collecting deidentified data, linked data are needed for all more elaborate forms of retrospective study and public health research. Specific consent requirements undermine secondary data analysis in medical and other areas of inquiry; they would close down epidemiology. However, a claim that specific consent is ethically required for retrospective study of linked patient information and tissues is neither intuitive nor plausible, provided that standard safeguards such as the approval of ethics committees and anonymization are in place. What would we think of a patient who asks his doctor how he knows that a medicine will prove helpful, is told that it helped nearly all patients with the same condition, accepts the treatment but then reflises consent to the further study of information—even anonymized information—about the clinical outcome in his own case? 3. Public Health and Public Goods There are further and more general reasons for rejecting the current tendency to suppose that informed consent procedures are the touchstone of ethically acceptable medical provision. One of the most significant is that informed consent procedures are inapplicable whenever the goods or benefits to be provided are public goods. Certain types of goods—consumer goods, clinical care—can be provided for individuals, and their provision can in principle be made to be contingent on individual consent. The difficulties that informed consent requirements raise may prove irresolvable in some cases and resolvable in others. By contrast, if public goods are provided for any, they have to be provided for many. Some types of public good must be provided (or not provided) for whole populations; others may be provided (or not provided) for more restricted groups. For present purposes I leave these differences aside, to make the simple point that the provision of public goods cannot be made contingent on individual consent. For example, road safety, food safety water safety safe medicines, and measures that protect against infection cannot be tailored to individual choice. Because there are no obligations to do the impossible ("ought implies can"), informed consent cannot be ethically required for the provision of public goods. The implications of these thoughts about public goods are wider than may at first seem to be the case. For example, clinical care itself has to be provided to

standards and formats that are also largely fixed and uniform and so cannot be treated as a matter for informed consent. The scaffolding of professional training, of institutional structures, of public funding, of physical facilities are all public goods. The public provision of health care can reflect democratic process and thereby certain forms of collective choice; but its basic structures cannot be geared to individual choice. Unavoidably there are large areas of medical ethics in which informed consent can play no part, or at most a minor part. What then are the appropriate ethical and other normative issues in these areas of medical ethics? 4. Health and Justice The ethical reasoning most commonly used in medical ethics focuses on transactions between individual professionals and individual patients or between individual professionals and individual research subjects. Individualistic approaches are not likely to prove to be useful for analysing ethical questions about the provision of public goods, such as public health provision. However, it may seem that theories of justice will also provide an inappropriate account of the normative reasoning most relevant to the provision of public goods or to public health ethics. Most uses of theories of justice in health care ethics have addressed distributive issues, such as the just distribution of clinical care. Discussions of health care allocation decisions—of rationing—are discussions of the just distribution of a good that can be made contingent on individual choice. Theories of distributive justice also fail to address the distinctive ethical questions that arise in providing public goods and so are not helpful for public health ethics. But there is more to justice than distributive justice. All theories of justice also address the justification of compulsion and in particular the justification of the forms of compulsion on which any legal order depends. There are many differing theories of justice, but by way of illustration I shall instance one theory, because it gives a particularly large—supposedly maximal—role to individual liberty and so to individual consent procedures. This theory is John Stuart Mill's form of liberalism, especially as developed in On Liberty. I choose Mill's account of justice not because I assume or argue that it is more plausible than other accounts but because he explicitly opposes compulsion except in very limited circumstances. Mill famously claimed that . . . the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of any of their number is self-protection. (Mill 1989, ch. 1, p. 13)

O'Neill: Informed Consent and Public Health

Compulsion, on this view, is permitted only where needed so as to protect others: it is unjust unless needed to prevent harm to others. Public health provision is an obvious area where Mill's arguments are relevant, because compulsion may be needed to prevent individual action that might harm others' health. Prevention of transmission of disease is a central case for Mill's justification of compulsion. Given the death rate from severe acute respiratory syndrome and the seriousness of the illness, Mill would view it as legitimate to make certain forms of action or treatment compulsory because the risk of transmission can be moderately high and the risk of death for those who succumb high for some age groups. Depending on the gravity of the risk, it might be permissible to institute mandatory monitoring of those who may have been exposed, mandatory vaccination (if a vaccine is developed), restrictions on free movement, or quarantine. Similarly, where vaccination is safe and effective, Mill's argument would suggest that it could legitimately be made compulsory to produce the herd immunity that protects vulnerable individuals who cannot (yet) be vaccinated. Making public health measures such as

these compulsory would hardly have seemed controversial a century ago. It has come to seem controversial on the basis of an illusory assumption that all medical provision, and with it public health provision, can be organized on the basis of informed consent of individuals. It cannot. Of course, there are always difficulties in judging how great a risk is and in deciding which forms of compulsion are most effective and most readily justified in a particular case. Such issues can be resolved only on a case-by-case basis, using the right expertise and the right information. Often there are deficits in information and in expertise, as with the many uncertainties about the mode of transmission and likely spread of vCJD and other new transmissible diseases. But where information and expertise point to the likelihood of harm to others, there are even on a very strong liberal account of justice no good ethical arguments to forbid all compulsion. To the contrary, appeals to individual consent do not offer a coherent, let alone an acceptable, way of approaching public health provision. Salus populi suprema lex is not an obsolete thought (Cicero 1928).

References Cicero, M. T. 1928 De Legibus, III, iii, 8. In De Re Publica, De Legibus (tr. Clinton Walker Keyes; ed. M. T. Cicero). Loeb Classical Library. Faden, R & Beauchamp, T. 1986 A history and theory of informed consent. Oxford University Press. Gamblin, S. J. (and 11 others) 2004 The structure and receptor-binding properties of the 1918 influenza hemagglutinin. Science 303, 1838-1842. (DOI 10.1126/science .1093155.) Mill, J. S. 1989 On liberty (1859). In On liberty and other writings (ed. S. Collini), p. 13. Cambridge University Press.

O'Neill, 0. 2002 Autonomy and trust in bioethics. Cambridge University Press. O'Neill, 0. 2002 Some limits of informed consent. J. Med. Ethics 28, 1-3. See ftp://ftpbmj.bma.org.uk/SJFauthor/ JME/ feb03/1Fme021800.pdf. Wolpe, P. R. 1998 The triumph of autonomy in American bioethics: a sociological view. In Bioethics and society: constructing the ethical (ed. R. DeVries & J. Subedi), pp. 38-59. Englewood Cliffs, NJ: Prentice-Hall.

2.4 Substituted Judgments

Enough: The Failure of the Living Will Angela Fagerlin and Carl E. Schneider By their fruits ye shall know them. Enough. The living will has failed, and it is time to say so.

We should have known it would fail: A notable but neglected psychological literature always provided arresting reasons to expect the policy of living wills to misfire. Given their alluring potential, perhaps they were worth trying. But a crescendoing empirical literature and persistent clinical disappointments reveal that the rewards of the campaign to promote living wills do

Source: From The Hastings Center Report 34,2 (2004), 30-42. Reprinted with permission of The Hastings Center and the author.

II

i311

111111

2 Medical Decision-Making

not justify its costs. Nor can any degree of tinkering ever make the living will an effective instrument of social policy. As the evidence of failure has mounted, living wills have lost some of their friends. We offer systematic support for their change of heart. But living wills are still widely and confidently urged on patients, and they retain the allegiance of many bioethicists, doctors, nurses, social workers, and patients. For these loyal advocates, we offer systematic proof that such persistence in error is but the triumph of dogma over inquiry and hope over experience. A note about the scope of our contentions: First, we reject only living wills, not durable powers of attorney: Second, there are excellent reasons to be skeptical of living wills on principle. For example, perhaps former selves should not be able to bind latter selves in the ways living wills contemplate.' And many people do and perhaps should reject the view of patients, their families, and their communities that informs living wills.2 But we accept for the sake of argument that living wills desirably serve a strong version of patients' autonomy: We contend, nevertheless, that living wills do not and cannot achieve that goal. And a stipulation: We do not propose the elimination of living wills. We can imagine recommending them to patients whose medical situation is plain, whose crisis is imminent, whose preferences are specific, strong, and delineable, and who have special reasons to prescribe their care. We argue on the level of public policy: In an attempt to extend patients' exercise of autonomy beyond their span of competence, resources have been lavished to make living wills routine and even universal. This policy has not produced results that recompense its costs, and it should therefore be renounced. Living wills are a bioethical idea that has passed from controversy to conventional wisdom, from the counsel of academic journals to the commands of law books, from professors' proposal to professional practice. Advance directives generally are embodied in federal policy by the Patient Self-Determination Act, which requires medical institutions to give patients information about their state's advance directives. In turn, the law of every state provides for advance directives, almost all states provide for living wills, and most states "have at least two statutes, one establishing a living will type directive, the other establishing a proxy or durable power of attorney for health care."3 Not only are all these statutes very much in effect, but new legislative activity is constant. Senators Rockefeller, Collins, and Specter have introduced bills to "strengthen" the PSDA and living wills,4 and state legislatures continue to amend living will statutes and to enact new ones.

Courts and administrative agencies too have become advocates of living wills. The Veterans Administration has proposed a rule to encourage the use of advance directives, including living wills.5 Where legislatures have not granted living wills legal status, some courts have done so as a matter of common law, and where legislatures have granted them legal status, courts have cooperated with eager enthusiasm.° Living wills have assumed special importance in states that prohibit terminating treatment in the absence of strong evidence of the patient's wishes.' One supreme court summarized a common theme: "[A] written directive would provide the most concrete evidence of the patient's decisions, and we strongly urge all persons to create such a directive." The grandees of law and medicine also give their benediction to the living will. The AMA's Council on Ethical and Judicial Affairs proclaims: "Physicians should encourage their patients to document their treatment preferences or to appoint a health care proxy with whom they can discuss their values regarding health care and treatment."9 The elite National Conference of Commissioners on Uniform State Laws continues to promulgate the Uniform Health-Care Decisions Act, a prestigious model statute that has been put into law in a still-growing number of states. Medical journals regularly admonish doctors and nurses to see that patients have advance directives, including living wills.10 Bar journals regularly admonish lawyers that their clients—all their clients—need advance directives, including living wills." Researchers demonstrate their conviction that living wills are important by the persistence of their studies of patients' attitudes toward living wills and ways of inveigling patients to sign them. Not only do legislatures, courts, administrative agencies, and professional associations promote the living will, but other groups unite with them. The Web abounds in sites advocating the living will to patients." The web site for our university's hospital plugs advance directives and suggests that it "is probably better to have written instructions because then everyone can read them and understand your wishes."" Our own experience in presenting this paper is that its thesis provokes some bioethicists to disbelief and indignation. It is as though they simply cannot bear to believe that living wills might not work. How can anything so intuitively right be proved so infuriatingly wrong? And indeed, bioethicists continue to investigate ways the living will might be extended (to deal with problems of the mentally ill and of minors, for example) and developed for other countries. Although some sophisticated observers have long doubted the wisdom of living wills," proponents have

Fag erli n/Schn eider: Enough:The Failure of the Living Will

tended to respond in one of three ways, all of which preserve an important role for living wills. First, proponents have supposed that the principal problem with living wills is that people just won't sign them. These proponents have persevered in the struggle to find ways of getting more people to sign up.13 Second, proponents have reasserted the usefulness of the living wills. For example, Norman Cantor, distinguished advocate of living wills, acknowledges that " [s] ome commentators doubt the utility or efficacy of advance directives" (by which he means the living will), but he concludes that "these objections don't obviate the importance of advance directives.' Other proponents are daunted by the criticisms of living wills but offer new justifications for them. Linda Emanuel, another eminent exponent of living wills, writes that "living wills can help doctors and patients talk about dying" and can thereby "open the door to a positive, caring approach to death."17 Third, some proponents concede the weaknesses of the living will and the advantages of the durable power of attorney and then propose a durable power of attorney that incorporates a living will. That is, the forms they propose for establishing a durable power of attorney invite their authors to provide the kinds of instructions formerly confined to living wills.'8 None of these responses fully grapples with the whole range of difficulties that confound the policy promoting living wills. In fairness, this is partly because the case against that policy has been made piecemeal and not in a full-fledged and full-throated analysis of the empirical literature on living wills. In sum, the law has embraced the principle of living wills and cheerfully continues to this moment to expound and expand that principle. Doctors, nurses, hospitals, and lawyers are daily urged to convince their patients and clients to adopt living wills, and patients hear their virtues from many other sources besides. Some advocates of living wills have shifted the grounds for their support of living wills, but they persist in believing that they are useful. The time has come to investigate those policies and those hopes systematically. That is what this article attempts. We ask an obvious but unasked question: What would it take for a regime of living wills to function as their advocates hope? First, people must have living wills. Second, they must decide what treatment they would want if incompetent. Third, they must accurately and lucidly state that preference. Fourth, their living wills must be available to people making decisions for a patient. Fifth, those people must grasp and heed the living will instructions. These conditions are unmet and largely unmeetable.

Do People Have Living Wills? At the level of principle, living wills have triumphed among the public as among the princes of medicine. People widely say they want a living will, and living wills have so much become conventional medical wisdom "that involvement in the process is being portrayed as a duty to physicians and others.' Despite this, and despite decades of urging, most Americans lack them." While most of us who need one have a property will, roughly 18 percent have living wills." The chronically or terminally ill are likelier to prepare living wills than the healthy, but even they do so fitfully." In one study of dialysis patients, for instance, only 35 percent had a living will, even though all of them thought living wills a "good idea."23 Why do people flout the conventional wisdom? The flouters advance many explanations." They don't know enough about living wills,25 they think living wills are hard to execute,26 they procrastinate," they hesitate to broach the topic to their doctors (as their doctors likewise hesitate).28 Some patients doubt they need a living will. Some think living wills are for the elderly or infirm and count themselves in neither group." Others suspect that living wills do not change the treatment people receive; 91 percent of the veterans in one study shared that suspicion.3° Many patients are content or even anxious to delegate decisions to their families,' often because they care less what decisions are made than that they are made by people they trust. Some patients find living wills incompatible with their cultural traditions.32 Thus in the large SUPPORT and HELP studies, most patients preferred to leave final resuscitation decisions to their family and physician instead of having their own preferences expressly followed (70.8% in HELP and 78.0% in SUPPORT). "This result is so striking that it is worth restating: not even a third of the HELP patients and hardly more than a fifth of the SUPPORT patients "would want their own preferences followed."33 If people lacked living wills only because of ignorance, living wills might proliferate with education. But studies seem not to "support the speculations found in the literature that the low level of advance directives use is due primarily to a lack of information and encouragement from health care professionals and family members."34 Rather, there is considerable evidence "that the elderly's action of delaying execution of advance directives and deferring to others is a deliberate, if not an explicit, refusal to participate in the advance directives process."35 The federal government has sought to propagate living wills through the Patient Self-Determination

• NI II

II

illl

111Li

2 Medical Decision-Making

Act,36 which essentially requires medical institutions to inform patients about advance directives. However, "empirical studies demonstrate that: the PSDA has generally failed to foster a significant increase in advance directives use; it is being implemented by medical institutions and their personnel in a passive manner, and the involvement of physicians in its implementation is lacking."37 One commentator even thinks "the PSDA's legal requirements have become a ceiling instead of a floor."38 In short, people have reasons, often substantial and estimable reasons, for eschewing living wills, reasons unlikely to be overcome by persuasion. Indeed, persuasion seems quickly to find its limits. Numerous studies indicate that without considerable intervention, approximately 20 percent of us complete living wills, but programs to propagate wills have mixed results.39 Some have achieved significant if still limited increases in the completion of living wills," while others have quite failed to do so.' Thus we must ask: If after so much propaganda so few of us have living wills, do we really want them, or are we just saying what we think we ought to think and what investigators want to hear?

Do People Know What They Will Want? Suppose, counterfactually, that people executed living wills. For those documents to work, people would have to predict their preferences accurately. This is an ambitious demand. Even patients making contemporary decisions about contemporary illnesses are regularly daunted by the decisions' difficulty. They are human. We humans falter in gathering information, misunderstand and ignore what we gather, lack well-considered preferences to guide decisions, and rush headlong to choice." How much harder, then, is it to conjure up preferences for an unspecifiable future confronted with unidentifiable maladies with unpredictable treatments? For example, people often misapprehend crucial background facts about their medical choices. Oregon has made medical policy in fresh and controversial ways, has recently had two referenda on assisted suicide, and alone has legalized it. Presumably, then, its citizens are especially knowledgeable. But only 46 percent of them knew that patients may legally withdraw life-sustaining treatment. Even experience is a poor teacher "Personal experience with illness . . . and authoring an advance directive . . . were not significantly associated with better knowledge about options."43 Nor do people reliably know enough about illnesses and treatments to make prospective life-ordeath decisions about them. To take one example

from many, people grossly overestimate the effectiveness of CPR and in fact hardly know what it is.44 For such information, people must rely on doctors. But doctors convey that information wretchedly even to competent patients making contemporaneous decisions. Living wills can be executed without even consulting a doctor," and when doctors are consulted, the conversations are ordinarily short, vague, and tendentious. In the Tulsky study, for example, doctors only described either "dire scenarios . . . in which few people, terminally ill or otherwise, would want treatment" or "situations in which patients could recover with proper treatment."' Let us put the point differently The conventional— legal and ethical wisdom—insists that candidates for even a flu shot give "informed consent." And that wisdom has increasingly raised the standards for disclosure." If we applied those standards to the information patients have before making the astonishing catalog of momentous choices living wills can embody, the conventional wisdom would be left shivering with indignation. Not only do people regularly know too little when they sign a living will, but often (again, we're human) they analyze their choices only superficially before placing them in the time capsule. An ocean of evidence affirms that answers are shaped by the way questions are asked. Preferences about treatments are influenced by factors like whether success or failure rates are used," the level of detail employed," and whether long- or short-term consequences are explained first.5° Thus in one study, "201 elderly subjects opted for the intervention 12 % of the time when it was presented negatively, 18% of the time when it was phrased as in an advance directive already in use, and 30% of the time when it was phrased positively. Seventy-seven percent of the subjects changed their minds at least once when given the same case scenario but a different description of the intervention. "51 If patients have trouble with contemporaneous decisions, how much more trouble must they have with prospective ones. For such decisions to be "true," patients' preferences must be reasonably stable. Surprisingly often, they are not. A famous study of eighteen women in a "natural childbirth" class found that preferences about anesthesia and avoiding pain were relatively stable before childbirth, but at "the beginning of active labor (4-5 cm dilation) there was a shift in the preference toward avoiding labor pains. . . . During the transition phase of labor (8-10 cm) the values remained relatively stable, but then . . . the mothers' preferences shifted again at postpartum toward avoiding the use of anesthesia during the delivery of her next child."32 And not only are preferences surprisingly labile, but

Fagerlin/Schneider: Enough:The Failure of the Living Will ()

people have trouble recognizing that their views have changed.53 This makes it less likely they will amend their living wills as their opinions develop and more likely that their living wills will treasonously misrepresent their wishes. Instability matters. The healthy may incautiously prefer death to disability. Once stricken, competent patients can test and reject that preference. They often do.54 Thus Wilfrid Sheed "quickly learned [that] cancer, even more than polio, has a disarming way of bargaining downward, beginning with your whole estate and then letting you keep the game warden's cottage or badminton court; and by the time it has tried to frighten you to death and threatened to take away your very existence, you'd be amazed at how little you're willing to settle for."55 At least sixteen studies have investigated the stability of people's preferences for life-sustaining treatment.56 A meta-analysis of eleven of these studies found that the stability of patients' preferences was 71 percent (the range was 57 percent to 89 percent).57 Although stability depended on numerous factors (including the illness, the treatment, and demographic variables), the bottom line is that, over periods as short as two years, almost one-third of preferences for life-sustaining medical treatment changed. More particularly, illness and hospitalization change people's preferences for life-sustaining treatments.58 In a prospective study, the desire for life-sustaining treatment declined significantly after hospitalization but returned almost to its original level three to six months later.59 Another study concluded that the "will to live is highly unstable among terminally ill cancer patients."6° The authors thought their findings "perhaps not surprising, given that only 10-14% of individuals who survive a suicide attempt commit suicide during the next 10 years, which suggests that a desire to die is inherently changeable." The consistent finding that interest in lifesustaining treatment shifts over time and across contexts coincides tellingly with research charting people's struggles to predict their own tastes, behavior, and emotions even over short periods and under familiar circumstances.61 People mispredict what poster they will like,62 how much they will buy at the grocery store,' how sublimely they will enjoy an ice cream,64 and how they will adjust to tenure decisions.65 And people "miswant" for numerous reasons.66 They imagine a different event from the one that actually occurs, nurture inaccurate theories about what gives them pleasure,67 forget they might outwit misery, concentrate on salient negative events and ignore offsetting happier ones,68 and misgauge the effect of physiological sensations like pain.69 Given this rich stew of research on people's

missteps in predicting their tastes generally, we should expect misapprehensions about end-of-life preferences. Indeed, those preferences should be especially volatile, since people lack experience deciding to die.

Can People Articulate What They Want? Suppose, arguendo, that patients regularly made sound choices about future treatments and write living wills. Can they articulate their choices accurately? This question is crucially unrealistic, of course, because the assumption is false. People have trouble reaching well-considered decisions, and you cannot state clearly on paper what is muddled in your mind. And indeed people do, for instance, issue mutually inconsistent instructions in living wills.70 But assume this difficulty away and the problem of articulation persists. In one sense, the best way to divine patients' preferences is to have them write their own living wills to give surrogates the patient's gloriously unmediated voice. This is not a practical policy. Too many people are functionally illiterate,71 and most of the literate cannot express themselves clearly in writing. It's hard, even for the expert writer. Furthermore, most people know too little about their choices to cover all the relevant subjects. Hence living wills are generally forms that demand little writing. But the forms have failed. For example, "several studies suggest that even those patients who have completed AD forms . . . may not fully understand the function of the form or its language."72 Living wills routinely baffle patients with their "syntactic complexity, concept density, abstractness, organization, coherence, sequence of ideas, page format, length of line of print, length of paragraph, punctuation, illustrations, color, and reader interest." Unfortunately, most advance directive forms . . . often have neither a reasonable scope nor depth. They do not ask all the right questions and they do not ask those questions in a manner that elicits clear responses.73 Doctors and lawyers who believe their clients are all above average should ask them what their living will says. One of us (CES) has tried the experiment. The modal answer is, in its entirety: "It says I don't want to be a vegetable." No doubt the forms could b e improved, but not enough to matter. The world abounds in dreadfully drafted forms because writing complex instructions for the future is crushingly difficult. Statutes read horribly because their authors are struggling to (1) work out

II

.11

0

Jill

III Vil

2 Medical Decision-Making

exactly what rule they want, (2) imagine all the circumstances in which it might apply, and (3) find language to specify all those but only those circumstances. Each task is ultimately impossible, which is why statutes explicitly or implicitly confide their enforcers with some discretion and why courts must interpret—rewrite?— statutes. However, these skills and resources are not available to physicians or surrogates. One might retort that property wills work and that living wills are not that far removed from property wills. But wills work as well as they do to distribute property because their scope is—compared to living wills—narrow and routinized. Most people have little property to distribute and few plausible heirs. As property accumulates and ambitions swell, problems proliferate. Many of them are resolvable because experts—lawyers--exclusively draft and interpret wills. Lawyers have been experimenting for centuries with testamentary language in a process which has produced standard formulas with predictable meanings and standard ways of distributing property into which testators are channeled. Finally, if testators didn't say it clearly enough in the right words and following the right procedures, courts coolly ignore their wishes and substitute default rules. The lamentable history of the living will demonstrates just how recalcitrant these problems are. There have been, essentially, three generations of living wills. At first, they stated fatuously general desires in absurdly general terms. As the vacuity of over-generality became clear, advocates of living wills did the obvious: Were living wills too general? Make them specific. Were they "one size fits all"? Make them elaborate questionnaires. Were they uncritically signed? "Require" probing discussions between doctor and patient. However, the demand for specificity forced patients to address more questions than they could comprehend. So, generalities were insufficiently specific and insufficiently considered. Specifics were insufficiently general and perhaps still insufficiently considered. What was a doctor—or lawyer—to do? Behold the "values history," a disquisition on the patient's supposed overarching beliefs from which to infer answers to specific questions.74 That patients can be induced to trek through these interminable and imponderable documents is unproved and unlikely. That useful conclusions can be drawn from the platitudes they evoke is false. As Justice Holmes knew, "General propositions do not decide concrete cases."75 The lessons of this story are that drafting instructions is harder than proponents of living wills seem to believe and that when you move toward one blessing in structuring these documents, you walk away from

another. The failure to devise workable forms is not a failure of effort or intelligence. It is a consequence of attempting the impossible. Where Is the Living Will? Suppose that, mirabile dictu, people executed living wills, knew what they will want, and could say it. That will not matter unless the living will reaches the people responsible for the incompetent patient. Often, it does not. This should be no surprise, for long can be the road from the drafter's chair to the icu bed. First, the living will may be signed years before it is used, and its existence and location may vanish in the mists of time." Roughly half of all living wills are drawn up by lawyers and must somehow reach the hospital, and 62 percent of patients do not give their living will to their physician." On admission to the hospital, patients can be too assailed and anxious to recall and mention their advance directives.78 Admission clerks can be harried, neglectful, and loath to ask patients awkward questions. Thus when a team of researchers reviewed the charts of 182 patients who had completed a living will before being hospitalized, they found that only 26 percent of the charts accurately recorded information about those directives79 and only 16 percent of the charts contained the form. And in another study only 35 percent of the nursing home patients who were transferred to the hospital had their living wills with them.8° Will Proxies Read It Accurately? Suppose, per impossibile, that patients wrote living wills, correctly anticipated their preferences, articulated their desires lucidly, and conveyed their document to its interpreters. How acutely will the interpreters analyze their instructions? Living wills are not self-executing: someone must decide whether the patient is incompetent, whether a medical situation described in the living will has arisen, and what the living will then commands. Usually, the patient's intimates will be central among a living will's interpreters. We might hope that intimates already know the patient's mind so that only modest demands need be made on their interpreting skills. But many studies have asked such surrogates to predict what treatment the patient would choose.81 Across these studies, approximately 70 percent of the predictions were correct—not inspiring success for life and death decisions. Do living wills help? We know of only one study that addresses that question. In a randomized trial,

Fagerlin/Schneider: Enough:The Failure of the Living Will ()

researchers asked elderly patients to complete a diseaseand treatment-based or a value-based living will.82 A control group of elderly patients completed no living will. The surrogates were generally spouses or children who had known the patient for decades. Surrogates who were not able to consult their loved one's living will predicted patients' preferences about 70 percent of the time. Strikingly, surrogates who consulted the living will did no better than surrogates denied it. Nor were surrogates more successful when they discussed living wills with patients just before their prediction. What is more, a similar study found that primary care physicians' predictions were similarly unimproved by providing them with patients' advance directives.83 On the other hand, emergency room doctors (complete strangers) given a living will more accurately predicted patients' preferences than ER doctors without one.84

Do Living Wills Alter Patient Care? Our survey of the mounting empirical evidence shows that none of the five requisites to making living wills successful social policy is met now or is likely to be. The program has failed and indeed is impossible. That impossibility is confirmed by studies of how living wills are implemented, which show that living wills seem not to affect patients' treatments. For instance, one study concluded that living wills "do not influence the level of medical care overall. This finding was manifested in the quantitatively equal use of diagnostic testing, operations, and invasive hemodynamic monitoring among patients with and without advance directives. Hospital and icu lengths of stay, as well as health care costs, were also similar for patients with and without advance directive statements.' Another study found that in thirty of thirty-nine cases in which a patient was incompetent and the living will was in the patient's medical record, the surrogate decision-maker was not the person the patient had appointed.86 In yet a third study, a quarter of the patients received care that was inconsistent with their living will.87 But all this is normal. Harry Truman rightly predicted that his successor would "sit here, and he'll say, `Do this! Do that!' And nothing will happen. Poor Ike— it won't be a bit like the army. He'll find it very frustrating." (Of course, the army isn't like the army either, as Captain Truman surely knew) Indeed, the whole law of bioethics often seems a whited sepulchre for slaughtered hopes, for its policies have repeatedly fallen woefully short of their purposes. Informed consent is a "fairytale."88 Programs to increase organ donation have persistently disappointed. Laws regulating DNR orders are hardly better. Legal definitions of brain death are

misunderstood by astonishing numbers of doctors and nurses. And so on.89 But why don't living wills affect care?9° Joan Teno and colleagues saw no evidence "that a physician unilaterally decided to ignore or disregard an AD." Rather, there was "a complex interaction of . . . three themes." First (as we have emphasized), "the contents of ADs were vague and difficult to apply to current clinical situations." The imprecision of living wills not only stymies interpreters, it exacerbates their natural tendency to read documents in light of their own preferences. Thus "(e)ven with the therapy-specific AD accompanied by designation of a proxy and prior patient— physician discussion, the proportion of physicians who were willing to withhold therapies was quite variable: cardiopulmonary resuscitation, 100%; administration of artificial nutrition and hydration, 82%; administration of antibiotics, 80%; simple tests, 70%; and administration of pain medication, 13%."91 Second, the Teno team found that "patients were not seen as 'absolutely, hopelessly ill,' and thus, it was never considered the time to invoke the AD." Living wills typically operate when patients become terminally ill, but neither doctors nor families lightly conclude patients are dying, especially when that means ending treatment. And understandably. For instance, "on the day before death, the median prognosis for patients with heart failure is still a 50% chance to live 6 more months because patients with heart failure typically die quickly from an unpredictable complication like arrhythmia or infection."92 So by the time doctors and families finally conclude the patient is dying, the patient's condition is already so dire that treatment looks pointless quite apart from any living will. "In all cases in which life-sustaining treatment was withheld or withdrawn, this decision was made after a trial of life-sustaining treatment and at a time when the patient was seen as 'absolutely, hopelessly ill' or 'actively dying' Until patients crossed this threshold, ADs were not seen as applicable." Thus "it is not surprising that our previous research has shown that those with ADs did not differ in timing of DNR orders or patterns of resource utilization from those without ADs."93 Third, "family members or the surrogate designated in a [durable power of attorney] were not available, were ineffectual, or were overwhelmed with their own concerns and did not effectively advocate for the patient." Family members are crucial surrogates because they should be: patients commonly want them to be; they commonly want to be; they specially cherish the patient's interests. Doctors ordinarily assume families know the patient's situation and preferences and may not relish responsibility for life-and-death

II 1111

111111

2 Medical Decision-Making

decisions, and doctors intent on avoiding litigation may realize that the only plausible plaintiffs are families. The family, however, may not direct attention to the advance directive and may not insist on its enforcement. In fact, surrogates may be guided by either their own treatment preferences or an urgent desire to keep their beloved alive.94 In sum, not only are we awash in evidence that the prerequisites for a successful living wills policy are unachievable, but there is direct evidence that living wills regularly fail to have their intended effect. That failure is confirmed by the numerous convincing explanations for it. And if living wills do not affect treatment, they do not work. Do Living Wills Have Beneficial Side Effects?

Even if living wills do not effectively promote patients' autonomy, they might have other benefits that justify their costs. There are three promising candidates. First, living wills might stimulate conversation between doctor and patient about terminal treatment. However, at least one study finds little association between patients' reports of executing an advance directive and their reports of such conversations.95 Nor do these conversations, when they occur, appear satisfactory." James Tulsky and colleagues asked experienced clinicians who had relationships with patients who were over sixty-five or seriously ill to "discuss advance directives in whatever way you think is appropriate" with them. Although the doctors knew they were being taped, the conversations were impressively short and one-sided: The median discussion "lasted 5.6 minutes (range, 0.9 to 15.0 minutes). Physicians spoke for a median of 3.9 minutes (range, 0.6 to 10.9 minutes), and patients spoke for the remaining 1.7 minutes (range, 0.3 to 9.6 minutes). . . . Usually, the conversation ended without any specific follow-up plan." The "[p]atients' personal values, goals for care, and reasons for treatment preferences were discussed in 71% of cases and were explicitly elicited by 34% of physicians." But doctors commonly "did not explore the reasons for patientb preferences and merely determined whether they wanted specific interventions."97 Nor were the conversations conspicuously informative: "Physicians used vague language to describe scenarios, asking what patients would want if they became 'very very sick' or 'had something that was very serious' . . ." Further, "[v]arious qualitative terms were used loosely to describe outcome probabilities." In addition, these brief conversations considered almost exclusively the two ends of the continuum—the most

hopeless and the most hopeful cases. Conversations tended to ignore "the more common, less clear-cut predicaments surrounding end-of-life care." True, the patients all thought "their physicians 'did a good job talking about the issues,— but this only suggests that patients did not understand how little they were told. The second candidate for beneficial side effect arises from evidence that living wills may comfort patients and surrogates. People with a living will apparently gain confidence that their surrogates will understand their preferences and will implement them comfortably, and the surrogates concur.98 Improved satisfaction with decisions was also a rare positive effect of the SUPPORT study (which devoted enormous resources to improving end-of-life decisions and care but made dismayingly little difference).99 In another study, living wills reduced the stress and unhappiness of family members who had recently withdrawn life support from a relative. w° But even if living wills make patients and surrogates more confident and comfortable, those qualities are apparently unrelated to the accuracy of surrogates' decisions. Thus we are left with the irony that one of the best arguments for a tool for enhancing peopleb autonomy is that it deceives them into confidence. Third, because living wills generally constrain treatment, they might reduce the onerous costs of terminal illness. Although several studies associated living wills with small decreases in those costs,' several studies have reached the opposite conclusion.102 The old Scotch verdict, "not proven," seems apt. The Costs

There is no free living will, and the better (or at least more thorough and careful) the living will, the more it costs. Living wills consume patients' time and energy. When doctors or lawyers help, costs soar. On a broader view, Jeremy Sugarman and colleagues estimated that the Patient Self-Determination Act imposed on all hospitals a start-up cost of $101,569,922 and imposed on one hospital (Johns Hopkins) initial costs of $114,528,103. These figures omit the expenses, paid even as we write and you read, of administering the program. And this money has bought only pro forma compliance. These are real costs incurred when over 40 million people lack health insurance and when we are spending more of our gross domestic product on health care than comparable countries without buying commensurately better health. If programs to promote and provide living wills showed signs of achieving the

Fagerlin/Schneider: Enough: The Failure of the Living Will

goals cherished for them, we would have to decide whether their valuable but incalculable rewards exceeded their diffuse but daunting costs. However, since those programs have failed, their costs plainly outweigh their benefits. What Is to Be Done? Living wills attempt what undertakers like to call "pre-need planning," and on inspection they are as otiose as the mortuary version. Critically, empiricists cannot show that advance directives affect care. This is damning, but were it our only evidence, perhaps we might not be weary in well doing: for in due season we might reap, if we faint not. However, our survey of the evidence suggests that living wills fail not for want of effort, or education, or intelligence, or good will but because of stubborn traits of human psychology and persistent features of social organization. Thus, when we reviewed the five conditions for a successful program of living wills, we encountered evidence that not one condition has been achieved or, we think, can be. First, despite the millions of dollars lavished on propaganda, most people do not have living wills. And they often have considered and considerable reasons for their choice. Second, people who sign living wills have generally not thought through its instructions in a way we should want for life-and-death decisions. Nor can we expect people to make thoughtful and stable decisions about so complex a question so far in the future. Third, drafters of living wills have failed to offer people the means to articulate their preferences accurately. And the fault lies primarily not with the drafters; it lies with the inherent impossibility of living wills' task. Fourth, living wills too often do not reach the people actually making decisions for incompetent patients. This is the most remediable of the five problems, but it is remediable only with unsustainable effort and unjustifiable expense. Fifth, living wills seem not to increase the accuracy with which surrogates identify patients' preferences. And the reasons we surveyed when we explained why living wills do not affect patients' care suggest that these problems are insurmountable. The cost-benefit analysis here is simple: If living wills lack detectable benefits, they cannot justify any cost, much less the considerable costs they now exact. Any attempt to increase their incidence and their availability to surrogates must be expensive. And the evidence suggests that broader use of living wills can actually disserve rather than promote patients' autonomy: If, as we have argued, patients sign living wills

without adequate reflection, lack necessary information, and have fluctuating preferences anyway then living wills will not lead surrogates to make the choices patients would have wanted. Thus, as Pope suggests, the "PSDA, rather than promoting autonomy has 'done a disservice to most real patients and their families and caregivers.' It has promoted the execution of uninformed and under-informed advance directives, and has undermined, not protected, self-determination.',104 If living wills have failed, we must say so. We must say so to patients. If we believe our declamations about truth-telling, we should frankly warn patients how faint is the chance that living wills can have their intended effect. More broadly, we should abjure programs intended to cajole everyone into signing living wills. We should also repeal the PSDA, which was passed with arrant and arrogant indifference to its effectiveness and its costs and which today imposes accumulating paperwork and administrative expense for paltry rewards.'°5 Of course we recognize the problems presented by the decisions that must be made for incompetent patients, and our counsel is not wholly negative. Patients anxious to control future medical decisions should be told about durable powers of attorney. These surely do not guarantee patients that their wishes will blossom into fact, but nothing does. What matters is that powers of attorney have advantages over living wills. First, the choices that powers of attorney demand of patients are relatively few, familiar, and simple. Second, a regime of powers of attorney requires little change from current practice, in which family members ordinarily act informally for incompetent patients. Third, powers of attorney probably improve decisions for patients, since surrogates know more at the time of the decision than patients can know in advance. Fourth, powers of attorney are cheap; they require only a simple form easily filled out with little advice. Fifth, powers of attorney can be supplemented by legislation (already in force in some states) akin to statutes of intestacy. These statutes specify who is to act for incompetent patients who have not specified a surrogate. In short, durable powers of attorney are—as these things go—simple, direct, modest, straightforward, and thrifty. In social policy as in medicine, plausible notions can turn out to be bad ideas. Bad ideas should be renounced. Bloodletting once seemed plausible, but when it demonstrably failed, the course of wisdom was to abandon it, not to insist on its virtues and to scrounge for alternative justifications for it. Living wills were praised and peddled before they were fully developed, much less studied. They have now failed repeated tests of practice. It is time to say, "enough."

71

II

8

IN

111111

2 Medical Decision-Making

Notes 1. R. Drescer, "Missing Persons: Legal Perceptions of Incompetent Patients," Rutgers Law Review 46 (1994): 609-695. 2. J. Lynn, 'Why I Don't Have a Living Will," Law, Medicine & Health Care 19, nos. 1-2 (1991): 101-104. 3. C.P. Sabatino, "End-of-Life Legal Trends," ABA Commission on Legal Problems of the Elderly 2 (2000). 4. Health Care Assurance of 2001. S. 26. 107th Congress ed; 2001; The Advance Planning and Compassionate Care Act of 1999. S. 628. 106th Congress ed; 1999. 5. 38 CFR Part 17 RIN. 2900-AJ28. November 2,1998. 6. Knight v. Beverly Health Care. 820 Std 92; 2001. 7. See Conservatorship of Wendland, where the California Supreme Court construed the state's Health Care Decisions Law as "requiring clear and convincing evidence of a conscious conservatee's wish to justify withholding lifesustaining treatment" but held that decision did not affect patients who had left "formal directions for health care." 28 P.3d 151; 2001. 8. In re Martin. 538 NW2d 399; Mich 1995. 9. Council on Ethical and Judicial Affairs of the American Medical Association, Surrogate Decision Making E8.081. http:// www.ama-assn.org. 10. P.J. Aitken, "IncorporatingAdvance Care Planning into Family Practice," American Family Physician," 59 (1999): 605-620; A.O. Calvin and A.P. Clark, "How Are You Facilitating Advance Directives in Your Clinical Nurse Specialist Practice?' Clinical Nur se Specialist 16, no. 6 (2002): 292-294. 11. A document that "1,0 ives person responsible for making medical decisions greater information, specificity and insight about your specific health-care related decisions, wishes, and objectives" is "A MUST FOR NEARLY EVERYONE" (P.A. Meints, "A Trust and Estate Planning Questionnaire for Families with Minor Children," The Practical Tax Lawyer 16, no. 1, 12001]: 33). Providing living wills has also become a pro bono activity "Wills on Wheels was established by a committee of paralegals and consulting attorneys determined to provide . . . low-income adults with simple wills and living wills" G.M. Price, "pro Bono and Paralegals: Helping to Make a Difference," Colorado Lawyer (September 30, 2000), 55-56. 12. See www.aarp.org/confacts/programs/endoflife.html. 13. The form's critical paragraph reads: "My desires concerning medical treatment are-." Then it leaves fourteen blank lines the patient may fill in. Available at www.med.umich .edu/llibr/aha/umlega104.htm. 14. R. Dresser, "Precommitment: A Misguided Strategy for Securing Death with Dignity," Texas Law Review 81 (2003): 1823-1847. 15. A.R. Eiser and M.D. Weiss, "The Underachieving Advance Directive: Recommendations for Increasing Advance Directive Completion," American Journal of Bioethics 1 (2001): 1-5. 16. N.L. Cantor, "Twenty-Five Years after Quinlan: A Review of the Jurisprudence of Death and Dying," Journal of Law, Medicine & Ethics 29 (2001): 182-196.

17. L. Emanuel, "Living Wills Can Help Doctors and Patients Talk about Dying," Western Journal of Medicine 173 (2000): 368. 18. For example, the form provided by a consortium of the American Bar Association, the American Medical Association, and the American Association of Retired Persons "combines and expands the traditional Living Will and Health Care Power of Attorney into a single, comprehensive document" (http://www.ama-assn.org/ public/booklets/livgwill.htm). 19. D.M. High, "Why Are Elderly People Not Using Advance Directives?" Journal of Aging and Health 5, no. 4 (1993): 497-515. 20. L.L. Emanuel, "Advance Directives for Medical Care; Reply," NEJM 325 (1991): 1256; N.L. Cantor, "Making Advance Directives Meaningful," Psychology, Public Policy, and Law 4, no. 3 (1998): 629-652; D.M. Cox and G.A. Sachs, "Advance Directives and the Patient Self-Determination Act," Clinics in Geriatric Medicine 10 (1994): 431-443; G.A.D. Havens, "Differences in the Execution/Nonexecution of Advance Directives by Community Dwelling Adults," Research in Nursing and Health 23 (2000): 319-333; D.M. High, "Advance Directives and the Elderly: A Study of Intervention Strategies to Increase Use," Gerontologist 33, no. 3 (1993): 342-349; S.H. Miles, R. Koepp, and E.P. Weber, "Advance End-of-Life Treatment Planning: A Research Review," Archives of Internal Medicine 156, no. 10 (1996): 10621068; S.R. Steiber, "Right to Die: Public Balks at Deciding for Others," Hospitals 61 (1987): 572; J. Teno et al., "Do Advance Directives Provide Instructions that Direct Care? SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment," Journal of the American Geriatrics Society 45, no. 4 (1997): 508-512. 21. Emanuel, "Advance Directives for Medical Care; Reply" 22. Miles, Koepp, and Weber, "Advance End-of-Life Treatment Planning"; J.L. Holley et al., "Factors Influencing Dialysis Patients' Completion of Advance Directives," American Journal of Kidney Diseases 30, no. 3 (1997): 356-360; L.C. Hanson and E. Rodgman, "The Use of Living Wills at the End of Life: A National Study," Archives of Internal Medicine 156, no. 9 (1996): 1018-1022; Teno et al., "Do Advance Directives Provide Instructions that Direct Care?" 23. Holley et al., "Factors Influencing Dialysis Patients' Completion of Advance Directives." 24. Cox and Sachs, "Advance Directives and the Patient SelfDetermination Act"; Miles, Koepp, and Weber, "Advance End-of-Life Treatment Planning"; D.M. High, "All in the Family: Extended Autonomy and Expectations in Surrogate Health Care Decision-Making," Gerontologist 28 (suppl) (1988): 46-51. 25. L.L. Emanuel and E. J. Emanuel, "The Medical Directive: A New Comprehensive Advance Care Document,"JAMA 261 (1989): 3288-3293.

Fagerlin/Schneider: Enough:The Failure of the Living Will

26. High, "Advance Directives and the Elderly"; J.M. Roe et al., "Durable Power of Attorney for Health Care: A Survey of Senior Center Participants," Archives of Internal Medicine 152 (1992): 292-296. 27. High, "Why Are Elderly People Not Using Advance Directives?"; Roe et al., "Durable Power of Attorney for Health Care." 28. High, "Why Are Elderly People Not Using Advance Directives?"; Roe et al., "Durable Power of Attorney for Health Care"; E.J. Emanuel, L.L. Emanuel, and D. Orentlicher, "Advance Directives," JAMA 266 (1991): 2563-63; G.A. Sachs, C.B. Stocking, and S.H. Miles, "Empowerment of the Older Patient? A Randomized, Controlled Trial to Increase Discussion and Use of Advance Directives," Journal of the American Geriatrics Society 40, no. 3 (1992): 269-273; L.L. Brunetti, S.D. Carperos, and R.E. Westlund, "Physicians' Attitudes towards Living Wills and Cardiopulmonary Resuscitation," Journal of General Internal Medicine 6 (1991): 323-329; T.E. Finucane et al., "Planning with Elderly Outpatients for Contingencies of Severe Illness: A Survey and Clinical Trial," Journal of General Internal Medicine 3, no. 4 (1988): 322-325; B. Lo, G.A. McLeod, and G. Saika, "Patient Attitudes to Discussing Life-Sustaining Treatment," Archives of Internal Medicine 146, no. 8 (1986): 1613-1615; R. Yamada et al., "A Multimedia Intervention on Cardiopulmonary Resuscitation and Advance Directives," Journal of General Internal Medicine 14 (1999): 559-563. 29. Cox and Sachs, "Advance Directives and the Patient Self-Determination Act"; L.L. Emanuel and E Emanuel, "Advance Directives," Annals of Internal Medicine 116 (1992): 348-349; B.B. Ott, "Advance Directives: The Emerging Body of Research," AmericanJournal of Critical Care 8 (1999): 514-519. 30. J. Sugarman, M. Weinberger, and G. Samsa, "Factors Associated with Veterans' Decisions about Living Wills," Archives of Internal Medicine 152 (1992): 343-347. 31. Cox and Sachs, "Advance Directives and the Patient SelfDetermination Act"; Holley et al., "Factors Influencing Dialysis Patients' Completion of Advance Directives," High, "All in the Family"; Roe et al., "Durable Power of Attorney for Health Care"; Ott, "Advance Directives"; N.A. Hawkins et al., "Do Patients Want to Micro-manage Their Own Deaths? Process Preferences, Values and Goals in End-ofLife Medical Decision Making," Unpublished manuscript; P.B. Terry et al., "End-of-Life Decision Making: When Patients and Surrogates Disagree," Journal of Clinical Ethics 10, no. 4 (1999): 286-293. 32. J. Carrese and L. Rhodes, "Western Bioethics on the Navajo Reservation: Benefit or Harm?" JAMA 274 (1995): 826-829; L. J. Blackhall et al., "Ethnicity and Attitudes toward Patient Autonomy," JAMA 274 (1995): 820-825. 33. C.M. Puchalski et al., Patients Who Want Their Family and Physician to Make Resuscitation Decisions for Them: Observations from SUPPORT and HELP, JAGS 48 (2000): S84. 34. High, "Why Are Elderly People Not Using Advance Directives?"

35. Ibid. 36. Patient Self-Determination Act of 1990, of the Omnibus Reconciliation Act of 1990. 37. J.L. Yates and H.R. Glick, "The Failed Patient SelfDetermination Act and Policy Alternatives for the Right to Die,"Journal of Aging and Social Policy 29 (1997): 29,31. 38. M.T. Pope, "The Maladaptation of Miranda to Advance Directives: A Critique of the Implementation of the Patient Self-Determination Act," Health Matrix 9 (1999): 139. 39. Cox and Sachs, "Advance Directives and the Patient SelfDetermination Act." 40. J. Hare and C. Nelson, "Will Outpatients Complete Living Wills? A Comparison of Two Interventions," Journal of General Internal Medicine 6 (1991): 41-46. 41. Yamada et al., "A Multimedia Intervention on Cardiopulmonary Resuscitation and Advance Directives"; G.A. Sachs, S.H. Miles, and R.A. Levin, "Limiting Resuscitation: Emerging Policy in the Emergency Medical System," Annals of Internal Medicine 114 (1991): 151-154. 42. C.E. Schneider, The Practice of Autonomy: Patients, Doctors, and Medical Decisions (New York Oxford University Press, 1998). 43. M J. Silveira et al., "Patient's Knowledge of Options at the End of Life Ignorance in the Face of Death," JAMA 284 (2000): 2483,2486-2487. 44. Yamada et al., "A Multimedia Intervention on Cardiopulmonary Resuscitation and Advance Directives"; S.H. Miles, "Advanced Directives to Limit Treatment: The Need for Portability," Journal of the American Geriatrics Society 35, no. 1 (1987): 74-76; K.M. Coppola et al., "Perceived Benefits and Burdens of Life-Sustaining Treatments: Differences among Elderly Adults, Physicians, and Young Adults," Journal of Ethics, Law, and Aging 4, no. 1 (1998): 3-13. 45. Roe et al., "Durable Power of Attorney for Health Care." 46. J.A. Tulsky et al., "Opening the Black Box: How Do Physicians Communicate about Advance Directives?" Annals of Internal Medicine 129 (1998): 441,444. 47. C.E. Schneider and M. Farrell, Information, Decisions, and the Limits of Informed Consent (New York: Oxford University Press, 2000). 48. B.J. McNeil et al., "On the Elicitation of Preferences for Alternative Therapies," NE JM 306 (1982): 1259-1262. 49. T.R. Malloy et al., "The Influence of Treatment Descriptions on Advance Medical Directive Decisions," Journal of the American Geriatrics Society 40, no. 12 (1992): 1255-1260; D.J. Mazur and D.H. Hickman, "Patient Preferences: Survival versus Quality-of-Life Considerations,"Journal of General Internal Medicine 8, no. 7 (1993): 374-377; D.J. Mazur and J.F. Merz, "How Age, Outcome Severity, and Scale Influence General Medicine Clinic Patients' Interpretations of Verbal Probability Terms" (See comments), Journal of General Internal Medicine 9 (1994): 268-271. 50. Miles, Koepp, and Weber, "Advance End-of-Life Treatment Planning." 51. Ott, "Advance Directives," pp. 514,517.

8

uu

8

I

Al

2 Medical Decision-Making

52. J.J. Christensen-Szalanski, "Discount Functions and the Measurement of Patients' Values: Women's Decisions during Childbirth," Medical Decision Making 4, no. 1 (1984): 47-58. 53. R.M. Gready et al., "Actual and Perceived Stability of Preferences for Life-Sustaining Treatment,"Journal of Clinical Ethics 11, no. 4 (2000): 334-346. 54. A. Upadya et al., "Patient, Physician, and Family Member Understanding of Living Wills," American Journal of Respiratory and Critical Care Medicine 166 (2002): 1433. 55. W. Sheed, In Love with Daylight: A Memoir of Recovery (New York: Simon and Schuster, 1995): 14. 56. Gready et al., "Actual and Perceived Stability of Preferences for Life-Sustaining Treatment"; J.T. Berger and D. Majerovitz, "Stability of Preferences for Treatment among Nursing Home Residents," Gerontologist 28, no. 2 (1998): 217-223; S. Carmel and E Mutran, "Stability of Elderly Persons' Expressed Preferences Regarding the Use of Life-Sustaining Treatments," Social Science and Medicine 49, no. 3 (1999): 303-311; M. Danis et al., "Stability of Choices about Life-Sustaining Treatments," Annals of Internal Medicine 120, no. 7 (1994): 567-573; P.H. Ditto et al., "A Prospective Study of the Effects of Hospitalization on Life-Sustaining Treatment Preferences: Context Changes Choices," Unpublished manuscript; P.H. Ditto et al., "The Stability of Older Adults' Preferences for Life-Sustaining Medical Treatment," Unpublished manuscript; E.J. Emanuel, "Commentary on Discussions about Life-Sustaining Treatments,"Journal of Clinical Ethics 5, no. 3 (1994): 250-251; L.L. Emanuel et al., "Advance Directives: Stability of Patients' Treatment Choices," Archives of Internal Medicine 154 (1994): 209-217; M.A. Everhart and R.A. Pearlman, "Stability of Patient Preferences Regarding Life-Sustaining Treatments," Chest 97 (1990): 159-164; L. Ganzini et al., "The Effect of Depression Treatment on Elderly Patients' Preferences for Life-Sustaining Medical Therapy," American Journal of Psychiatry 151, no. 11 (1994): 1631-1636; N. Kohut et al., "Stability of Treatment Preferences: Although Most Preferences Do Not Change, Most People Change Some of Their Preferences," Journal of Clinical Ethics 8, no. 2 (1997): 124-135; M.D. Silverstein et al., "Amyotrophic Lateral Sclerosis and Life-Sustaining Therapy: Patients' Desires for Information, Participation in Decision Making, and Life-Sustaining Therapy," Mayo Clinic Proceedings 66 (1991): 906-913; J.S. Weissman et al., "The Stability of Preferences for Life-Sustaining Care among Persons with AIDS in the Boston Health Study," Medical Decision Making 19 (1999): 16-26; K.M. Coppola et al., "Are LifeSustaining Treatment Preferences Stable over Time? An Analysis of the Literature," Unpublished manuscript. 57. Coppola et al., "Are Life-Sustaining Treatment Preferences Stable over Time?" 58. Danis et al., "Stability of Choices about Life-Sustaining Treatments"; Ditto et al., "A Prospective Study of the Effects of Hospitalization"; Weissman et al., "The Stability of Preferences for Life-Sustaining Care."

59. Ditto et al., "A Prospective Study of the Effects of Hospitalization." 60. H.M. Chochinov et al., "Will to Live in the Terminally Ill," Lancet 354 (1999): 816,818. 61. D.T. Gilbert and T.D. Wilson, "Miswanting: Some Problems in the Forecasting of Future Affective States," in

Feeling and Thinking: The Role of Affect in Social Cognition,

62. 63. 64.

65. 66. 67.

68.

69. 70.

71.

72.

73. 74.

75. 76.

77.

ed. J.P. Forgas (New York: Cambridge University Press, 2000): 178-197; C.H. Griffith 3rd et al., "Knowledge and Experience with Alzheimer's Disease: Relationship to Resuscitation Preference," Archives of Family Medicine 4, no. 9 (1995): 780-784; T.M. Osberg and J.S. Shrauger, "Self-Prediction: Exploring the Parameters of Accuracy," Journal of Personality and Social Psychology 51, no. 5 (1986): 1044-1057. Griffith 3rd et al., "Knowledge and Experience with Alzheimer's Disease." Gilbert and Wilson, "Miswanting." D. Kahneman and J. Snell, "Predicting a Changing Taste: Do People Know What They Will Like?"Journal of Behavioral Decision Making 5, no. 3 (1992): 187-200. Gilbert and Wilson, "Miswanting." Ibid. G. Loewenstein and D. Schkade, "Wouldn't It Be Nice? Predicting Future Feelings," in Hedonic Psychology: Scientific Approaches to Enjoyment, Suffering and Wellbeing, ed. N. Schwartz and D. Kahneman (New York: Russell Sage Foundation, 1997). D. Schkade, "Does Living in California Make People Happy? A Focusing Illusion in Judgements of Life Satisfaction," Psychological Science 9 (1998): 340-346. Loewenstein and Schkade, "Wouldn't It Be Nice?" A.S. Brett, "Limitations of Listing Specific Medical Interventions in Advance Directives," JAMA 266 (1991): 825-828. I.S. Kirsch et al., Adult Literacy in America: A First Look at the Results of the National Adult Literacy Survey, US Department of Education, August 1993, NCES 93275. Cox and Sachs, "Advance Directives and the Patient SelfDetermination Act"; Miles, Koepp, and Weber, "Advance End-of-Life Treatment Planning"; Silveira et al., "Patient Knowledge of Options at the End of Life"; Coppola et al., "Perceived Benefits and Burdens of Life-Sustaining Treatments." Pope, "The Maladaptation of Miranda to Advance Directives," pp. 139,165-166. D.J. Doukas and L.B. McCullough, "The Values History: The Evaluation of the Patients Values and Advance Directives," Journal of Family Practice 32, no. 2 (1991): 145-153. Lochner v. New York N. 198 U.S. 45: Supreme Court of the United States; 1905. HI Silverman et al., "Implementation of the Patient Self-Determination Act in a Hospital Setting: An Initial Evaluation," Archives of Internal Medicine 155, no. 5 (1995): 502-510. Roe et al., "Durable Power of Attorney for Health Care."

Fagerlin/Schneider: Enough:The Failure ofthe Living Will ®

78. R.S. Morrison et al., "The Inaccessibility of Advance Directives on Transfer from Ambulatory to Acute Care Settings," JAMA 274 (1995): 478-482. 79. Ibid. 80. M. Danis et al., "A Prospective Study of the Impact of Patient Preferences on Life-Sustaining Treatment and Hospital Cost," Critical Care Medicine 24,11 (1996): 1811-1817. 81. J.A. Druley et al., "Physicians' Predictions of Elderly Outpatients' Preferences for Life-Sustaining Treatment," Journal of Family Practice 37 (1993): 469-475; J. Hare, C. Pratt, and C. Nelson, "Agreement between Patients and Their Self-Selected Surrogates on Difficult Medical Decisions," Archives of Internal Medicine 152, no. 5 (1992): 1049-1054; P.M. Layde et al., "Surrogates' Predictions of Seriously Ill Patients' Resuscitation Preferences," Archives of Family Medicine 4, no. 6 (1995): 518-523; J.G. Ouslander, A. J. Tymchuk, and B. Rahbar, "Health Care Decisions among Elderly Long-Term Care Residents and Their Potential Proxies," Archives of Internal Medicine 149, no. 6 (1989): 1367-1372; A.B. Seckler et al., "Substituted Judgment: How Accurate Are Proxy Predictions?" Annals of Internal Medicine 115 (1991): 92-98; D.P. Sulmasy et al., "The Accuracy of Substituted Judgments in Patients with Terminal Diagnoses," Annals of Internal Medicine 128, no. 8 (1998): 621-629; R.F. Uhlmann, R.A. Pearlman, and K.C. Cain, "Physicians' and Spouses' Predictions of Elderly Patients' Resuscitation Preferences," Journal of Gerontology 43, no. 5 (1988): M115-M121; R.F. Uhlmann, RA. Pearlman, and K.C. Cain, "Understanding of Elderly Patients' Resuscitation Preferences by Physicians and Nurses," Western Journal of Medicine 150 (1989): 705-707; N.R. Zweibel and C.K. Cassell, 'Treatment Choices at the End of Life: A Comparison of Decisions by Older Patients and Their Physician-Selected Proxies," Gerontologist 29, no. 5 (1989): 615-621. 82. L. Emanuel, "The Health Care Directive: Learning How to Draft Advance Care Documents," Journal of the American Geriatrics Society 39, no. 12 (1991): 1221-1228; P.H. Ditto et al., "Fates Worse Than Death: The Role of Valued Life Activities in Health-State Evaluations," Health Psychology 15, no. 5 (1996): 332-343. 83. K.M. Coppola et al., "Accuracy of Primary Care and Hospital-Based Physicians Predictions of Elderly Outpatients' Treatment Preferences with and without Advance Directives," Archives of Internal Medicine 161, no. 3 (2001): 431-440. 84. Ibid. 85. M.D. Goodman, M. Tamoff, and G.J. Slotman, "Effect of Advance Directives on the Management of Elderly Critically Ill Patients," Critical Care Medicine 26, no. 4 (1998): 701-704. 86. Morrison et al., "The Inaccessibility of Advance Directives." 87. M. Danis and J.M. Garrett, "Advance Directives for Medical Care: Reply," NEJM 325 (1991). 88. J. Katz, "Informed Consent-A Fairy Tale? Laws Vision," University of Pittsburgh Law Review 39, no. 2 (1977): 137-174;

89.

90. 91.

92. 93.

94.

95.

96.

97. 98.

99.

100.

101. 102.

103.

104. 105.

C.H. Braddock 3rd et al., "Informed Decision Making in Outpatient Practice: Time to Get Back to Basics," JAMA 282, no. 24 (1999): 2313-2320. C.E. Schneider, 'The Best-Laid Plans," Hastings Center Report 30, no. 4 (2000): 24-25; C.E. Schneider, "Gang Aft Agley," Hastings Center Report 31, no. 1 (2001): 27-28. Teno et al., "Do Advance Directives Provide Instructions That Direct Care?" W.R. Mower and L.J. Baraff,"Advance Directives: Effect of Type of Directive on Physicians' Therapeutic Decisions," Archives of Internal Medicine 153 (1993): 375,378. J. Lynn, "Learning to Care for People with Chronic Illness Facing the End of Life," JAMA 284 (2000): 2508-2509. J. Teno et al., "The Illusion of End-of-Life Resource Savings with Advance Directives SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatment," Journal of the American Geriatrics Society 45, no. 4 (1997): 513-518. A. Fagerlin et al., "Projection in Surrogate Decisions about Life-Sustaining Medical Treatments," Health Psychology 20, no. 3 (2001): 166-175. J. Virmani, U. Schneiderman, and R.M. Kaplan, "Relationship of Advance Directives to Physician-Patient Communication," Archives of Internal Medicine 154 (1994): 909-913. J.A. Tulsky, M.A. Chesney, B. Lo, "How Do Medical Residents Discuss Resuscitation with Patients?"Journal of General Internal Medicine 10, no. 8 (1995): 436-442. Tulsky et al., "Opening the Black Box," pp. 441,445. P.H. Ditto et al., "Advance Directives as Acts of Communication: A Randomized Controlled Trial, Archives of Internal Medicine 161, no. 3 (2001): 421-430. R. Baker et al., "Family Satisfaction with End-of-Life Care in Seriously Ill Hospitalized Adults," Journal of the American Geriatrics Society 48, no 5 (suppl) (2000): S61-S69. V.P. Tilden et al., "Family Decisionmaking to Withdraw Life-Sustaining Treatments from Hospitalized Patients," Nursing Research 50, no. 2 (2001): 105-115. Miles, Koepp, and Weber, "Advance End-of-Life Treatment Planning." Teno et al., "The Illusion of End-of-Life Resource Savings with Advance Directives"; E.J. Emanuel and L.L. Emanuel, "The Economics of Dying: The Illusion of Cost Savings at the End of Life," NEJM 330 (1994): 540-544; L.J. Schneiderman et al., "Effects of Offering Advance Directives on Medical Treatments and Costs," Annals of Internal Medicine 117, no. 7 (1992): 599-606. J. Sugarman et al., "The Cost of Ethics Legislation: A Look at the Patient Self-Determination Act," Kennedy Institute of Ethics Journal 3, no. 4 (1993): 387-399. Pope, "The Maladaptation of Miranda to Advance Directives," pp. 139,167. Yates and Glick, "The Failed Patient Self-Determination Act"; Sugarman et al., "The Cost of Ethics Legislation."

II

1111

1111ild

2 Medical Decision-Making

Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures Canadian Medical Association

Some people want to specify in advance the types of medical procedures they would or would not want to undergo in the event that they became incompetent. They can fulfill this desire through a written advance directive, or by appointing a proxy decision-maker, or both. Physicians should assist their patients in these endeavours. They should honour a patient's advance directives unless there are reasonable grounds for not doing so. In recent years, patients' concerns over decisionmaking in the medical setting have increasingly focused on advance directives for cardiopulmonary resuscitation, resuscitation in general, and other life-saving or -sustaining measures. The CMA holds that the right to accept or reject any treatment or procedure ultimately resides with the patient or appropriate proxy. This includes the right to accept or ref use resuscitative as well as other life-saving or -sustaining measures should they become medically indicated. Furthermore, under certain circumstances it may be appropriate for a patient to indicate to the physician and other relevant people, by means of an advance directive, whether he or she wants such resuscitative measures taken should the need arise. Patients frequently believe that an advance directive to ref use life-saving or -sustaining measures will be honoured under all circumstances. The reality of medical practice makes this impossible. If an advance directive is specific to a particular set of circumstances the directive will have no force when these circumstances

or ones essentially similar to them do not exist. On the other hand, if an advance directive is so general that it applies to all possible events that could arise it is usually too vague to give any usable direction to the physician. In either case physicians will have to rely on their professional judgment to reach a decision. Implementation

A physician should assist a patient in a consultative capacity in the preparation of an advance directive concerning life-saving or -sustaining measures if the patient requests such assistance. In the course of this consultative process, the physician should try to ensure that the patient understands the limits of such documents. Also, the physician should impress upon the patient the need to make advance directives reasonable and accessible. Any such directive should be in writing. A patient's duly executed advance directive shall be honoured by the attending physician unless there are reasonable grounds to suppose that it no longer represents the wishes of the patient or that the patient's understanding was incomplete at the time the directive was prepared. Some patients may not wish to execute an advance directive but are concerned about who will make health care decisions for them when they are no longer able to do so. Physicians should explore with these patients the possibility of identifying a specific person who will have the legal power to make health care decisions on their behalf in such an eventuality. Physicians whose patients do wish to draw up advance directives should explore with them the possibility of identifying a specific person who will have the legal power to act as their proxy decision-maker should the need arise for clarification of the directive.

Source: From wwwcma.ca/index.cfm/ci_id/33230/la_id/l.htm (1992). Copyright © 1992. This work is protected by copyright and the making of this copy was with the permission of Access Copyright. Any alteration of its content or further copying in any form whatsoever is strictly prohibited unless otherwise permitted by law.

2.5 Best-Interest Judgments

Involving Children in Medical Decisions Christine Harrison, Nuala R. Kenny, Mona Sidarous, and Mary Rowell

Eleven-year-old Samantha is a bright, loving child who was treated for osteosarcoma in her left arm. The arm had to be amputated, and Samantha was given a course of chemotherapy. She has been cancer-free for 18 months

Source: From Canadian Medical Association Journal 156, 6 (1997), 825-8. Copyright © 1997. This work is protected by copyright and the making of this copy was with the permission of Access Copyright. Any alteration of its content or further copying in any form whatsoever is strictly prohibited unless otherwise permitted by law.

Harrison/Kenny/Sidarous/Rowell: Involving Children in Medical Decisions

and is doing well in school. She is self-conscious about her prosthesis and sad because she had to give away her cat, Snowy, to decrease her risk of infection. Recent tests indicate that the cancer has recurred and metastasized to her lungs. Her family is devastated by this news but do not want to give up hope. However, even with aggressive treatment Samantha's chances for recovery are less than 20 per cent. Samantha adamantly refuses further treatment. On earlier occasions she had acquiesced to treatment only to struggle violently when it was administered. She distrusts her health care providers and is angry with them and her parents. She protests, "You already made me give up Snowy and my arm. What more do you want?" Her parents insist that treatment must continue. At the request of her physician, a psychologist and psychiatrist conduct a capacity assessment. They agree that Samantha is probably incapable of making treatment decisions; her understanding of death is immature and her anxiety level very high. Nursing staff are reluctant to impose treatment; in the past, Samantha's struggling and the need to restrain her upset them a great deal. Why Is It Important to Include Children in Medical Decision-Making? Ethics

Traditionally, parents and physicians have made all medical decisions on behalf of children. However, just as the concept of informed consent has developed over the last 30 years with respect to competent adult patients, so new ways of thinking about the role of children in medical decision-making have evolved. Ethical principles that provide guidance in the care of adults are insufficient in the context of caring for childrent-3 Issues related to the voluntariness of consent, the disclosure of information, capacity assessment, treatment decisions, and bereavement are more complex, as is the physician's relationship with the patient and the patient's family.3'4 Adult models presume that the patient is autonomous and has a stable sense of self, established values, and mature cognitive skills; these characteristics are undeveloped or underdeveloped in children. Although it is important to understand and respect the developing autonomy of a child, and although the duty of beneficence provides a starting point for determining what is in the child's best interest, a family-centred ethic is the best model for understanding the interdependent relationships that bear upon the child's situation.5 A family-centred approach considers the effects of a decision on all

family members, their responsibilities toward one another, and the burdens and benefits of a decision for each member, while acknowledging the special vulnerability of the child patient. A family-centred approach presents special challenges for the health care team, particularly when there is disagreement between parent and child. Such a situation raises profound questions about the nature of the physician—patient relationship in pediatric practice. Integrity in this relationship is fundamental to the achievement of the goal of medicine,6 which has been defined as "right and good healing action taken in the interest of a particular patient."' In the care of adults, the physician's primary relationship is with the particular capable patient. The patient family may be involved in decision-making, but it is usually the patient who defines the bounds of such involvement. The care of children, on the other hand, has been described in terms of a "triadic" relationship in which the child, his or her parents, and the physician all have a necessary involvement (Dr Abbyann Lynch, Director, Ethics in Health Care Associates, Toronto: personal communication, 1992). When there is disagreement between parent and child, the physician may experience some moral discomfort in having to deal separately with the child and parent. The assumption that parents best understand what is in the interest of their child is usually sound. However, situations can arise in which the parents' distress prevents them from attending carefully to the child's concerns and wishes. Simply complying with the parents' wishes in such cases is inadequate. It is more helpful to and respectful of the child to affirm the parents' responsibility for the care of their child while allowing the child to exercise choice in a measure appropriate to his or her level of development and experience of illness and treatment. This approach does not discount the parents' concerns and wishes, but recognizes the child as the particular patient to whom the physician has a primary duty of care. This approach seeks to harmonize the values of everyone involved in making the decision.6 Law The legal right to refuse medical treatment is related to, but not identical with, the right to consent to treatment. The patient's right to refuse even life-saving medical treatment is recognized in Canadian law8,9 and is premised on the patient's right to exercise control over his or her own body. Providing treatment despite a patienth valid refusal can constitute battery and, in some circumstances, negligence.

77

II 111

■ ui I

111111

2 Medical Decision-Making

To be legally valid, the refusal of medical treatment must be given by a person deemed capable of making health care choices, that is, capable of understanding the nature and consequences of the recommended treatment, alternative treatments, and non-treatment. In common law the notion of the "mature minor" recognizes that some children are capable of making their own health care choices despite their age.10 In common law and under the statutory law of some provinces patients are presumed capable regardless of age unless shown otherwise; in other provinces an age at which patients are presumed capable is specified." When a child's capacity is in doubt an assessment is required. In the case of children who are incapable of making their own health care decisions, parents or legal guardians generally have the legal authority to act as surrogate decision-makers. The surrogate decision-maker is obliged to make treatment decisions in the best interest of the child. Health care providers who believe that a surrogate's decisions are not in the child's best interest can appeal to provincial child welfare authorities. The courts have the authority to assume a parens patriae role in treatment decisions if the child is deemed to be in need of protection. This issue has arisen most commonly with respect to Jehovah's Witnesses who refuse blood transfusions for their children on religious grounds, and courts have authorized treatment in recognition of the state's interest in protecting the health and well-being of children." Every province has child welfare legislation that sets out the general parameters of the "best interest" standard. Courts are reluctant to authorize the withholding or withdrawal of medical treatment, especially in the face of parental support for such treatment. A special point to consider involves the use of patient restraints. The wrongful or excessive use of restraints could prompt an action of false imprisonment or battery. Restraint can involve the use of force, mechanical means, or chemicals. The use of restraint compromises the dignity and liberty of the patient, including the child patient. Restraints should never be used solely to facilitate care but, rather, only when the patient is likely to cause serious bodily harm to himself or herself or to another. If restraint is required, the health care provider should use the least restrictive means possible, and the need for the restraint (as well as its effect on the patient) should be assessed on an ongoing basis. Policy

The Canadian Paediatric Society has no policy regarding the role of the child patient in medical decision-making The American Academy of Pediatrics statement on this question articulates the joint responsibility of physicians

and parents to make decisions for very young patients in their best interest and states that " [p] arents and physicians should not exclude children and adolescents from decision-making without persuasive reasons.' Empirical Studies

As they grow, children develop decision-making skills, the ability to reason using complex concepts, an understanding of death,14 and the ability to imagine a future for themselves.15 Children with a chronic or terminal illness may have experiences that endow them with insight and maturity beyond their years. Families often encourage children to participate in decision-making. Allowing even young children to make decisions about simple matters facilitates the development of skills that they will need to make more complex decisions later on.' '8 Because tools developed to assess the capacity of adults have not been tested with children, health care professionals working with children should be sensitive to the particular capacity of each child. Children are constantly developing their physical, intellectual, emotional, and personal maturity. Although developmental milestones give us a general sense of capacities, two children of the same age will not necessarily have the same ability to make choices. Even when they are deemed capable of making health care choices, children need support for their decisions from family members and the health care team. How Should I Determine the Appropriate Role of a Child in Medical Decision-Making?

Most children fall into one o f three groups with respect to their appropriate involvement in decision-making.19,20 Infants and Young Children

Preschool children have no significant decision-making capacity and cannot provide their own consent. As surrogate decision-makers, parents should authorize (or refuse authorization) on their child's behalf, basing their decisions on what they believe to be in the child's best interest. Primary-School Children

Children of primary-school age may participate in medical decisions but do not have full decision-making capacity. They may indicate their assent or dissent without fully understanding its implications. Nonetheless they should be provided with information appropriate to their level of comprehension. Although the child's parents should authorize or refuse to authorize treatment, the child's assent should be sought and any strong and sustained dissent should be taken seriously.21

Harrison/Kenny/Sidarous/Rowell: Involving Children in Medical Decisions

Adolescents

Many adolescents have the decision-making capacity of an adult.22.23 This capacity will need to be determined for each patient in light of his or her • • •

•

ability to understand and communicate relevant information; ability to think and choose with some degree of independence; ability to assess the potential for benefits, risks, or harms as well as to consider consequences and multiple options; and achievement of a fairly stable set of values.24

Many children and adolescents, particularly those who have been seriously ill, will need assistance in developing an understanding of the issues and in demonstrating their decision-making capacity. Age-appropriate discussions, perhaps with the assistance of teachers, chaplains, play therapists, nurses, psychologists, or others skilled in communicating with children, are helpful. The child's participation may be facilitated by the use of art activities, stories, poems, role-playing, and other techniques.25,28 Physicians should ensure that good decisions are made on behalf of their child patients. Although the interests of other family members are important and will influence decision-making, the child's interests are most important and are unlikely to be expressed or defended by the child himself or herself. Anxious, stressed, or grieving family members may need assistance in focusing on what is best for the child. This may be especially difficult when a cure is no longer possible; in such cases a decision to stop treatment may seem like a decision to cause the child's death. Whether or not the child participates, the following considerations should bear upon a treatment decision concerning that child: • •

•

the potential benefits to the child; the potential harmful consequences to the child, including physical suffering, psychological or spiritual distress, and death; and the moral, spiritual, and cultural values of the child's family.

The Case

For Samantha, resuming aggressive treatment will have a serious negative effect on her quality of life. The chances of remission are small, yet a decision to discontinue treatment will likely result in her death. Because death is an irreversible harm, and decisions with serious consequences require a high level of competence in decision-making,27 the capacity required would be very high. It has been determined that Samantha does not have this capacity. Nevertheless, Samantha is included in discussions about her treatment options, and her reasons for refusing treatment are explored. 28 Members of the team work hard to re-establish trust. They and Samantha's parents come to agree that refusing treatment is not necessarily unreasonable; a decision by an adult patient in similar circumstances to discontinue treatment would certainly be honoured. Discussions address Samantha's and her parents' hopes and fears, their understanding of the possibility of cure, the meaning for them of the statistics provided by the physicians, Samantha's role in decision-making, and her access to information. They are assisted by nurses, a child psychologist, a psychiatrist, a member of the clergy, a bioethicist, a social worker, and a palliative care specialist. Discussions focus on reaching a common understanding about the goals of treatment for Samantha. Her physician helps her to express her feelings and concerns about the likely effects of continued treatment. Consideration is given to the effects on her physical wellbeing, quality of life, self-esteem, and dignity of imposing treatment against her wishes. Spiritual and psychological support for Samantha and her family is acknowledged to be an essential component of the treatment plan. Opportunities are provided for Samantha and her family to speak to others who have had similar experiences, and staff are given the opportunity to voice their concerns. Ultimately, a decision is reached to discontinue chemotherapy and the goal of treatment shifts from "cure" to "care." Samantha's caregivers assure her and her family that they are not "giving up" but are directing their efforts toward Samantha's physical comfort and her spiritual and psychological needs. Samantha returns home, supported by a community palliative care program, and is allowed to have a new kitten. She dies peacefully.

Notes 1.

Ruddick, W. 1979. "Parents and Life Prospects" in Having Children: Philosophical and Legal Reflections on Parenthood, ed. 0. O'Neill and W. Ruddick. Oxford University Press: New York. 124.

2. 3.

Nelson, J.L. 1992. "Taking Families Seriously" in Hastings Center Report 22: 6. Hardwig, J. 1990. "What about the Family?," in Hastings Center Report 20 (2): 5-10.

I I ;ill

111111

C

I 2 Medical Decision-Making

4. Leikin, S. 1989. "A Proposal Concerning Decisions to Forgo Life-Sustaining Treatment for Young People," in Journal of Pediatrics 115: 17-22. 5. Mahowald, M. 1993. Women and Children in Health Care. Oxford University Press: New York. 187,189. 6. Hellmann, J. 1996. "In Pursuit of Harmonized Values: Patient/Parent-Pediatrician Relationships" in The "Good" Pediatrician: An Ethics Curriculum for Use in Canadian Pediatrics Residency Programs, ed. A. Lynch. Pediatric Ethics Network: Toronto. 7. Pellegrino, E.D. 1979. 'Toward a Reconstruction of Medical Morality. The Primacy of the Act of Profession and the Fact of Illness," in Journal of Medical Philosophy 4: 47. 8. Malette v. Shulman 119901,67 DLR (4th) (Ont CA). 9. Art 11 CCQ. 10. Rozovsky, L.E., and Rozovsky, F.A. 1992. The Canadian Law of Consent to Treatment. Butterworths: Toronto. 53-7. 11. Etchells, E., Sharpe, G., Elliott, C., and Singer, PA. 1996. "Bioethics for Clinicians 3: Capacity," in Canadian Medical Association Journal 155: 657-61. 12. R.B. v. Children's Aid Society of Metropolitan Toronto, (1995] 1 SCR 315 (SCC). 13. American Academy of Pediatrics. 1995. "Informed Consent, Parental Permission and Assent in Pediatric Practice," in Pediatrics 95: 314-17. 14. Matthews, G.R. "Children's Conceptions of Illness and Death," in Children and Health Care: Moral and Social Issues, ed. L.M. Kopelman and J.C. Moskop. Kluwer Academic Publishers: Dordrecht (Holland). 133-46. 15. Koocher, G.P, and DeMaso, D.R. 1990. "Children's Competence to Consent to Medical Procedures," in Pediatrician 17: 68-73. 16. King, N.M.P, and Cross, A.W. 1989. "Children as Decision-Makers: Guidelines for Pediatricians," in Journal of Pediatrics 115: 10-16.

17. Lewis, M.A., and Lewis, C.E. 1990. "Consequences of Empowering Children to Care for Themselves," in Pediatrician 17: 63-7. 18. Yoos, H.L. 1994. "Children's Illness Concepts: Old and New Paradigms," in Pediatric Nursing 20: 134-45. 19. Broome, M.E., and Stieglitz, K.A. 1992. "The Consent Process and Children," in Research in Nursing and Health 15: 147-52. 20. Erlen, J.A. 1987. "The Child's Choice: An Essential Component in Treatment Decisions," in Child Health Care 15: 156-60. 21. Baylis, F 1993. "The Moral Weight of a Child's Dissent," in Ethics in Medical Practice 3 (1): 2-3. 22. Weithorn, L.A., and Campbell, S.B. 1982. "The Competency of Children and Adolescents to Make Informed Treatment Decisions," in Child Development 53: 1589-98. 23. Lewis, C.C. 1981. "How Adolescents Approach Decisions: Changes over Grades Seven to Twelve and Policy Implications," in Child Development 52: 538-44. 24. Brock, D.W. 1989. "Children's Competence for Health Care Decision-Making," in Children and Health Care: Moral and Social Issues, ed. L.M. Kopelman and J.C. Moskop. Kluwer Academic Publishers: Dordrecht (Holland). 181-212. 25. Adams, PL., and Fras, I. 1988. Beginning Child Psychiatry. Bruner/Mazel: New York. 26. Kestenbaum, C.J., and Williams, D., eds. 1988. Handbook of Clinical Assessment of Children and Adolescents. University Press: New York 27. Drane, J.F. 1985. "The Many Faces of Competency," in Hastings Center Report 15 (2): 17-21. 28. Freyer, D.R. 1992. "Children with Cancer: Special Considerations in the Discontinuation of Life-Sustaining Treatment," in Medical and Pediatric Oncology 20: 136-42.

Position Statement: Treatment Decisions Regarding Infants, Children, and Adolescents

may occur when there is consensus that there is a high degree of probability that:

Christine Harrison; Canadian Paediatric Society, Bioethics Committee

[...] Withholding or Withdrawing Treatment

Some of the principal goals of health care are to maintain life and prevent pain and suffering and not to unthinkingly prolong the dying process. Exceptions to the general duty to provide life-sustaining treatment

• • •

•

there is irreversible progression to imminent death; treatment is clearly ineffective or harmful; life will be severely shortened regardless of treatment, and the limitation or withdrawal of interventions will allow greater palliative and comfort care; lives will be filled with intolerable distress and suffering that cannot be prevented or alleviated.

To withhold or withdraw life-sustaining treatment may be ethically appropriate when it is clear that this treatment will not benefit the child or adolescent. Such

Source: From Paediatric Child Health 9 (2004) no. 2,99-103. Posted Feb 1, 2004; reaffirmed Feb 1,2016. Reprinted with permission of the Canadian Paediatric Society (CPS).

Harrison/Canadian Paediatric Society: Position Statement

treatments include aggressive measures aimed at cure, resuscitation, mechanical ventilation, and so on. Decisions to withhold or withdraw antibiotics or artificial nutrition and hydration are more controversial but may also be considered if they are detrimental to the comfort of the dying child or adolescent.',2 These difficult and often controversial decisions should not be made in isolation and should include members of the interdisciplinary team and, when possible and desired, an ethics committee

or consultant. All discussions and decisions to withhold or withdraw treatment should be well documented and reviewed after the child or adolescent's death. Some families will be comfortable with decisions to withhold or withdraw treatment, while others may take longer to reach this decision. The religious, spiritual, cultural, and moral background of families should be recognized in these situations because they frequently influence families and their decisions. F. . .1

Notes 1. Miraie ED. Withholding nutrition from seriously ill newborn infants: A parentS perspective. J Pediatr 1988;113:262-5.

2. Nelson LJ, Rushton CH, Cranford RE, Nelson, RM, Glover JJ, Truog RD. Forgoing medically provided nutrition and hydration in pediatric patients. J Law Med Ethics 1995;23:33-46.

with the outside world. It may be appropriate to withdraw or withhold life-sustaining treatment. 3. The "No Chance" Situation. The child has such severe disease that life-sustaining treatment simply Royal College of Paediatrics and Child Health delays death without significant alleviation of suffering. Treatment to sustain life is inappropriate. [. . .1 The RCPCH acknowledges that all members of 4. The "No Purpose" Situation. Although the patient the child health team, in partnership with parents, may be able to survive with treatment, the degree of physical or mental impairment will be so great have a duty to act in the best interests of the child. This includes sustaining life and restoring health to an that it is unreasonable to expect them to bear it. acceptable standard. However, there are circumstances 5. The "Unbearable" Situation. The child and/or in which treatments that merely sustain "life" neither family feel that in the face of progressive and irrestore health nor confer other benefit and hence are reversible illness further treatment is more than can be borne. They wish to have a particular treatno longer in the child's best interests. There are five situations where it may be ethical ment withdrawn or to refuse further treatment irand legal to consider withholding or withdrawal of liferespective of the medical opinion that it may be of sustaining medical treatment': some benefit.

Withholding or Withdrawing LifeSustaining Treatment in Children*

1. The "Brain Dead" Child.' In the older child3 where criteria of brain-stem death are agreed by two practitioners in the usual way4 it may still be technically feasible to provide basal cardio-respiratory support by means of ventilation and intensive care. It is agreed within the profession that treatment in such circumstances is futile and the withdrawal of current medical treatment is appropriate. 2. The "Permanent Vegetative" State.5,6 The child who develops a permanent vegetative state following insults, such as trauma or hypoxia, is reliant on others for all care and does not react or relate

In situations that do not fit with these five categories, or where there is uncertainty about the degree of future impairment or disagreement, the child's life should always be safeguarded in the best way possible by all in the Health Care Team, until these issues are resolved. Decisions must never be rushed and must always be made by the team with all evidence available. In emergencies it is often doctors in training who are called to resuscitate. Rigid rules, even for conditions which seem hopeless, should be avoided, and life-sustaining treatment should be administered and continued until a senior or more experienced doctor arrives.

Source: From Withholding or Withdrawing life Sustaining Treatment in Children, 2nd edn (2004), 10-11. Reprinted with permission of the Royal College of Paediatrics and Child Health.

•

II

LI

111111

2 Medical Decision-Making

The decision to withhold or withdraw lifesustaining therapy should always be associated with consideration of the child's overall palliative or

terminal care needs. These include symptom alleviation and care, which maintains human dignity and comfort. [. .1

Notes *In 2015 the Royal College published a revised edition of this framework, in a document titled Malting decisions to limit treatment in life-limiting and life-threatening conditions in children: A framework for practice (Larcher V, Craig F., Bhogal K. et al., Arch Dis Child 2015; 100(Suppl 2): sl-s23). In the 2015 edition, the five situations presented in the 2004 framework are replaced by a more formal classification based on quality or quantity of life (see Sec. 3 in the 2015 edition), but there is no indication that the Royal College has changed its view on this matter. We use the 2004 statement in this book because it identifies more clearly the situations in which the limitation of treatment can be plausibly considered. Of particular interest and importance is situation 5: The "Unbearable" Situation. 1.

2.

Withdrawal of curative medical treatment should signal the initiation of palliative care if this has not already been introduced. Definition—Brain death occurs when a child has sustained either (i) irreversible cessation of circulatory and respiratory functions or (ii) irreversible cessation of all functions of the entire brain including the brain stem. A

Deciding to Forego Life-Sustaining Treatment Judicial Council, American Medical Association

[...] 2.10 Quality of Life In the making of decisions for the treatment of seriously deformed newborns or persons who are severely deteriorated victims of injury, illness, or advanced age, the primary consideration should be what is best for the individual patient and not the avoidance of a burden to the family or to society. Quality of life is a factor to be considered in determining what is best for the individual. Life should be cherished despite disabilities and handicaps, except when prolongation would be inhumane and unconscionable. Under these circumstances, withholding or

3.

4.

5.

6.

determination of death must be made in accordance with accepted medical standards. Original definitions of brain death were not applied to neonates as criteria were thought to be affected by brain immaturity. Task force for the determination of brain death in children. "Guidelines for the determination of brain death in children." Annals of Neurology. (1987)21:616-617. Pediatrics. (1987)80:298-299. The vegetative state—guidance on diagnosis and management A Report of a working party of the Royal College of Clinical Medicine (2003)3:249-254. Defines the vegetative state and uses the terms "persistent" to mean a vegetative state that has persisted for four weeks or more and "permanent" when the vegetative state is deemed to be permanent and it is predicted that awareness will never recover. "The persistent vegetative state." Conference of Medical Royal Colleges and Their Faculties of the United Kingdom. Journal of the Royal College of Physicians, London. (1996)30:119-121.

removing life supporting means is ethical provided that the normal care given an individual who is ill is not discontinued. In desperate situations involving newborns, the advice and judgment of the physician should be readily available, but the decision whether to exert maximal efforts to sustain life should be the choice of the parents. The parents should be told the options, expected benefits, risks, and limits of any proposed care; how the potential for human relationships is affected by the infant's condition; and relevant information and answers to their questions. The presumption is that the love which parents usually have for their children will be dominant in the decisions which they make in determining what is in the best interest of their children. It is to be expected that parents will act unselfishly, particularly where life itself is at stake. Unless there is convincing evidence to the contrary, parental authority should be respected. [. . .1

Source: From Current Opinions of the Judicial Council of the American Medical Association (1982).

Macklin: Ethical Relativism in a Multicultural Society

Ethical Relativism in a Multicultural Society Ruth Macklin

Cultural pluralism poses a challenge to physicians and patients alike in the multicultural United States [and Canada], where immigrants from many nations and diverse religious groups visit the same hospitals and doctors. Multiculturalism is defined as "a social-intellectual movement that promotes the value of diversity as a core principle and insists that all cultural groups be treated with respect and as equals" (Fowers and Richardson 1996: 609). This sounds like a value that few enlightened people could fault, but it produces dilemmas and leads to results that are, at the least, problematic if not counterintuitive. Critics of mainstream bioethics . . . have complained about the narrow focus on autonomy and individual rights. Such critics argue that much—if not most—of the world embraces a value system that places the family the community, or the society as a whole above that of the individual person. The prominent American sociologist Renee Fox is a prime example of such critics: "From the outset, the conceptual framework of bioethics has accorded paramount status to the valuecomplex of individualism, underscoring the principles of individual rights, autonomy, self-determination, and their legal expression in the jurisprudential notion of privacy" (Fox 1990: 206). The emphasis on autonomy at least in the early days of bioethics in the United States, was never intended to cut patients off from their families by focusing monistically on the patient. Instead, the intent was to counteract the predominant and longstanding paternalism on the part of the medical profession. In fact, there was little discussion of where the family entered in and no presumption that a family-centred approach to sick patients was somehow a violation of the patientb autonomy Most patients want and need the support of their families, regardless of whether they seek to be autonomous agents regarding their own care. Respect for autonomy is perfectly consistent with recognition of the important role that families play when a loved one is ill. Autonomy has fallen into such disfavour among some bioethicists that the pendulum has begun to

swing in the direction of families, with urgings to "take families seriously" (Nelson 1992) and even to consider the interests of family members equal to those of the competent patient (Hardwig 1990). The predominant norm in the United States of disclosing a diagnosis of serious illness to the patient is not universally accepted even among longstanding citizens comprising ethnic or religious subcultures. . . . Perspectives of Health Care Workers and Patients

A circumstance that arises frequently in multicultural urban settings is one that medical students bring to ethics teaching conferences. The patient and family are recent immigrants from a culture in which physicians normally inform the family rather than the patient of a diagnosis of cancer. The medical students wonder whether they are obligated to follow the family's wish, thereby respecting their cultural custom, or whether to abide by the ethical requirement at least to explore with patients their desire to receive information and to be a participant in their medical care. When medical students presented such a case in one of the conferences I co-direct with a physician, the dilemma was heightened by the demographic picture of the medical students themselves. Among the 14 students, 11 different countries of origin were represented. Those students either had come to the United States themselves to study or their parents had immigrated from countries in Asia, Latin America, Europe, and the Middle East. The students began their comments with remarks like, "Where I come from, doctors never tell the patient a diagnosis of cancer," or "In my country, the doctor always asks the patient's family and abides by their wishes." The discussion centred on the question of whether the physician's obligation is to act in accordance with what contemporary medical ethics dictates in the United States or to respect the cultural difference of their patients and act according to the family's wishes. Not surprisingly, the medical students were divided on the answer to this question. Medical students and residents are understandably confused about their obligation to disclose information to a patient when the patient comes from a culture in which telling a patient she has cancer is rare or unheard

Source: From The Kennedy Institute of Ethics Journal 9 (1998) no. 1,1-22. Copyright CO The Johns Hopkins University Press. Reprinted with permission of The Johns Hopkins University Press.

83

II

1111

1111;111

2 Medical Decision-Making

of. They ask: "Should I adhere to the American custom of disclosure or the Argentine custom of withholding the diagnosis?" That question is miscast, since there are some South Americans who want to know if they have cancer and some North Americans who do not. It is not, therefore, the cultural tradition that should determine whether disclosure to a patient is ethically appropriate, but rather the patient's wish to communicate directly with the physician, to leave communications to the family or something in between. It would be a simplistic, if not unethical, response on the part of doctors to reason that "This is the United States, we adhere to the tradition of patient autonomy, therefore I must disclose to this immigrant from the Dominican Republic that he has cancer." Most patients in the United States do want to know their diagnosis and prognosis, and it has been amply demonstrated that they can emotionally and psychologically handle a diagnosis of cancer. The same may not be true, however, for recent immigrants from other countries, and it may be manifestly untrue in certain cultures. Although this, too, may change in time, several studies point to a cross-cultural difference in beliefs and practice regarding disclosure of diagnosis and informed consent to treatment. One survey examined differences in the attitudes of elderly subjects from different ethnic groups toward disclosure of the diagnosis and prognosis of a terminal illness and regarding decision-making at the end of life (Blackhall et al. 1995). This study found marked differences in attitudes between Korean Americans and Mexican Americans, on the one hand, and African Americans and Americans of European descent, on the other. The Korean Americans and Mexican Americans were less likely than the other two groups to believe that patients should be told of a prognosis of terminal illness and also less likely to believe that the patient should make decisions about the use of life-support technology. The Korean and Mexican Americans surveyed were also more likely than the other groups to have a family-centred attitude toward these matters; they believed that the family and not the patient should be told the truth about the patient's diagnosis and prognosis. The authors of the study cite data from other countries that bear out a similar gap between the predominant "autonomy model" in the United States and the family-centred model prevalent in European countries as well as in Asia and Africa. The study cited was conducted at 31 senior citizen centres in Los Angeles. In no ethnic group did 100 per cent of its members favour disclosure or non-disclosure to the patient. Forty-seven per cent of Korean Americans believed that a patient with metastatic cancer should

be told the truth about the diagnosis, 65 per cent of Mexican Americans held that belief, 87 per cent of European Americans believed patients should be told the truth, and 89 per cent of African Americans held that belief. It is worth noting that the people surveyed were all 65 years old or older. Not surprisingly, the Korean and Mexican American senior citizens had values closer to the cultures of their origin than did the African Americans and European Americans who were born in the United States. Another finding was that among the Korean American and Mexican American groups, older subjects and those with lower socioeconomic status tended to be opposed to truth-telling and patient decision-making more strongly than the younger, wealthier, and more highly educated members of these same groups. The authors of the study draw the conclusion that physicians should ask patients if they want to receive information and make decisions regarding treatment or whether they prefer that their families handle such matters. Far from being at odds with the "autonomy model," this conclusion supports it. To ask patients how much they wish to be involved in decision-making does show respect for their autonomy: patients can then make the autonomous choice about who should be the recipient of information or the decision-maker about their illness. What would fail to show respect for autonomy is for physicians to make these decisions without consulting the patient at all. If doctors spoke only to the families but not to the elderly Korean American or Mexican American patients without first approaching the patients to ascertain their wishes, they would be acting in the paternalistic manner of the past in America, and in accordance with the way many physicians continue to act in other parts of the world today. Furthermore, if physicians automatically withheld the diagnosis from Korean Americans because the majority of people in that ethnic group did not want to be told, they would be making an assumption that would result in a mistake almost 50 per cent of the time. Intolerance and Overtolerance

A medical resident in a New York hospital questioned a patient's ability to understand the medical treatment he had proposed and doubted whether the patient could grant truly informed consent. The patient, an immigrant from the Caribbean islands, believed in voodoo and sought to employ voodoo rituals in addition to the medical treatment she was receiving. "How can anyone who believes in that stuff be competent to consent

Macklin: Ethical Relativism in a Multicultural Society

to the treatment we offer?" the resident mused. The medical resident was an observant Jew who did not work, drive a car, or handle money on the Sabbath and adhered to Kosher dietary laws. Both the Caribbean patient and the Orthodox Jew were devout believers in their respective faiths and practised the accepted rituals of their religions. The patient's voodoo rituals were not harmful to herself or to others. If the resident had tried to bypass or override the patient's decision regarding treatment, the case would have posed an ethical problem requiring resolution. Intolerance of another's religious or traditional practices that pose no threat of harm is, at least, discourteous and at worst, a prejudicial attitude. And it does fail to show respect for persons and their diverse religious and cultural practices. But it does not (yet) involve a failure to respect persons at a more fundamental level, which would occur if the doctor were to deny the patient her right to exercise her autonomy in the consent procedures. At times, however, it is the family that interferes with the patient's autonomous decisions. Two brothers of a Haitian immigrant were conducting a conventional Catholic prayer vigil for their dying brother at his hospital bedside. The patient, suffering from terminal cancer and in extreme pain, had initially been given the pain medication he requested. Sometime later a nurse came in and found the patient alert, awake, and in excruciating pain from being undermedicated. When questioned, another nurse who had been responsible for the patient's care said that she had not continued to administer the pain medication because the patient's brothers had forbidden her to do so. Under the influence of the heavy dose of pain medication, the patient had become delirious and mumbled incoherently. The brothers took this as an indication that evil spirits had entered the patient's body and, according to the voodoo religion of their native culture, unless the spirit was exorcised it would stay with the family forever, and the entire family would suffer bad consequences. The patient manifested the signs of delirium only when he was on the medication, so the brothers asked the nurse to withhold the pain medication, which they believed was responsible for the entry of the evil spirit. The nurse sincerely believed that respect for the family's religion required her to comply with the patient's brothers' request, even if it contradicted the patient's own expressed wish. The person in charge of pain management called an ethics consultation, and the clinical ethicist said that the brothers' request, even if based on their traditional religious beliefs, could not override the patient's own request for pain medication that would relieve his suffering.

There are rarely good grounds for failing to respect the wishes of people based on their traditional religious or cultural beliefs. But when beliefs issue in actions that cause harm to others, attempts to prevent those harmful consequences are justifiable. An example that raises public health concerns is a ritual practised among adherents of the religion known as Santeria, practised by people from Puerto Rico and other groups of Caribbean origin. The ritual involves scattering mercury around the household to ward off bad spirits. Mercury is a highly toxic substance that can harm adults and causes grave harm to children. Shops called "botanicas" sell mercury as well as herbs and other potions to Caribbean immigrants who use them in their healing rituals. The public health rationale that justifies placing limitations on people's behaviour in order to protect others from harm can justify prohibition of the sale of mercury and penalties for its domestic use for ritual purposes. Yet the Caribbean immigrants could object: "You are interfering with our religious practices, based on your form of scientific medicine. This is our form of religious healing and you have no right to interfere with our beliefs and practices." It would not convince this group if a doctor or public health official were to reply: "But ours is a well-confirmed, scientific practice while yours is but an ignorant, unscientific ritual." It may very well appear to the Caribbean group as an act of cultural imperialism: "These American doctors with their Anglo brand of medicine are trying to impose it on us." This raises the difficult question of how to implement public health measures when the rationale is sufficiently compelling to prohibit religious or cultural rituals. Efforts to eradicate mercury sprinkling should enlist members of the community who agree with the public health position but who are also respected members of the cultural or religious group. Belief System of a Subculture

Some widely held ethical practices have been transformed into law, such as disclosure of risks during an informed consent discussion and offering to patients the opportunity to make advanced directives in the form of a living will or appointing a health care agent. Yet these can pose problems for adherents of traditional cultural beliefs. In the traditional culture of Navajo Native Americans, a deeply rooted cultural belief underlies a wish not to convey or receive negative information. A study conducted on a Navajo Indian reservation in Arizona demonstrated how Western biomedical and bioethical concepts and principles can come into conflict with traditional Navajo values and ways of thinking (Carrese and Rhodes 1995). In March

85

H kill

86

111111

2 Medical Decision-Making

1992, the Indian Health Service adopted the require- opportunity to make a living will or appoint a health ments of the Patient Self-Determination Act, but the care agent. A physician or nurse working for the Indian Indian Health Service policy also contains the follow- Health Service could easily fulfill this requirement by ing proviso: "Tribal customs and traditional beliefs that asking Navajo patients if they wish to discuss their relate to death and dying will be respected to the extent future care or options, without introducing any of possible when providing information to patients on the negative thinking. This approach resolves one of these issues" (Carrese and Rhodes 1995: 828). the limitations of the published study. As the authors The relevant Navajo belief in this context is the acknowledge, the findings reflect a more traditional notion that thought and language have the power to perspective and the full range of Navajo views is not shape reality and to control events. The central con- represented. So it is possible that some patients who cern posed by discussions about future contingencies use the Indian Health Service may be willing or even is that traditional beliefs require people to "think and eager to have frank discussions about risks of treatment speak in a positive way" When doctors disclose risks and future possibilities, even negative ones, if offered of a treatment in an informed consent discussion, they the opportunity. speak "in a negative way," thereby violating the Navajo It is more difficult, however, to justify withholdprohibition. The traditional Navajo belief is that health ing from patients the risks of proposed treatment in is maintained and restored through positive ritual an informed consent discussion. The article about the language. This presumably militates against disclosing Navajo beliefs recounts an episode told by a Navajo risks of treatment as well as avoiding mention of future woman who is also a nurse. Her father was a candidate illness or incapacitation in a discussion about advance for bypass surgery. When the surgeon informed the pacare planning. Western-trained doctors working with tient of the risks of surgery, including the possibility that the traditional Navajo population are thus caught in a he might not wake up, the elderly Navajo man refused dilemma. Should they adhere to the ethical and legal the surgery altogether. If the patient did indeed require standards pertaining to informed consent now in force the surgery and refused because he believed that telling in the rest of the United States and risk harming their him of the risk of not waking up would bring about patients by "talking in a negative way"? Or should they that result, then it would be justifiable to withhold that adhere to the Navajo belief system with the aim of risk of surgery. Should not that possibility be routinely avoiding harm to the patients but at the same time vio- withheld from all patients, then, since the prospect of lating the ethical requirement of disclosure to patients not waking up could lead other people—Navajos and of potential risks and future contingencies? non-Navajos alike—to refuse the surgery? The answer The authors of the published study draw several is no, but it requires further analysis. conclusions. One is that hospital policies complying Respect for autonomy grants patients who have with the Patient Self-Determination Act are ethically been properly informed the right to refuse a proposed troublesome for the traditional Navajo patients. Since medical treatment. An honest and appropriate disclophysicians who work with that population must decide sure of the purpose, procedures, risks, benefits, and how to act, this problem requires a solution. A second available alternatives, provided in terms the patient can conclusion is that "the concepts and principles of West- understand, puts the ultimate decision in the hands of ern bioethics are not universally held" (Carrese and the patient. This is the ethical standard according to Rhodes 1995: 829). This comes as no surprise. It is a Western bioethics. A clear exception exists in the case straightforward statement of the thesis of descriptive of patients who lack decisional capacity altogether, and ethical relativism, the evident truth that a wide variety debate continues regarding the ethics of paternalisticof cultural beliefs about morality exist in the world. ally overriding the refusal of marginally competent paThe question for normative ethics endures: What fol- tients. This picture relies on a key feature that is lacking lows from these particular facts of cultural relativity? in the Navajo case: a certain metaphysical account of A third conclusion the authors draw, in light of their the way the world works. Western doctors and their findings, is that health care providers and institutions patients generally do not believe that talking about caring for Navajo patients should reevaluate their poli- risks of harm will produce those harms (although there cies and procedures regarding advance care planning. have been accounts that document the "dark side" of This situation is not difficult to resolve, ethically the placebo effect). It is not really the Navajo values or practically. The Patient Self-Determination Act does that create the cross-cultural problem but rather their not mandate patients to actually make an advance metaphysical belief system holding that thought and directive; it requires only that health care institutions language have the power to shape reality and control provide information to patients and give them the events. In fact, the Navajo values are quite the same as

Macklin: Ethical Relativism in a Multicultural Society

the standard Western ones: fear of death and avoidance of harmful side effects. To understand the relationship between cultural variation and ethical relativism, it is essential to distinguish between cultural relativity that stems from a difference in values and that which can be traced to an underlying metaphysics or epistemology. Against this background, only two choices are apparent: insist on disclosing to Navajo patients the risks of treatment and thereby inflict unwanted negative thoughts on them; or withhold information about the risks and state only the anticipated benefits of the proposed treatment. Between those two choices, there is no contest. The second is clearly ethically preferable. It is true that withholding information about the risks of treatment or potential adverse events in the future radically changes what is required by the doctrine of informed consent. It essentially removes the "informed" aspect, while leaving in place the notion that the patient should decide. The physician will still provide some information to the Navajo patient, but only the type of information that is acceptable to the Navajos who adhere to this particular belief system. True, withholding certain information that would typically be disclosed to patients departs from the ethical ideal of informed consent, but it does so in order to achieve the ethically appropriate goal of beneficence in the care of patients. The principle of beneficence supports the withholding of information about risks of treatment from Navajos who hold the traditional belief system. But so, too, does the principle of respect for autonomy. Navajos holding traditional beliefs can act autonomously only when they are not thinking in a negative way. If doctors tell them about bad contingencies, that will lead to negative thinking, which in their view will fail to maintain and restore health. The value of both doctor and patient is to maintain and restore health. A change in the procedures regarding the informed consent discussion is justifiable based on a distinctive background condition: the Navajo belief system about the causal efficacy of thinking and talking in a certain way. The less-than-ideal version of informed consent does constitute a "lower" standard than that which is usually appropriate in today's medical practice. But the use of a "lower" standard is justified by the background assumption that that is what the Navajo patient prefers. What is relative and what is non-relative in this situation? There is a clear divergence between the Navajo belief system and that of Western science. That divergence leads to a difference in what sort of discussion is appropriate for traditional Navajos in the medical setting and that which is standard in

Western medical practice. According to one description, "always disclose the risks as well as the benefits of treatment to patients," the conclusion points to ethical relativism. But a more general description, one that heeds today's call for cultural awareness and sensitivity, would be: "Carry out an informed consent discussion in a manner appropriate to the patient's beliefs and understanding." That obligation is framed in a non-relative way. A heart surgeon would describe the procedures, risks, and benefits of bypass surgery in one way to a patient who is another physician, in a different way to a mathematician ignorant of medical science, in yet another way to a skilled craftsman with an eighth grade education, and still differently to a traditional Navajo. The ethical principle is the same; the procedures differ.

Obligations of Physicians The problem for physicians is how to respond when an immigrant to the United States acts according to the cultural values of her native country values that differ widely from accepted practices in American medicine. Suppose an African immigrant asks an obstetrician to perform genital surgery on her baby girl. Or imagine that a Laotian immigrant from the Iu Mien culture brings her four-month-old baby to the pediatrician for a routine visit and the doctor discovers burns on the baby's stomach. The African mother seeks to comply with the tradition in her native country, Somalia, where the vast majority of women have had clitoridectomies. The Iu Mien woman admits that she has used a traditional folk remedy to treat what she suspected was her infant's case of a rare folk illness. What is the obligation of physicians . . . when they encounter patients in such situations? At one extreme is the reply that . . . physicians are obligated to follow the ethical and cultural practices accepted here and have no obligation to comply with patients' requests that embody entirely different cultural values. At the other extreme is the view that cultural sensitivity requires physicians to adhere to the traditional beliefs and practices of patients who have emigrated from other cultures. A growing concern on the part of doctors and public health officials is the increasing number of requests for genital cutting and defence of the practice by immigrants to the United States and European countries. A Somalian immigrant living in Houston said he believed his Muslim faith required him to have his daughters undergo the procedure; he also stated his belief that it would preserve their virginity. He was quoted as saying, "It's my responsibility. If I don't do

II

;III

IIIIi

2 Medical Decision-Making

it, I will have failed my children" (Dugger 1996: 1). Another African immigrant, living in Houston, sought a milder form of the cutting she had undergone for her daughter. The woman said she believed it was necessary so her daughter would not run off with boys and have babies before marriage. She was disappointed that Medicaid would not cover the procedure, and planned to go to Africa to have the procedure done there. A New York City physician was asked by a father for a referral to a doctor who would do the procedure on his three-year-old daughter. When the physician told him this was not done in America, the man accused the doctor of not understanding what he wanted (Dugger 1996: 1, 9). However, others in our multicultural society consider it a requirement of "cultural sensitivity" to accommodate in some way to such requests of African immigrants. Harborview Medical Center in Seattle sought just such a solution A group of doctors agreed to consider making a ritual nick in the fold of skin that covers the clitoris, but without removing any tissue. However, the hospital later abandoned the plan after being flooded with letters, postcards, and telephone calls in protest (Dugger 1996). A physician who conducted research with East African women living in Seattle held the same view as the doctors who sought a culturally sensitive solution. In a talk she gave to my medical school department, she argued that Western physicians must curb their tendency to judge cultural practices different from their own as "rational" or "irrational." Ritual genital cutting is an "inalienable" part of some cultures, and it does a disservice to people from those cultures to view it as a human rights violation. She pointed out that in the countries where female genital mutilation (FGM) is practised, circumcised women are "normal." Like some anthropologists who argue for a "softer" linguistic approach (Lane and Rubinstein 1996), this researcher preferred the terminology of "circumcision" to that of "female genital mutilation." One can understand and even have some sympathy for the women who believe they must adhere to a cultural ritual even when they no longer live in the society where it is widely practised. But it does not follow that the ritual is an "inalienable" part of that culture, since every culture undergoes changes over time. Furthermore, to contend that in the countries where FGM is practised, circumcised women are "normal" is like saying that malaria or malnutrition is "normal" in parts of Africa. That a human condition is statistically normal implies nothing whatever about whether an obligation exists to seek to alter the statistical norm for the betterment of those who are affected.

Some Africans living in the United States have said they are offended that Congress passed a law prohibiting female genital mutilation that appears to be directed specifically at Africans. France has also passed legislation, but its law relies on general statutes that prohibit violence against children (Dugger 1996). In a recent landmark case, a French court sent a Gambian woman to jail for having had the genitals of her two baby daughters mutilated by a midwife. French doctors report an increasing number of cases of infants who are brought to clinics hemorrhaging or with severe infections. Views on what constitutes the appropriate response to requests to health professionals for advice or referrals regarding the genital mutilation of their daughters vary considerably. Three commentators gave their opinions on a case vignette in which several African families living in a US city planned to have the ritual performed on their daughters. If the procedure could not be done in the US, the families planned to have it done in Africa. One of the parents sought advice from health professionals. One commentator, a child psychiatrist, commented that professional ethical practice requires her to respect and try to understand the cultural and religious practices of the group making the request (Brant 1995). She then cited another ethical requirement of clinical practice: her need to promote the physical and psychological well-being of the child and refusal to condone parenting practices that constitute child abuse according to the social values and laws of her city and country. Most of what this child psychiatrist would do with the mother who comes to her involves discussion, mutual understanding, education, and the warning that in this location performing the genital cutting ritual would probably be considered child abuse. The psychiatrist would remain available for a continuing dialogue with the woman and others in her community, but would stop short of making a child-abuse report since the woman was apparently only considering carrying out the ritual. However, the psychiatrist would make the report if she had knowledge that the mother was actually planning to carry out the ritual or if it had already been performed. She would make the child-abuse report reluctantly, however, and only if she believed the child to be at risk and if there were no other option. She concluded by observing that the mother is attempting to act in the best interest of her child and does not intend to harm her. The psychiatrist's analysis demonstrates the possible ambiguities of the concept of child abuse. Is abuse determined solely by the intention of the adult? Should child abuse be judged by the harmful consequences to the child, regardless of the adult's intention? Of course,

Macklin: Ethical Relativism in a Multicultural Society

if a law defines the performance of female genital mutilation as child abuse, then it is child abuse, from a legal point of view, and physicians are obligated to report any case for which there is a reasonable suspicion. Legal definitions aside, intentions are relevant for judging the moral worth of people, but not for the actions they perform. This means that the good intentions of parents could exonerate them from blame if their actions cause harm to their children, but the harmful actions nevertheless remain morally wrong. The second commentator, a clinical psychologist and licensed sex therapist, would do many of the same things as the child psychiatrist, but would go a bit further in finding others from the woman's community and possibly another support network (Wyatt 1995). Like most other commentators on female genital mutilation, this discussant remarked that "agents of change must come from within a culture" (Wyatt 1995: 289). The third commentator on this case vignette was the most reluctant to be critical. A British historian and barrister, he began with the observation that "a people's culture demands the highest respect" (Martin 1995). On the one hand, he noted that custom, tradition, and religion are not easily uprooted. But on the other hand, he pointed out that no human practice is beyond questioning. He contended that the debate over the nature and impact of female circumcision is a "genuine debate," and the ritual probably had practical utility when it was introduced into the societies that still engage in it. Of the three commentators, he voiced the strongest opposition to invoking the child abuse laws because it "would be an unwarranted criminalization of parents grappling in good faith with a practice that is legal and customary in their home country" (Martin 1995: 291). In the end, this discussant would approach the parents "much as a lawyer would address a jury," leaving the parents (like a jury) to deliberate and come to an informed decision. He would also involve the girls in this process, since they are adolescents, and should have input into the deliberations. It is tempting to wonder whether the involvement of adolescent girls in deliberations of their parents would, in traditional Gambian culture, be even remotely considered, much less accepted. The "lawyer-jury-adolescent involvement" solution looks to be very Western. If these families living in the United States still wish to adhere to their cultural tradition of genital mutilation, is it likely that they will appreciate the reasoned, deliberative approach this last commentator proposed? Exactly where to draw the line in such cases is a difficult matter. Presumably, one could go farther than any of these commentators and inform the African families

that since US law prohibits female genital mutilation, which has been likened to child abuse, a health professional would be obligated to inform relevant authorities of an intention to commit child abuse. Conceivably, US authorities could prevent immigrants from returning to this country if they have gone to Africa to have a procedure performed that would be illegal if done within the United States. But this is a matter of law, not ethics, and would involve a gross invasion of privacy since to enforce the ruling it would be necessary to examine the genitals of the adolescent girls when these families sought re-entry into the United States. That would be going too far and probably deserves condemnation as "ethical imperialism." Since the cutting would already have been done, punitive action toward the family could not succeed in preventing the harm. Another case vignette describes a Laotian woman from the Mien culture who immigrated to the United States and married a Mien man. When she visited her child's pediatrician for a routine four-month immunization, the doctor was horrified to see five red and blistered quarter-inch round markings on the child's abdomen (Case Study: Culture, Healing, and Professional Obligations 1993). The mother explained that she used a traditional Mien "cure" for pain, since she thought the infant was experiencing a rare folk illness among Mien babies characterized by incessant crying and loss of appetite, in addition to other symptoms. The "cure" involves dipping a reed in pork fat, lighting the reed, and passing the burning substance over the skin, raising a blister that "pops like popcorn." The popping indicates that the illness is not related to spiritual causes; if no blisters appear, then a shaman may have to be summoned to conduct a spiritual ritual for a cure. As many as 11 burns might be needed before the end of the "treatment." The burns are then covered with a mentholated cream. The Mien woman told the pediatrician that infection is rare and the burns heal in a week or so. Scars sometimes remain but are not considered disfiguring. She also told the doctor that the procedure must be done by someone skilled in burning, since if a burn is placed too near the line between the baby's mouth and navel, the baby could become mute or even retarded. The mother considered the cure to have been successful in the case of her baby since the child had stopped crying and regained her appetite. Strangely enough, the pediatrician did not say anything to the mother about her practice of burning the baby, no doubt from the need to show "cultural sensitivity" She did, however, wonder later whether she should have said something since she thought the practice was dangerous and also cruel to babies.

II

311

II 1111

2 Medical Decision-Making

One commentator who wrote about this case from other cultures who employ painful and potenproposed using "an ethnographic approach" to ethics tially harmful rituals that have no scientific basis. This in the cross-cultural setting (Carrese 1993). This ap- attitude of tolerance is appropriate against the backproach need not result in a strict ethical relativism, ground knowledge that the parents do not intend to however, since one can be respectful of cultural dif- harm the child and are simply using a folk remedy ferences and at the same time acknowledge that there widely accepted in their own culture. But tolerance are limits. What is critical is the perceived degree of of these circumstances must be distinguished from a harm; some cultural practices may constitute atroci- judgment that the actions harmful to children should ties and violations of fundamental human rights. The be permitted to continue. What puzzles me is the commentator argued that the pediatrician must first notion that "cultural sensitivity" must extend so far seek to understand the Mien woman in the context of as to refrain from providing a solid education to these her world before trying to educate her in the ways of parents about the potential harms and the infliction of Western medicine. The commentator stopped short gratuitous pain. . . . We ought to be able to respect of providing a solution, but noted that many possible cultural diversity without having to accept every single resolutions can be found for cross-cultural ethical con- feature embedded in traditional beliefs and rituals. flicts. Be that as it may, we still need to determine which The reluctance to impose modern medicine on of the pediatrician's obligations should take preced- immigrants from a fear that it constitutes yet another ence: to seek to protect her infant patient (and possibly instance of "cultural imperialism" is misplaced. Is it not also the Mien woman's other children) from harmful possible to accept non-Western cultural practices side rituals or to exhibit cultural sensitivity and refrain from by side with Western ones, yet condemn those that are attempts at re-education or critical admonitions. manifestly harmful and have no compensating benefit A second pair of commentators assumed a non- except for the cultural belief that they are beneficial? judgmental stance. These commentators urged respect The commentators who urged respect for the Mien for cultural diversity and defended the Mien woman's woman's burning treatment on the grounds that it is belief system as entirely rational: "It is well grounded in practised widely, the reasons for it are widely underher culture it is practised widely; the reasons for it are stood among the Mien, and the procedure works, from widely understood among the Iu Mien; the procedure, a Mien point of view, seemed to be placing that practice from a Mien point of view, works" (Brown and Jameton on a par with practices that "work" from the point of 1993: 17). This is a culturally relative view of rational- view of Western medicine. Recall that if the skin does ity. The same argument could just as well be used to not blister, the Mien belief holds that the illness may justify female genital mutilation. Nevertheless, the com- be related to spiritual causes and a shaman might have mentators rejected what they said was the worst choice: to be called. Should the pediatrician stand by and do simply to tolerate the practice as a primitive cultural nothing, if the child has a fever of 104 degrees and the artifact and do nothing more. They also rejected the op- parent calls a shaman because the skin did not blister? posite extreme: a referral of child abuse to the appropri- Recall also that the Mien woman told the pediatrician ate authorities. The mother's actions did not constitute that if the burns are not done in the right place, the intentional abuse, since she actually believed she was baby could become mute or even retarded. Must we helping the child by providing a traditional remedy. reject the beliefs of Western medicine regarding causalHere I think the commentators are correct in rejecting ity and grant equal status to the Mien beliefs? To refrain a referral to the child-abuse authorities, since a charge from seeking to educate such parents and to not exhort of child abuse can have serious consequences that may them to alter their traditional practices is unjust, as it ultimately run counter to the best interests of the child. exposes the immigrant children to health risks that are What did these commentators recommend? Not to not borne by children from the majority culture. It is heresy in today's postmodern climate of retry to prohibit the practice directly, which could alienate the parent. Instead, the pediatrician could discuss spect for the belief systems of all cultures to entertain the risk of infection and suggest safer pain remedies. the notion that some beliefs are demonstrably false and The doctor should also learn more about the rationale others, whether true or false, lead to manifestly harmful for and technique of the traditional burning "cure." The actions. We are not supposed to talk about the evolumost she should do, according to these commentators, tion of scientific ideas or about progress in the Western is consider sharing her concerns with the local Mien world, since that is a colonialist way of thinking. If it is simply "the white man's burden, medicalized" (Morsy community, but not with the mother alone. There is in these commentaries a great reluctance 1991) to urge African families living in the United to criticize, scold, or take legal action against parents States not to genitally mutilate their daughters, or to

Macklin: Ethical Relativism in a Multicultural Society

attempt to educate Mien mothers about the harms of burning their babies, then we are doomed to permit ethical relativism to overwhelm common sense. Multiculturalism, as defined at the beginning of this paper, appears to embrace ethical relativism and yet is logically inconsistent with relativism. The second half of the definition states that multiculturalism "insists that all cultural groups be treated with respect and as equals." What does this imply with regard to cultural groups that oppress or fail to respect other cultural groups? Must the cultural groups that violate the mandate to treat all cultural groups with respect and as equals be respected themselves? It is impossible to insist that all such groups be treated with respect and as equals, and at the same time accept any particular group's attitude toward and treatment of another group as inferior. Every cultural group contains subgroups within the culture: old and young, women and men,

people with and people without disabilities. Are the cultural groups that discriminate against women or people with disabilities to be respected equally with those that do not? What multiculturalism does not say is whether all of the beliefs and practices of all cultural groups must be equally respected. It is one thing to require that cultural, religious, and ethnic groups be treated as equals; that conforms to the principle of justice as equality. It is quite another thing to say that any cultural practice whatever of any group is to be tolerated and respected equally. This latter view is a statement of extreme ethical relativism. If multiculturalists endorse the principle of justice as equality however, they must recognize that normative ethical relativism entails the illogical consequence of toleration and acceptance of numerous forms of injustice in those cultures that oppress women and religious and ethnic minorities.

References Blackhall, Leslie, Murphy, Sheila T., Frank, Gelya, Michel, Vicki, and Azen, Stanley 1995. "Ethnicity and Attitudes Toward Patient Autonomy," in Journal of the American Medical Association 274: 820-5. Brant, Renee. 1995. "Child Abuse or Acceptable Cultural Norms: Child Psychiatrist's Response," in Ethics & Behavior 5: 284-7. Brown, Kate, and Jameton, Andrew. 1993. "Culture, Healing, and Professional Obligations: Commentary," in Hastings Center Report 23 (4): 17. Carrese, Joseph. 1993. "Culture, Healing, and Professional Obligations: Commentary," in Hastings Center Report 23 (4): 16. , and Rhodes, Lorna A. 1995. "Western Bioethics on the Navajo Reservation: Benefit or Harm?," in Journal of the American Medical Association 274: 826-9. "Case Study: Culture, Healing, and Professional Obligations." 1993. In Hastings Center Report 23 (4): 15. Dugger, Celia W. 1996. "Tug of Taboos: African Genital Rite vs US Law," in New York Times (28 Dec.): 1,9.

Fowers, Blaine J., and Richardson, Frank C. 1996. "Why is Multiculturalism Good?," in American Psychologist 51: 609-21. Fox, Renee C. 1990. "The Evolution of American Bioethics: A Sociological Perspective," in Social Science Perspectives on Medical Ethics, ed. George Weisz. University of Pennsylvania Press Philadelphia 201-20. Hardwig, John. 1990. "What About the Family?," in Hastings Center Report 20 (2): 5-10. Lane, Sandra D., and Rubinstein, Robert A. 1996. "Judging the Other: Responding to Traditional Female Genital Surgeries," in Hastings Center Report 26 (5): 31-40. Martin, Tony. 1995. "Cultural Contexts," in Ethics & Behavior 5: 290-2. Morsy, Soheir A. 1991. "Safeguarding WomeiA Bodies: The White Man's Burden Medicalized," in Medical Anthropology Quarterly 5 (1): 19-23. Nelson, James Lindemann. 1992. "Taking Families Seriously," in Hastings Center Report 22 (4): 6-12. Wyatt, Gail Elizabeth. 1995. "Ethical Issues in Culturally Relevant Interventions," in Ethics & Behavior 5: 288-90.

2.6 Cases

Case 1

Scott Starson: Refusing Treatment while Incompetent Physics prodigy Scott Starson had been in and out of psychiatric facilities in the US and Canada since

1985 with a history of erratic and threatening behaviour. He was variously diagnosed as having bipolar disorder and/or schizophrenia. In January 1999, the Ontario Review Board ordered Mr Starson's mandatory 12-month detention at the Centre for Addiction and Mental Health in Toronto after he was found not criminally responsible by reason of mental illness for

91

II

0

'ail

111111

2 Medical Decision-Making

uttering death threats to fellow residents of his rooming house. Mr Starson's physicians proposed treatment for the bipolar disorder that included anti-psychotic, antianxiety, and mood-stabilizing drugs. Mr Starson recognized that he was mentally ill and that he could not be released until he took the medication, but he ref used to consent because he said the treatment would slow his thinking and prevent him from engaging in the scientific research that gave his life meaning—a state he described as "worse than death." His physicians were convinced that taking the medication would be beneficial to the patient's mental well-being and would alleviate any concerns about his behaviour in society. They also looked to the Canadian Health Care Consent Act, which states that "individuals are considered to be capable with respect to treatment if they are able to understand the information relevant to making decisions about a proposed treatment, and able to appreciate the reasonably foreseeable consequences of a decision or a lack of decision."' It appeared to them that Mr Starson could not adequately appreciate the value of treatment, so they brought the case before a medical review board, which subsequently determined that Mr Starson was not competent to make medical decisions for himself. A surrogate decision-maker was appointed and approved the treatment. Mr Starson applied to the Ontario Consent and Capacity Board (ccB) for a review of the physician's decision. The CCB agreed with the physician that Mr Starson was unable to fully appreciate the benefits and risks of treatment. Next, the case went to judicial review at the Ontario Superior Court of Justice, then to the Court of Appeal, which upheld the findings of the reviewing judge. Finally, the case was appealed to the Supreme Court of Canada.

In a 6-3 split decision handed down in June 2003, the Supreme Court overturned the finding of incapacity, stating that the CCB had been guided too much by its interpretation of Mr Starson's best interests rather than by a strict interpretation of the law.' According to the ruling, patients are not held to a best-interests standard when being tested for competence and may therefore make decisions that are contrary to their physician's advice, nor do they have to agree with a physician's precise diagnosis in order to be deemed competent. The court concluded that mentally ill patients are presumptively entitled to make their own decisions regarding treatment unless a "balance of probabilities" exists to override the decisions. Mr Starson's mother, Jeanne Stevens, believed the Supreme Court decision had ruined her son's life. The medication would have taken away his erratic behaviour, allowing him to work and live in the community rather than remaining confined to the Brockville Psychiatric Hospital with no prospects for release. "I'm devastated and I truthfully believe that the Supreme Court did not have sufficient information. It's the end of his life."2 As it turned out, it was not the end of his life. Mr Starson developed severe paranoid delusions, refused food, and fell into a life-threatening situation. His mother then authorized treatment, and his condition improved. Two years later the board ordered the hospital to assist Mr Starson to become established in an apartment in the community. Was the Supreme Court of Canada right to allow him to refuse his medications? Ontario, Manitoba, Prince Edward Island, the Northwest Territories, and Nunavut are the only jurisdictions that allow capable patients admitted under the Mental Health Act to refuse treatment. Should the mother have been allowed to authorize the treatment that he earlier refused?

Notes 1.

Starson v. Swayze, 2003 5CC 32, [2003] 1 SCR 722.

Case 2

No Chemotherapy for Anael: Surrogate Refusal of Treatment for a Minor Child In July 2007, three-year-old Anael IlEsperanceNascimento was being treated at the Children's Hospital

2.

News, "Mentally Ill Man May Refuse Treatment, Court Rules," 6 June 2003, www.cbc.ca/canada/story/2003/06/06/ starson030606.html. CBC

of Eastern Ontario in Ottawa for cancerous tumours in his brain and spinal cord. Physicians recommended chemotherapy, even though the tumours were not then life-threatening, but Anael's parents refused. They decided instead to feed the boy a vegan diet of organic vegetables, without sugar or animal products, because they believed that the body could heal itself if given the right nutrition. They also were concerned about

Cases

the toll prior treatments had taken on the child, who had lost weight and had a greenish complexion: "So we could not imagine continuing it [the treatment]. . . . We would probably have lost him." Officials at the hospital asked the province's child protection agency to intervene to force treatment, but the agency declined because the boy's condition was not at that point life-threatening.' According to Jean-Pierre Menard, a Quebec medical lawyer, the court rarely overrides the right of parents to refuse treatment for their child unless the advantages of

treatment far outweigh the harm.' Chemotherapy is toxic and highly aggressive, and the evidence to establish that treatment is necessary has so far not been produced. The option remains for hospital officials to petition the court to order treatment at a later date if Anael's condition worsens. "The parents are making what is doubtless a loving decision but it's not a rational decision and I think that public authorities will override their wishes if there is an effective alternative," said Arthur Schafer, a University of Manitoba medical ethicist.'

Notes 1.

CBC News, "Quebec Parents Refuse Cancer Treatment

for Boy, 3," 25 July 2007, wwwcbc.ca/canada/ottawa/ story/2007/07/25/ot-chemo-070725.html.

Case 3

Do Everything for Mom: Advance Directives and a Surrogate's Right to Demand Treatment In June 2003, 81-year-old Joyce Holland was living with advanced Alzheimer's disease in a Toronto longterm care facility when staff noticed that she was swallowing into her lungs rather than into her stomach. After being admitted to Toronto Western Hospital, Mrs Holland was diagnosed with aspiration pneumonia, a condition that required repeated suction of her lungs. She also could not eat or speak, and her hips and back were marked with deep, painful bedsores, while her knees, elbows, and fingers were permanently bent. When her condition deteriorated, she was transferred to the intensive care unit, where she was placed on a ventilator and given drugs to raise her blood pressure. This treatment was temporarily successful, and Mrs Holland was released, but she was forced to return one week later with an infection in her feeding tube. She stayed in the ICU for one month before being moved to a ward where she continued to develop fevers and infections and required lung suctioning as frequently as every 45 minutes.' Mrs Holland's medical team, which included Dr Laura Hawryluck, believed that continued treatment would not stop the progress of her disease and would unnecessarily prolong her suffering. The team met with Mrs Holland's two daughters, Patricia and Margaret, to discuss palliative care. Dr Hawryluck suggested that Mrs Holland not be returned to the ICU

2.

c-ry News, "Quebec Refuses to Force Chemo on 3-YearOld Boy," 27 July 2007, www.ctv.ca/servlet/ArticleNews/ story/ctvnews/20070727/cancerkid_treatment_070727/20 070727?hub=TopStories.

when her condition inevitably deteriorated; instead, she would be offered palliative care in the form of pain and sedative medications until her death. The physician then sent notice to Ontario's Consent and Capacity Board (CCB) that she would be seeking an order allowing her to withhold life-sustaining treatment from Mrs Holland, specifically cardiopulmonary resuscitation, ventilation, and adrenaline-type drugs to raise blood pressure. The daughters together held Mrs Holland's power-of-attorney for her personal care and acted as her proxy decision-maker. They disagreed with Dr Hawryluck's recommendation on the basis of their mother's previously stated wishes, which arose from her deeply held religious convictions. At the CCB, the daughters explained that Mrs Holland "believed strongly in the sanctity of human life" and argued that "She would want to fight for that life, regardless of the pain or loss of dignity."' Their mother still watched television and interacted with them, communicating in non-verbal ways—signs, they believed, that her life was worth saving. In contrast, Dr Hawryluck explained to the board that even with heroic efforts in the icu, Mrs Holland would likely die within six months. The board agreed, stating that Mrs Holland had not left sufficiently clear instructions to her children about how to be treated in the current circumstances and that the CCB would decide what was in her best interests. That decision was to permit the withholding-of-treatments request on the grounds that the "pain, discomfort and loss of dignity" Mrs Holland would suffer in the course of further treatment outweighed any potential benefits.3

93

II III

94

II

II

ei

1111

.11111J

2 Medical Decision-Making

The daughters appealed the board's decision to Ontario's Superior Court, claiming that they had acted according to their mother's expressed wishes as the proxies and that the ca had inappropriately usurped their position as proxy decision-maker for their mother. They further argued that the law that gave the CCB the ability to override them was unconstitutional, denying their mother her Charter rights to "life, liberty, and security of the person." When the case was heard on 20 January 2004, Mrs Holland was back in icu undergoing successful treatment: her lungs cleared, her pneumonia abated, and ventilator support was decreased. Judge Cullity determined that the Health Care Consent Act does not directly address whether doctors have to seek consent to withhold or withdraw treatment they

consider inappropriate. He also stated that the CCB had not given enough consideration to Mrs Holland's expressed wishes, her desire for and enjoyment of life, and the fact that she wanted everything possible to be done.' Medical decision-making was given back to her daughters; the hospital chose not to appeal. Mrs Holland remained alive in the icu at Toronto Western Hospital for more than a year, at an estimated cost of $1500 per day.' It should be noted that Ontario is unique among Canadian provinces in legally upholding a patient's previously expressed competent wishes. The CCB is allowed to determine a patient's best interests only when the patientb prior competent wishes were not clearly expressed.'

Notes 1.

2.

Andrew Duffy and Pauline Tam, "Patients, Doctors in Ethical 'Grey Zone,— Ottawa Citizen 28 April 2005, www .canada.com/components/print.aspx?id=13185d20-d6d74b33-92fc-a259eec714aftsponsor=. Andrew Duffy and Pauline Tam. "End-of-life Dilemma." Ottawa Citizen 18 April 2005, www.canada.com/ottawacitizen/ story.html?id=b2c44167-a987-44bd-bacb-c293bb53a806.

Case 4

Treatment for Jehovah's Witnesses: Adults and Children In June of 1979, Georgette Malette, 57, was involved in a head-on car collision that killed her husband. She was unconscious when she arrived at a hospital in her Kirkland Lake, Ontario, community. Her attending physician, Dr Shulman, determined that she was suffering from shock due to blood loss and that a blood transfusion was necessary to save her life. A nurse going through Mrs Malette's purse discovered a card that identified the patient as a Jehovah's Witness and requested that no blood be administered under any circumstances. After getting a second opinion, Dr Shulman accepted full responsibility for his actions and administered the transfusions. Mrs Malette recovered from her injuries and subsequently successfully sued Dr Shulman for assault and battery, winning an award of $20,000 in damages. Dr Shulman appealed to the Ontario Court of Appeal, which heard the case in 1990. The basis of Dr Shulman's appeal was that the presence of a card on the patient's body is not an effective informed refusal: "the card cannot be effective when

3. 4.

Scardoni v. Hawtyluch (2004), CanL11 34326 (ON sc). Daniel L. Ambrosini and Anne G. Crocker. "Psychiatric Advance Directives and the Right to Refuse Treatment in Canada," Canadian Journal of Psychiatry 52, no. 6 (2007): 397-402.

the doctor is unable to provide the patient with the information she would need before making a decision to withhold consent in this specific emergency situation."' Mr Justice Robins, in his decision to reject the appeal, stated that a doctor is not free to disregard a patient's advance instructions, even if the choice is "contrary to the mandates of his own conscience and professional judgment."' "The principles of self-determination and individual autonomy compel the conclusion that the patient may reject blood transfusions even if harmful consequences may result and even if the decision is generally regarded as foolhardy."3 Although the refusal of treatment was not verbally made, since there was no reason to doubt that the card validly expressed Mrs Malette's desire to withhold consent to blood transfusions, the card was considered by Justice Robins as a legitimate advance directive. In 2006, a 14-year-old Manitoba girl was hospitalized in Winnipeg suffering from Crohn's disease, which can cause gastrointestinal bleeding, therefore requiring blood transfusions. As a Jehovah's Witness, the youth refused treatment, but Manitoba Child and Family Services obtained a court order forcing the transfusion based on her doctor's recommendation. The province followed legislation which sets the threshold age for medical decision making at 16. The patient's lawyers

Cases

claimed that this decision violated her constitutional rights and argued that decisions should be made on a case-by-case basis according to competence and not some arbitrary age scale. In a 6-1 ruling, the Manitoba Supreme Court declared that such medical

interventions are constitutionally sound, striking a balance between protecting the child and allowing the child to choose. However, they also declared that lower courts must consider minor's maturity and decisionmaking skills in all future enforced-treatment cases.'

Notes 1. 2.

Malette v. Shulman, 1990, Ontario Court of Appeal, (1990) 72 O.R. (2d) 417. Ibid.

Case 5

Markayla Sault: Caring for an Aboriginal Patient Markayla Sault was an 11-year-old Aboriginal girl who had acute lymphoblastic leukemia Doctors gave Markayla a 72 per cent chance of survival if she continued with chemotherapy. A two-year course of treatment was planned, but after 11 weeks Markayla experienced severe side effects that put her in the intensive care unit. She begged for the chemo to be stopped; the family acceded to her wish; Family Services in Ontario did not contest the decision, and she died in January 2015. On the day Markayla died, an editorial appeared in the Canadian Medical Association Journal (CMAJ) arguing that doctors must respect the desire of parents to substitute "native" medicine for scientific medicine. In that editorial, "Caring for Aboriginal Patients Requires Trust and Respect Not Courtrooms," Dr Lisa Richardson, a specialist in internal medicine and co-leader of Toronto's Office of Indigenous Medical Education, and Dr Matthew Stanbrook, deputy editor of CMAJ, argued:

3. 4.

Ibid. CBC News, 26

June 2009, http://www.cbc.ca/news/canada/ story/2009/06/26/supreme-blood026.htm1.

To make medical treatment acceptable to our Aboriginal patients the health care system must earn their trust by delivering respect. We must ensure that our Aboriginal patients, their families and communities feel welcome, are comfortable selfidentifying as Aboriginal and do not fear judgment based on stereotypes E. . .1 Medical science is not specific to a single culture but is shared by Aboriginal and nonAboriginal people alike. Most Aboriginal people seek care from health professionals—but nearly half also use traditional medicines. Aboriginal healing traditions are deeply valued ancestral practices that emphasize plant-based medicines, culture and ceremony, multiple dimensions of health (physical, mental, emotional and spiritual), and relationships between healer, patient, community and environment. These beliefs create expectations that Aboriginal patients bring to their health care encounters; these must be respected. Doing so is not political correctness—it is patientcentered care.'

Note 1.

Quoted passages taken CMAJ, 187, no. 3 (17 February 2015): 162.

2.7 Study Questions 1. What is meant by "decision-making capacity"? 2. Who decides for those who lack decision-making capacity? According to what standards or criteria? 3. Anyone who has had a surgery and a flu shot understands that there is a difference in what risks are disclosed. What reasons can be given for this difference, and do you think they justify the difference in disclosure? There is likewise a difference

95

II

UM

111111

2 Medical Decision-Making

in what risks are disclosed when surgery is recommended and a medication prescribed. Do you think that standards of disclosure should be different for surgery and medication? 4. How do instructional advance directives differ from proxy directives? What are the advantages and risks of instructional advance directives? Under what circumstances, if any, would you make an instructional advance directive? 5. Under what conditions does the Canadian Paediatric Society think that life-sustaining treatment can be withdrawn or withheld from infants? How does this differ from the view of the American Medical Association? Which view do you prefer? Do you think that families should be able to WLST from their child when suffering is severe, as the Royal College allows? Would you qualify this in any way, and if so, how? 6. How much should health care providers respect the cultural or religious views of patients and families? Should female circumcision be allowed? Should male circumcision? To what extent should the views of Jehovah's Witnesses or Christian Scientists be taken into account in making medical decisions?

2.8 Suggested Further Reading Medical Decision-Making Brock, Dan W. "What Is the Moral Authority of Family Members t o Act a s Surrogates for Incompetent Patients?" Millbank Quarterly 74 (1996): 599-618. Gawande, Atul. "Whose Body Is It Anyway?" The New Yorker 4 October 1999,84-91. Groopman, Jerome. How Doctors Think (New York: Houghton Mifflin Harcourt, 2007). Jonsen, Albert R., Mark Siegler, and William J. Winslade. Clinical Ethics: A Practical Approach to Ethical Decisions in Clinical Medicine. Updated in many subsequent editions (New York: McGraw-Hill, 1992). Marway, Herjeet, and Heather Widdows. 2015. "Philosophical Feminist Bioethics: Past, Present, and Future." Cambridge Quarterly of Healthcare Ethics 24, no. 2 (April): 165-83. Ross, Lainie Friedman. 1998. Children, Families, and Health Care Decision-Making (New York: Oxford University Press, 1998). Sullivan, Mark D., and Stuart J. Youngner. "Depression, Competence, and the Right to Refuse Livesaving Medical Treatment." American Journal of Psychiatry 151 (July 1994): 971-8. Weir, Robert F, and Charles Peters. "Affirming the Decisions Adolescents Make about Life and Death." Hastings Center Report 27 (November/December 1997): 29-40. Advance Directives Brock, Dan W. "A Proposal for the Use of Advance Directives in the Treatment of Incompetent Mentally Ill Persons." Bioethics 7 (April 1993): 247-56. Browne, Alister, and Bill Sullivan. 2006. "Advance Directives in Canada." Cambridge Quarterly of Healthcare Ethics 19, no. 3 (summer): 256-60. Dresser, Rebecca. 1994. "Confronting the 'Near Irrelevance' of Advance Directives." Journal of Clinical Ethics 5: 55-6. King, Nancy. Making Sense of Advance Directives (Dordrecht: Kluwer Academic Publishers, 1991) Lynn, Joanne. 1991. "Why I Don't Have a Living Will." Law, Medicine & Health Care 19, no. 1-2: 101-4. Olick, Robert S. 2001. Taking Advance Directives Seriously: Prospective Autonomy and Decisions Near the End of Life (Washington: Georgetown University Press, 2001).

Suggested Further Reading

Teno, Joan M., Joanne Lynn, et al. 1994. "Do Formal Advance Directives Affect Resuscitation Decisions and the Use of Resources for Seriously Ill Patients?" Journal of Clinical Ethics 5 (1994): 23-30 (with following commentary). Informed Consent

Bok, Sissela. "Shading the Truth in Seeking Informed Consent." Kennedy Institute of Ethics Journal 5 (1995): 1-17. Buchanan, Allen E., and Dan W. Brock. Deciding for Others: The Ethics of Surrogate Decision Making (Cambridge, MA: Cambridge University Press: 1989). Dolgin, Janet L. "The Legal Development of the Informed Consent Doctrine: Past and Present." Cambridge Quarterly of Healthcare Ethics 19, no. 1 (January 2010): 97-101. Faden, Ruth R. "Informed Consent and Clinical Research." Kennedy Institute of Ethics Journal 6 (1996): 356-9. and Tom L Beauchamp. A History and Theory of Informed Consent (New York: Oxford University Press, 1986). Gostin, Lawrence 0. "Informed Consent, Cultural Sensitivity, and Respect for Persons." Journal of the American Medical Association 274 (13 September 1995): 844-5. Howe, Edmund G. "Leaving Laputa: What Doctors Aren't Taught about Informed Consent." The Journal of Clinical Ethics 11 (2000): 3-13. Kondo, Douglas G., E Marian Bishop, and Jay A. Jacobson. "Residents' and Patients' Perspectives on Informed Consent in Primary Care Clinics." Journal of Clinical Ethics 11 (2000): 39-48. Meisel, Alan, and Mark Kuczewski. "Legal and Ethical Myths about Informed Consent." Archives of Internal Medicine 156 (December 1996): 2521-6. Veatch, Robert M. "Abandoning Informed Consent." Hastings Center Report 25 (MarchApril 1995): 477-99.

111

?Nfranagernent of viledka[ [Inforrinatjon 3.1 Introduction Patients and their health care providers have a special relationship. Health care providers should deliver health benefits to patients. They should do this while respecting patient autonomy, avoiding harm, and acting justly to all those concerned with a patient's well-being. But maximizing health benefits sometimes conflicts with these other values. For example, telling a patient that he or she has terminal cancer may have harmful effects on the patient's general well-being. Telling a patient his or her diagnosis may also be something frowned on in the patient's culture of origin. The patient's family, concerned with the patient's wellbeing, may say that they and not the patient are the ones to be given information about the diagnosis and possible treatments. Another patient may insist that information about their medical condition is private and not to be shared with others, even other family members. What should a health care provider do in these circumstances? Questions like this are about the management of patient medical information and are the topic of this chapter.

Access to Information Therapeutic Privilege

therapeutic privilege the idea that a medical professional may deceive or withhold the truth from a patient when providing the truth will produce harm.

As we are told in Chapter 2, there are two models or conceptions of how information is to be controlled in the relationship between health care providers and their patients. The traditional idea, known as therapeutic privilege, is that everything is in the hands of medical professionals. This conception denies that a patient has a right to complete and correct information about the diagnosis or treatments available or expectations for recovery after treatment. Complete disclosure might worry the patient unduly or even cause serious harm. Telling only part of the truth or even lying can sometimes be good medicine. A health care professional should decide what information is to be made available to the patient based on a judgment of what will advance good care. A more recent conception sees a health care professional as someone with professional skills and judgment but also a duty to make full disclosure to the patient. A medical professional on this model may not lie or withhold information from a competent patient. As Chapter 2 emphasizes, this is essential for a patient making an informed decision about the risks and benefits of proposed treatments. Patients with a terminal illness not only need accurate information to make choices about medical care but also how to use the remaining part of their lives.

Access to Information

Lying and Withholding the Truth We should begin with a distinction. •

•

A lies to B = A communicates in some way something to B that A believes to be false, with the goal of deceiving someone (usually B) and in circumstances in which A could have communicated what A believes to be true. A withholds the truth from B = A does not communicate something to B that A believes to be true, either with the goal of deceiving someone (usually B) or with the goal of leaving someone (usually B) in ignorance and in circumstances in which A could have communicated what A believes to be true.

Are there ever instances when a physician or other medical professional does one of the above without doing the other? The most commented on instances involve the use of placebos in treatment. Suppose a doctor prescribes something for a patient that she knows is just a placebo. The doctor might tell the patient that the treatment is not a placebo; the doctor lies. But the doctor might instead not say anything one way or the other or might say something vague like "This treatment has proved to be effective in your kind of case." If the patient is left with the belief that the treatment is not a placebo, then the doctor has withheld the truth but not lied. Does the difference between one way of deceiving a patient (lying) and the other (withholding the truth) make for a moral difference? It is easy to think that lying is morally worse than withholding the truth. One is actively doing something while the other is merely passively letting something happen. To return to the example of a placebo treatment, a doctor who tells a patient that a treatment is not a placebo when it really is seems to be behaving in a morally worse way than one who manages to allow a patient to think it is not a placebo without actually saying so. But a minor change in James Rachels's famous argument against the idea that active killing is morally worse than passively letting someone die in "Active and Passive Euthanasia" (see Chapter 6, page 301) should lead us to doubt that these forms of deception—one active and the other passive—are ethically that different. Returning to the example of a placebo treatment, in both cases the patient agreeing to the treatment is not fully informed by his or her doctor to the extent required by the standards outlined in Chapter 2.

placebo a treatment all of whose effects are placebo effects; placebo effects are a good or bad effect of a treatment that occurs because the recipient believes the treatment will have the effect.

LA.

The Ethics of Lying Is it ever proper for a medical professional to lie to a patient, at least when the intent is -to do good for the patient, as the traditional model of the patient—professional relationship says it is? The simplest answer to this question comes from philosopher Immanuel Kant in the reading "On the Supposed Right to Lie from Benevolent Motives" (see page 106). According to a common (though not universally agreed upon) interpretation of Kant's essay, it is always wrong to lie to anyone about anything, no matter what the consequences and no matter what the ultimate goal of the deception.' On this understanding of what Kant is saying, there is nothing special about the relationship between a doctor and patient or the truth about a patient's medical condition when it comes to this issue. Why is it always wrong to deceive anyone or at least a patient about his or her diagnosis or prognosis and leave them not fully informed? Kant seems to be saying that lying to anyone even once about the truth would make it impossible to have any confidence that anyone is ever being honest rather than deceitful. In Kant's own words, by telling a lie, "I cause that declarations in general find no credit" (see page 106). This statement seems an exaggeration. If one person tells a lie to a patient, why would that by itself alter the believability of every other person? Kant might have in mind his idea that the maxim,

C /

maxim Kant's name for a personal rule for acting r a certain way. On some ideas aboutwhat Kant means, this must include a statement of the goal of acting and the kind of action that will achieve the goal.

YU I

ai

II 'al

iuw

3 Management of Medical Information

or personal rule, governing a choice must be one that the person telling the lie is able to successfully require everyone to follow. For example, the maxim or rule that doctors who keep the truth from a patient might be following is "If it would produce good results or avoid bad ones, then a patient (and anyone who might tell the patient) should be deceived about his or her true medical condition." Could the doctors require that everyone follow this rule? Would it be selfdefeating or impossible in some way for medical professionals and everyone in contact with a patient to follow this rule all the time? Well, if every one of them all the time followed this maxim or rule, no patient would be successfully deceived. The attempts at deception would be discovered frequently enough that no patient would give any "credit," to use Kant's word, to what they are told by the medical staff treating them or to what anyone else tells them about their medical condition. So the maxim is not universalizable, and deception is wrong. This thinking, however, tells us directly only about deception by lying. It does not tell us about deception achieved by withholding the truth. On the other hand, if these two types of deception are morally equal as suggested above, Kant's reasoning does say that withholding the truth as a way of deceiving a patient is also wrong. Are there any arguments for the conclusion that lying is sometimes the right thing for a medical professional to do? David Thomasma in "Telling the Truth to Patients: A Clinical Ethics Exploration" (see page 107) and Robert Stewart in "Telling the Patients the Truth" (see page 110) list many cases in which medical professionals may be doing the right thing when they deceive a patient or keep a patient ignorant of his or her true medical state. One case listed as one in which physicians are allowed or even morally required to deceive involves the harm that would result from the patient being informed. As mentioned above, deception is allowed on the traditional conception of the patient— professional relationship as a therapeutic privilege. The harm need not be medical harm; it could be psychological or emotional. But recent research (briefly mentioned in "Ethical Relativism in a Multicultural Society" by Ruth Macklin in Chapter 2; see page 83) concludes that harms of this kind really don't result from patients being told the truth. So this argument for deceiving a patient or even keeping a patient in the dark (and not fully informed about their diagnosis or the chances of their medical condition being successfully treated) won't work. Macklin also reports that doctors who have trained in countries other than the US or Canada say that in their country, medical professionals are expected to provide the truth to members of the patient's family but not to the patient himself or herself—at least when the diagnosis or prognosis is bad (as one for terminal lung cancer might be). A reason for this stance in these other countries and cultures, however, is that it may be understood and accepted that members of the family become the substitute decision-makers when a patient is seriously ill, even if the patient is still competent enough to understand what is happening and to make his or her own decisions about treatments.

Access to Medical Records The traditional model of the relationship between a medical professional and a patient denies a patient access to his or her medical records or chart. Sometimes these records contain statements of ideas considered by a professional as part of the process of diagnosing an illness. These statements can be a source of worry to a layperson patient. They also often contain technical jargon not everyone understands. But as Arthur Schafer, in "Medical Secrecy: Patients' Right of Access to Medical Records" (see page 113) points out, there are many good reasons to allow a patient access to these records. For example, such access is essential if a patient is to make an informed decision about authorizing the release of confidential information in the records to third parties.

Confidentiality

Confidentiality The Duty of Confidentiality Dave Unger's article, "Breaching Confidentiality," clearly explains (see page 128) that the duty of confidentiality generally means a physician or other health care professional cannot reveal any medical information about their patients to others without the patient's consent. The professional duty of confidentiality covers not only what patients reveal to doctors and other medical professionals during the course of treatment but also what they conclude based on their examination or assessment of a patient. It covers all medical records (including x-rays, lab reports, and so on) as well as direct communications between a patient and a doctor or others. Keeping confidences is one of the oldest principles of medical ethics. In the ancient Greek Hippocratic Oath, new physicians promise this: About whatever I may see or hear in treatment, or even outside of treatment, in the life of men and women that ought not to be spoken abroad, I will remain silent, holding such things to be unutterable.2 A newer version of the code, written by the dean of the Tufts University medical school, has new doctors promise this: I will respect the privacy of my patients, for their problems are not disclosed to me that the world may known

Exceptions to the Duty But almost all professional medical ethics codes as well as Canadian laws say that this duty to keep patient medical information secret and out of the hands of others is not absolute. According to the Canadian Nurses Association code of ethics, for example, nurses are acting properly when they disclose patient health information even without the patient's consent to those who need to know it.3 (The code does not say how to decide who needs to know) The code of ethics for Canadian psychologists says that information learned in confidence from a patient must be passed on when keeping it confidential would lead to serious harm to others or loss of their lives.4 The Canadian Medical Association tells its members that they should generally keep their patients' personal health information confidential; nevertheless, they may disclose it to others when the law requires them to do so.' The code also says that doctors should make sure that their patients understand that there are these exceptions to complete confidentiality. But what is the best policy here? Should the duty to respect confidences be absolute— without exceptions—or not? There are several possibilities, and some are considered in the readings in this chapter. Here is one: Unqualified confidentiality: Patients should understand that no information about them obtained during treatment will be passed on to anyone else under any circumstances without the consent of the patient. This policy is supported by Kenneth Kipnis in "A Defense of Unqualified Medical Confidentiality" (see page 117). Kipnis sometimes gives it the label "absolute confidentiality." He thinks better consequences for everyone will result from following this rule about confidentiality than by following any other possible rule. Kipnis calls other possible rules about confidentiality "qualified." Here is a statement of a general rule that qualifies confidentiality.

Hippocratic Oath this ancient Greek text requires a new physician to swear to the healing gods to follow specific ethical standards.

II

ill

11111,1

3 Management of Medical Information

Qualified confidentiality: Patients should understand that some information about them obtained during treatment will be passed on to others under some circumstances even without the consent of the patient. In theory, there are many qualifications or exceptions that fill in the details of this general rule. Here are three examples: Example 1: Patients should understand that some information about them obtained during treatment will be passed on to others in some circumstances even without their consent. These circumstances are ones in which passing on the information will prevent serious harms to others. Example 2: Patients should understand that some information about them obtained during treatment will be passed on to others in some circumstances even without their consent. These circumstances include ones in which the law requires that the information be passed on. Example 3: Patients should understand that some information about them obtained during treatment will be passed on to others in some circumstances even without their consent. These circumstances include ones in which passing on the information will benefit the public. The first two of these qualified rules can be found in Canadian law and professional codes as well as in law and professional codes in many other countries. The third is found in British law:6 Supporters of all three rules believe that the best, consequences will follow from their favoured rule. Consider the last qualified rule (example 3). What are the benefits to the public that are supposed to come about when information acquired during the treatment of patients is passed on without their consent? An important source of data for medical research will be opened up; surely this is a significant benefit, as Bonnie Kaplan notes in "How Should Health Data Be Used?" (see page 134). Other benefits include the production of reliable health statistics, such as how long the average Canadian might be expected to live or what percentage of people living in a certain area are HIV+. Big Data A final issue about the confidentiality of medical information is a particularly modern one. It arises when patient medical information is stored in a way that can be made easily available for what are sometimes called "secondary uses." The primary purpose or use for medical information is to benefit the patient, to help in his or her treatment. Pretty much anything else is a secondary use. Many of these secondary uses of medical data are realistically available only when medical records are stored electronically and can be processed and combined with other information stored by computers. Practically speaking, it would be impossible for such big amounts of data to be processed manually, so the issues arising from storing medical information electronically is a modern concern. Secondary uses of information from a patient's medical record include medical research (as mentioned above), the collection of health statistics (as mentioned above), and the management of health care costs. These uses are forbidden under the absolute confidentiality rule without the patient's explicit consent and are the sorts of things allowed by some of the qualified rules even without this consent. Another secondary use is allowing someone or a private company not primarily concerned with the health of the patient to make money. This is what happened in the two legal cases, one from the US and the other from the UK, mentioned by Kaplan in her article. In both cases, pharmacists sold the prescription information they had about patients to what are called "data mining" companies. These companies mine data in the same way

Confidentiality

that other companies mine the earth. They summarize the data or put it together with other information they have access to. In these two cases, the data mining companies weren't interested in what the prescription records said about the patients. (In fact, the prescription information sold to them by the pharmacists hid patient information in a way described in the following paragraphs.) Rather, the data mining companies were interested in the behaviour of the medical professionals writing the prescriptions. The records contained the names and other identifying information about these medical professionals. The companies searched the records for data about w.bia_doctors prescribed which drugs and how often. Then the data was sold to drug manufacturing companies. In_turn, the_ drug manufacturers used the information to help target their marketing and sales efforts. Kaplan points out that confidentiality practices commonly require medical records used in these secondary ways to be anonymized or de-identified. The idea is that when medical information is stripped of personally identifying information, any secondary use no longer breaks even the most absolute standard for confidentiality. Unfortunately, this anonymization is not always effective, given all the information available in public or other records about almost everyone in Canada. Imagine that a hospital or doctor's office has a number of electronic medical records that look like those in Table 3.1 below. These records (rows in the table) contain confidential medical information in the last column and the person's name in column 1. Table 3.2 shows the result of anonymization. The table still has information useful for medical and social science researchers and administrators. What do we now need to learn from Table 3.2 that it was John who had the symptoms of wheezing? If we know from elsewhere—for example, from public census data or the kind of information people provide every day when they complete forms on the Web—that there is someone who was born on 17 April 1994 and is male and lives in the postal code forward sortation area V6T, then the chance is better than 87 per cent that this is John.' If Table 3.2 contained the full Canadian postal code, then we could determine that it is John who is N3 with almost complete certainty.' Table 3.2 tells us the medical complaint of this individual. So we know—with a very high probability—that this was John's medical problem. Confidentiality of medical records is therefore not always completely protected by the common practice of anonymization.

Table 3.1 Data with Identifying Details Name

Birth Date

Sex

Visit Date

Postal Code

Medical Issue

Sean Phillips

21 Mar 1995

Male

14 Apr 2016

V6T1Z1

Fever

May Ong

10 Nov1994

Female

15 Apr 2016

V6Y2Z6

Chest pain

John Scatopolus

17 Apr 1994

Male

15 Apr 2016

V6T1Z1

Wheezing

Lisa Friedrich

01 Dec 1993

Female

16 Apr 2016

V6Y2Z6

Aching joints

Table 3.2 Data with Identifying Details Removed Name

Birth Date

Sex

Postal Code FSA

N1

21 Mar 1995

Male

V6T

Fever

N2

10 Nov 1994

Female

V6Y

Chest pain

N3

17 Apr 1994

Male

V6T

Wheezing

N4

01 Dec 1993

Female

V6Y

Aching joints

Medical Issue

II all

I

1111.1J

3 Management of Medical Information

Genetic Information Here are two examples of genetic tests that provide medically relevant information: The BRCA test. Using a blood sample, the test analyzes DNA to identify harmful changes (mutations) in either one of two genes, BRCA1 and BRCA2. Men and women with these harmful changes have a significantly greater risk of developing breast cancer and, in women, ovarian cancer compared to the average. Inherited BRCA gene mutations are responsible for 5 to 10 per cent of breast cancers and about 15 per cent of ovarian cancers. The Huntington's Disease (HD) test. Using a blood sample, the test analyzes DNA to identify harmful numbers of CAG repeats in parts of the Huntington gene. Individuals with 28 or fewer repeats usually will be free of the disease. Those with 40 or more repeats usually will have the disease, with more repeats meaning that the disease symptoms will start at a younger age.

genetic exceptionalism the view that genetic information is a special type of medical information and must therefore be treated with greater care—for example, with special laws--than other health care information.

In the US there are special laws giving greater protections to genetic health information like the results of the BRCA or HD tests than to other kinds of health information,8 a practice known as genetic exceptionalism. Some of these laws forbid discrimination by insurance companies or employers based on the results of these tests. For example, someone with a higher risk of breast cancer because of a BRCA mutation cannot be refused medical insurance or fired from a job because of this. Some of these laws in the US require that hospitals, doctors' offices, the providers of genetic testing services, and other places where this information might be created or stored make special efforts to keep it confidential. Canada currently does not have any laws that treat the confidentiality of genetic information like results of the BRCA or HD or other genetic tests differently from other personal health information. Are the results of genetic tests like the BRCA or HD test so different that they should be given special treatment compared to the results of, for example, a pregnancy test or a prostate cancer screening test? This is the topic of Thomas H. Murray's "Genetic Exceptionalism and 'Future Diaries'" (see page 147). Genetic test results can have what seem like unique and almost paradoxical consequences. Consider A and her genetically identical twin sister B. (Even identical twins are not completely genetically identical, but we'll suppose A and B are in this scenario.) A would like to have a test for the BRCA mutation done. Her family—and of course B's family—have a history of breast and ovarian cancer. Her cancer doctor thinks having this test is a good choice. It would also seem that A is perfectly within her rights to acquire this kind of medical information about herself. Now, if A has the test, she can figure out medical information about her genetically identical twin sister B. She can do this because she knows that B is her genetically identical twin and not because she peeks at B's confidential medical records. But this is private information about B's health state, information requiring B's consent to find out. So A may not have the test providing her with information about her own medical status without B's consent. Most of the time, however, another person like B will not be As exact genetic twin. But often A will be able to determine the chance that B has a particular genetic medical issue. Suppose A had the test for the Huntington's Disease mutation long before she has or might have any symptoms. If A finds she has a harmful number of CAG repeats, then she can figure out that her biological full brothers and sisters have at least a 50 per cent chance of having the harmful mutation as well. If this is information that any one of her brothers or sisters might not want anyone else to know, then the concern for the confidentiality and privacy of their medical state forbids A from finding this out about herself.

Genetic Information

Is any of this unique to medical information that comes from the analysis of someone's genetic makeup? Not really, says Murray in his article. Suppose A has a test for an HIV infection. She tests positive. A has a partner, C. Since A and C do not use protection methods that would prevent C from being infected by A, it is almost certain that C is infected too. But this is information about C's health that C might want to keep private. So again A may not find out information about her own HIV status, at least not without C's permission. Or suppose A and her sister B grew up in Waltham, Massachusetts, a suburb of Boston, in the 1980s. They both drank lots of water supplied by the city. This water was heavily contaminated with chemicals that significantly increased the risk of leukemia. Since A knows that B drank this city water, A knows something about B's risk for this kind of cancer. Is A allowed to figure this out about B or even herself without B's consent? One reason why health information based on a person's DNA is said to be special is its prophetic character, as Murray emphasizes: it tells us something about a person's likely future health. Someone with a harmfully large number of CAG repeats in his or her Huntington gene will get Huntington's Disease. But not all genetic information is like this. The presence of a harmful BRCA mutation doesn't mean that breast cancer is certain. It does mean, though, that the chance of getting breast cancer is much greater than average. Anyone who has one copy of the E4 mutation of the APOE gene—something else that can be found through DNA analysis—has a three times greater risk of developing Alzheimer's disease than someone with no copies of the mutation. Having two copies means that the risk is 15 times greater. But these predictions are not special to genetic information. Before treatments were available to keep multiplication of the virus that causes AIDS in check, anyone infected with HIV was almost certainly destined to have AIDS in the near future. It seems that genetic medical information is often not that different from other kinds of medical information, and there is no reason, as Murray concludes, to give it special legal protections.

Notes 1.

2. 3. 4. 5. 6. 7.

8.

Helga Varden, "Kant and Lying to the Murderer at the Door," Journal of Social Philosophy 41 (2010), 403-21; Christine Korsgaard, "The Right to Lie: Kant on Dealing with Evil," Philosophy and Public Affairs 15 (1986), 325-49; Thomas L. Carson, Lying and Deception: Theory and Practice (New York: Oxford University Press, 2010), ch. 3. Steven H. Miles, The Hippocratic Oath and the Ethics of Medicine (New York Oxford University Press, 2004). Canadian Nurses Association, Code of Ethics for Registered Nurses (Ottawa: Canadian Nurses Association, 2008). Canadian Psychological Association, Canadian Code of Ethics for Psychologists, 3r d edn. (Ottawa: Canadian Psychological Association, 2000). Canadian Medical Association, cma Code of Ethics, update 2004 (Ottawa: Canadian Medical Association, 2015). J. O'Brien and 0. Chantler, "Confidentiality and the Duties of Care," Journal of Medical Ethics 29 (2003), 36-40. The estimate is based on a study by privacy researcher Latanya Sweeney for the US. Latanya Sweeney, "Simple Demographics Often Identify People Uniquely," Carnegie Mellon University Data Privacy Working Paper 3 (Pittsburgh, 2000). The study estimated that there is a slightly better than 87 per cent chance of identifying a person in the US using his or her birthdate, sex, and zip code. A US zip code has about the same average number of residents as a Canadian postal code forward sortation area. A full Canadian postal code has a much smaller population (less than 1 per cent of a forward sortation area). Using Sweeney's calculations, identification of a person is nearly certain given just that person's birthdate, sex, and full Canadian postal code. Department of Health and Human Services, "Genetic Information Nondiscrimination Act of 2008" (Washington: Department of Health and Human Services, 2009).

I I IA

106

111111

3 Management of Medical Information

3.2 Patient Access to Information On the Supposed Right to Lie from I3enevolent Motives Immanuel Kant In the work called France, for the year 1797, Part VI. No. 1, on Political Reactions, by Benjamin Constant, the following passage occurs, p. 123:— "The moral principle that it is one's duty to speak the truth, if it were taken singly and unconditionally, would make all society impossible. We have the proof of this in the very direct consequences whi:h have been drawn from this principle by a German philosopher, who goes so far as to affirm that to tell a falsehood to a murderer who asked us whether our friend, of whom he was in pursuit, had not taken refuge in our house, would be a crime." The French philosopher opposes this principle in the following manner, p. 124:— "It is a duty to tell the truth. The notion of duty is inseparable from the notion of right. A duty is what in one being corresponds to the right of another. Where there are no rights there are no duties. To tell the truth then is a duty, but only towards him who has a right to the truth. But no man has a right to a truth that injures others." The 7rpofitru cpaic6os here lies in the statement that "To tell the truth is a duty, but only towards him who has a right to the truth." It is to be remarked, first, that the expression "to have a right to the truth" is unmeaning. We should rather say, a man has a right to his own truthfulness (veracitas), that is, to subjective truth in his own person. For to have a right objectively to truth would mean that, as in meum and tuum generally, it depends on his will whether a given statement shall be true or false, which would produce a singular logic. Now, the first question is whether a man—in cases where he cannot avoid answering Yes or No— has the right to be untruthful. The second question is whether, in order to prevent a misdeed that threatens

him or someone else, he is not actually bound to be untruthful in a certain statement to which an unjust compulsion forces him. Truth in utterances that cannot be avoided is the formal duty of a man to everyone, however great the disadvantage that may arise from it to him or any other; and although by making a false statement I do no wrong to him who unjustly compels me to speak, yet I do wrong to men in general in the most essential point of duty, so that it may be called a lie (though not in the jurist's sense), that is, so far as in me lies I cause that declarations in general find no credit, and hence that all rights founded on contract should lose their force; and this is a wrong which is done to mankind. If then, we define a lie merely as an intentionally false declaration towards another man, we need not add that it must injure another, as the jurists think proper to put in their definition (mendacium est falsiloquium in prajudicium alterius). For it always injures another, if not another individual, yet mankind generally, since it vitiates the source of justice. This benevolent lie may, however, by accident (casus) become punishable even by civil laws; and that which escapes liability to punishment only by accident may be condemned as a wrong even by external laws. For instance, if you have by a lie hindered a man who is even now planning a murder, you are legally responsible for all the consequences. But if you have strictly adhered to the truth, public justice can find no fault with you, be the unforeseen consequence what it may. It is possible th4 whilst you have honestly answered Yes to the murderer's question, whether his intended victim is in the house, the latter may have gone out unobserved, and so not have come in the way of the murderer, and the deed therefore have not been done; whereas, if you lied and said he was not in the house, and he had really gone out (though unknown to you) so that the murderer met him as he went, and executed his purpose on him, then you might with justice be accused as the cause of his death. For, if you had spoken the truth as well as you knew it, perhaps the murderer while seeking for his enemy in the house might have been caught by neighbours coming up and the deed been prevented. Whoever then tells a lie, however good his intentions may be, must answer for

Source: From Kant's Critique of Practical Reason and Other Works on the Theory of Ethics, trans. T.K. Abbott (London: Longmans, Green and Co., 1898). © SophiaOmni 2012.

Kant: On the Supposed Right to Lie from Benevolent Motives

the consequences of it, even before the civil tribunal, and must pay the penalty for them, however unforeseen they may have been; because truthfulness is a duty that must be regarded as the basis of all duties founded on contract, the laws of which would be

Telling the Truth to Patients: A Clinical Ethics Exploration David C. Thomasma

Reasons for Telling the Truth

In all human relationships, the truth is told for a myriad of reasons. A summary of the prominent reasons is that it is a right, a utility, and a kindness. It is a right to be told the truth because respect for the person demands it. As Kant argued, human society would soon collapse without truth-telling, because it is the basis of interpersonal trust, covenants, contracts, and promises. The truth is a utility as well, because persons need to make informed judgments about their actions. It is a mark of maturity that individuals advance and grow morally by becoming more and more self-aware of their needs, their motives, and their limitations. All these steps toward maturity require honest and forthright communication, first from parents and later also from siblings, friends, lovers, spouses, children, colleagues, co-workers, and caregivers.' Finally, it is a kindness to be told the truth, a kindness rooted in virtue precisely because persons to whom lies are told will of necessity withdraw from important, sometimes life-sustaining and life-saving relationships. Similarly, those who tell lies poison not only their relationships but themselves, rendering themselves incapable of virtue and moral growth.2 . . . Overriding the Truth . . . Not all of us act rationally and autonomously at all times. Sometimes we are under sufficient stress that others must act to protect us from harm. This is called necessary patemalism.,Should we become seriously ill, others must step in and rescue us if we are incapable of doing it ourselves. . . .

rendered uncertain and useless if even the least exception to them were admitted. To be truthful (honest) in all declarations is therefore a sacred unconditional command of reason, and not to be limited by any expediency.

In General Relationships

In each of the three main reasons why the truth must be told, as a right, a utility, and a kindness, lurk values that may from time to time become more important than the truth. When this occurs, the rule of truthtelling is trumped, that is, overridden by a temporarily more important principle. The ultimate value in all instances is the survival of the community and/or the well-being of the individual. Does this mean for paternalistic reasons, without the person's consent, the right to the truth, the utility and the kindness, can be shunted aside? The answer is "yes." The truth in a relationship responds to a multivariate complexity of values, the context for which helps determine which values in that relationship should nredominate. Nothing I have said thus far suggests that the truth may be treated in a cavalier fashion or that it can be withheld from those who deserve it for frivolous reasons. The only values that can trump the truth are recipient survival, community survival, and the ability to absorb the full impact of the truth at a particular time. All these are only temporary trump cards in any event. They can only be played under certain limited conditions because respect for persons is a foundational value in all relationships. In Health Care Relationships

It is time to look more carefully at one particular form of human relationship, the relationship between the doctor and the patient or sometimes between other health care providers and the patient. Early in the 1960s, studies were done that revealed the majority of physicians would not disclose a diagnosis of cancer to a patient. Reasons cited were mostly those that derived from nonmaleficence. Physicians were concerned that such a diagnosis might disturb the equanimity of a patient and might lead to desperate acts. Primarily physicians did not want to destroy their patients' hope. By the middle 1970s, however, repeat studies brought to light a radical shift in physician attitudes. Unlike earlier views, physicians now

Source: From Cambridge Quarterly of Healthcare Ethics 3 (1994), 375-82. Copyright 1994 Cambridge University Press. Reprinted with permission of the Cambridge University Press.

II

111413

3 Management of Medical Information

emphasized patient autonomy and informed consent temporarily in favour of more important long-term over paternalism. In the doctor—patient relation, this values, which are known in the relationship. Finally, the goal of health care relations is treatment meant the majority of physicians stressed the patient's right to full disclosure of diagnosis and prognosis. of an illness. An illness is far broader than its subset, disOne might be tempted to ascribe this shift of atti- ease. Illness can be viewed as a disturbance in the life of tudes to the growing patients' rights and autonomy move- an individual, perhaps due to many non-medical factors. ments in the philosophy of medicine and in public affairs. A disease, by contrast, is a medically caused event that No doubt some of the change can be attributed to this may respond to more interventionist strategies.5 movement. But also, treatment interventions for cancer Helping one through an illness is a far greater perled to greater optimism about modalities that could offer sonal task than doing so for a disease. A greater, more some hope to patients. Thus, to offer them full disclosure enduring bond is formed. The strength of this bond of their diagnosis no longer was equivalent to a death sen- may justify withholding the truth as well, although in tence. Former powerlessness of the healer was supplanted the end "the truth will always out." with technological and pharmaceutical potentialities. A more philosophical analysis of the reasons for a Clinical Case Categories shift conies from a consideration of the goal of medicine. The goal of all healthcare relations is to receive/ The general principles about truth-telling have been provide help for an illness such that no further harm reviewed, as well as possible modifications formed is done to the patient, especially in that patient's vul- from the particularities of the health care professional— nerable state.3 The vulnerability arises because of in- patient relationship. Now I turn to some contemporary creased dependency. Presumably, the doctor will not examples of how clinical ethics might analyze the_ biertake advantage of this vulnerable condition by adding archy of values surrounding truth-telling. to it through inappropriate use of power or the lack of There are at least five clinical case categories in compassion. Instead, the vulnerable person should be which truth-telling becomes problematic: intervention assisted back to a state of human equality, if possible, cases, long-term care cases, cases of dying patients, prefr,ee from the prior dependency.' vention cases, and non-intervention cases. 1 First, the goal of the health care giver—patient relation is essentially to restore the patient's autonomy. Intervention Cases Thus, respect for the right of the patient to the truth is Of all clinically difficult times to tell the truth, two typmeasured against this goal. If nothing toward that goal ical cases stand out. The first usually involves a mother can be gained by telling the truth at a particular time, of advanced age with cancer. The family might beg the still it must be told for other reasons. Yet, if the truth surgeon not to tell her what has been discovered for would impair the restoration of autonomy, then it may fear that "Mom might just go off the deep end." The be withheld on grounds of potential harm. Thus, the movie Dad, starring Jack Lemmon, had as its centregoal of the healing relationship enters into the calculus piece the notion that Dad could not tolerate the idea of _ of values that are to be protected. cancer. Once told, he went into a psychotic shock that Second, most healthcare relationships of an inter- ruptured standard relationships with the doctors, the , ventionis: character are temporary, whereas relationships hospital, and the family. However, because this diagnoinvolving primary care, prevention, and chronic or dying sis requires patient participation for chemotherapeutic care are more permanent. These differences also have a interventions and the time is short, the truth must be bearing cn truth-telling. During a short encounter with faced directly. Only if there is not to be intervention health care strangers, patients and health care provid- might one withhold the truth from the patient for a ers will of necessity require the truth more readily than while, at the family's request, until the patient is able during a long-term relation among near friends. In the to cope with the reality. A contract about the time alshort term, decisions, often dramatically important ones, lowed before telling the truth might be a good idea. need to be made in a compressed period. There is less opThe second case is that of ambiguous genitalia. portunity to manoeuvre or delay for other reasons, even if A woman, 19-years-old, comes for a checkup because she there are concerns about the truth's impact on the person. plans to get married and has not yet had a period. She is Over a longer period, the truth may be withheld very mildly retarded. It turns out that she has no vagina, for compassionate reasons more readily. Here, the pa- uterus, or ovaries but does have an undescended testicle tient and physician or nurse know one another. They in her abdomen. She is actually a he. Should she be told are more likely to have shared some of their values. In this fundamental truth about herself? Those who argue this context, it is more justifiable to withhold the truth for the truth do so on grounds that she will eventually

Thomasma: Telling the Truth to Patients: A Clinical Ethics Exploration

find out, and more of her subsequent life will have been ruined by the lies anddisingenuousness of others. Those who argue against the truth usually prevail. National standards exist in this regard. The young woman is told that she has something like a "gonadal mass" in her abdomen that might turn into cancer if not removed, and an operation is performed. She is assisted to remain a female. More complicated still is a case of a young Hispanic woman, a trauma accident victim, who is gradually coming out of a coma. She responds only to commands such as "move your toes." Because she is now incompetent, her mother and father are making all care decisions in her case. Her boyfriend is a welcome addition to the large, extended family. However, the physicians discover that she is pregnant. The fetus is about five weeks old. Eventually, if she does not recover, her surrogate decisionmakers will have to be told about the pregnancy, because they will be involved in the terrible decisions about continuing the life of the fetus even if it is a risk to the mother's recovery from the coma. This revelation will almost certainly disrupt current family relationships and the role of the boyfriend. Further, if the mother is incompetent to decide, should not the boyfriend, as presumed father, have a say in the decision about his own child? In this case, revelation of the truth must be carefully managed. The pregnancy should be revealed only on a "need to know" basis, that is, only when the survival of the young woman becomes critical. She is still progressing moderately toward a stable state. Long Term Cases

Rehabilitation medicine provides one problem of truthtelling in this category If a young man has been paralyzed by a football accident, his recovery to some level of function will depend upon holding out hope. As he struggles to strengthen himself, the motivation might be a hope that caregivers know to be false, that he may someday be able to walk again. Yet, this falsehood is not corrected, lest he slip into despair. Hence, because this is a long-term relationship, the truth will be gradually discovered by the patient under the aegis of encouragement by his physical therapists, nurses, and physicians, who enter his life as near friends. Cases of Dying Patients

Sometimes, during the dying process, the patient asks directly "Doctor, am I dying?" Physicians are frequently reluctant to "play God" and tell the patient how many days or months or years they have left. This reluctance sometimes bleeds over into a less-than-forthright answer to the question just asked. A surgeon with whom I make rounds once answered this question posed by a

terminally ill cancer patient by telling her that she did not have to worry about her insurance running out! Yet in every case of dying patients, the truth can be gradually revealed such that the patient learns about dying even before the family or others who are resisting telling the truth. Sometimes, without- directly saying "you are dying," we are able to use interpretative truth and comfort the patient. If a car driver who has been in an accident and is dying asks about other family members in the car who are already dead, there is no necessity to tell him the truth. Instead, he can be told that "they are being cared for" and that the important thing right now is that he be comfortable and not in pain. One avoids the awful truth because he may feel responsible and guilt-ridden during his own dying hours if he knew that the rest of his family were already dead. Prevention Cases

A good example of problems associated with truthtelling in preventive medicine might come from screening. The high prevalence of prostate cancer among men over 50-years-old may suggest the utility of cancer screening. An annual checkup for men over 40-years-old is recommended. Latent and asymptomatic prostate cancer is often clinically unsuspected and is present in approximately 30 per cent of men over 50 years of age. If screening were to take place, about 16.5 million men in the United States alone would be diagnosed with prostate cancer, or about 2.4 million men each year. As of now, only 120,000 cases are newly diagnosed each year. Thus, as Timothy Moon noted in a recent sketch of the disease, "a majority of patients with prostate cancer that is not clinically diagnosed will experience a benign course throughout their lifetime."6 The high incidence of prostate cancer coupled with a very low malignant potential would entail a whole host of problems if subjected to screening. Detection would force patients and physicians to make very difficult and life-altering treatment decisions. Among them are removal of the gland (with impotence a possible outcome), radiation treatment, and most effective of all, surgical removal of the gonads (orchiectomy). But why consider these rather violent interventions if the probable outcome of neglect will overwhelmingly be benign? For this reason the US Preventive Services Task Force does not recommen,c1 either for or against screening for prostate cancer.! Quality-of-life issues would take precedence over the need to know. Non-Intervention Cases

This last example more closely approximates the kind of information one might receive as a result of gene mapping. This information could tell you of the likelihood

II

1111i1

3 Management of Medical Information

or probability of encountering a number of diseases through genetic heritage, for example, adult onset or type II diabetes, but could not offer major interventions for most of them (unlike a probability for diabetes). Some evidence exists from recent studies that the principle of truth-telling now predominates in the doctor— patient relationship. Doctors were asked about revealing diagnosis for Huntington's disease and multiple sclerosis, neither of which is subject to a cure at present. An overwhelming majority would consider full disclosure. This means that, even in the face of diseases for which we have no cure, truth-telling seems to take precedence over protecting the patient from imagined harms. The question of full disclosure acquires greater poignancy in today's medicine, especially with respect to Alzheimer's disease and genetic disorders that may be diagnosed in utero. There are times when our own scientific endeavours lack a sufficient conceptual and cultural framework around which to assemble facts. The facts can overwhelm us without such conceptual frameworks. The future of genetics poses just such a problem. In consideration of the new genetics, this might be the time to stress values over the truth. Conclusion

First, truth is contextual. Its revelation depends upon the nature of the relationship between the doctor and patient and the duration of that relationship. Second, truth is a secondary good. Although important, other primary values take precedence over the truth. The most important of these values is survival of the individual and the community. A close second would be preservation of the relationship itself. Third, truth is essential for healing an illness. It may not be as important for curing a disease. That is why for example, we might withhold the truth from the woman with ambiguous genitalia, curing her disease (having a gonad) in favour of maintaining her health (being a woman). Fourth, withholding the truth is only a temporary measure. In vino, veritas, it is said. The truth will eventually come out, even if in a slip of the tongue. Its revelation, if it is to be controlled, must always aim at the good of the patient for the moment. At all times, the default mode should be that the truth is told. If for some important reason, it is not to be immediately revealed in a particular case, a truthmanagement protocol should be instituted so that all caregivers on the team understand how the truth will eventually be revealed.

Truth in the clinical relationship is factored in with knowledge and values. Notes 1. 2. 3.

Bok, S. 1989. Lying: Moral Choice in Public and Personal Life. Vintage Books: New York. Pellegrino, E.D., and Thomasma, D.C. 1993. The Virtues in Medical Practice. Oxford University Press: New York. Cassell, E. 1982. "The Nature of Suffering and the Goals of Medicine," in New England Journal of Medicine 306 (11): 639-45.

4.

5. 6.

See Nordenfelt, L., issue editor. 1993. "Concepts of Health and Their Consequences for Health Care," in Theoretical Medicine 14 (4). Moon, T.D. 1992. "Prostate Cancer," inJournal of the American Geriatrics Society 40: 622-7 (quote from 626). Ibid.

p; Telling Patients the Truth Robert Scott Stewart Introduction

David Thomasma (1994) suggests that there are three reasons why, in general, individuals ought to tell the truth and/or be told the truth: respect for persons, utility, and kindness. This research paper follows him in his assessment and discusses the issue of truth telling to patients within his model.

Respect for Persons

According to Immanuel Kant, respecting human dignity follows from the fact that humans have a special capacity to live autonomously. That is, humans live not only by laws that are imposed upon them, whether by nature or by legislatures, but humans also have the unique capacity to decide what laws or rules will govern them, which is the literal meaning of autonomy. To make decisions for others, then, is to _failto respect a person's autonomy. Over the past number of

Source: From Online Journal of Health Ethics 6 (2010), 1. Reprinted with permission of the Open Access Journal.

Stewart: Telling Patients the Truth

years in North America, society has seen the impact of this idea in many ways. In terms of models for the physician—patient relationship, society has moved from a paternalistic model to an informative one (Emanuel & Emanuel, 1992). In the paternalistic model, on the basis of his/her particular expertise, the physician was empowered to act in the best interests of his/her patients. As such, it was not uncommon for physicians, when working under this model, to withhold information from patients in some situations, such as when there was a bad prognoses, for the patienth "own good." Indeed, as Braddock notes (1998), "In 1961 only 10% of physicians surveyed believed it was correct to tell a patient of a fatal cancer diagnosis. . . . [By] 1979, 97% felt that such disclosure was correct." On the one hand, this is in keeping with what patients want. "For instance, 90% of patients surveyed said they would want to be told of a diagnosis of cancer or Alzheimer's disease" (Braddock, 1998). As health care has moved away from a paternalistic model toward an informative model over the past thirty years, the principle of patient autonomy has been taken more seriously (Emanuel & Emanuel, 1992). In this model, patients are conceived as having the right to be involved in decisions about their health care and indeed ultimately to provide informed consent for any treatment option. Despite disagreement regarding how much information is necessary for a person to have in order to provide such informed consent (Randall & Downie, 2006), clearly withholding information that the patient has a serious illness such as cancer, or that they are in fact palliative, is far beyond any reasonable standard of informed consent. Hence, if health professionals assume that cancer needs to be treated—even if this treatment is limited to palliative care—then morally and legally, informed consent must be provided, and it is impossible to achieve this when the patient's basic diagnosis and prognosis are withheld from him or her. (.-Th Hence, under the principle of respect for persons it is clear that at some point within the course of some process, which this paper discusses further below, competent patients must be informed of both their diagnosis and prognosis. Utility Utility concerns weighing costs and benefits and acting in such a way as to maximize benefits over costs. Traditional utilitarian theory equated costs with pain or unhappiness and benefits with pleasure or happiness (Bentham, 1789/2007; Mill, 1863/2007), and in contemporary bioethical theory utility is typically associated with the principles of non-maleficence and

beneficence, which obligate health care workers to "do no harm" on the one hand and to pursue positive outcomes on the other (Beauchamp & Childress, 2001). Under the paternalistic model, physicians often used this principle in order to justify withholding seriously bad diagnoses from patients, and patients' families continue to use this rationale when urging physicians and other health care workers to do so. Writing in 1979, physician Mack Lipkin employs this sort of reasoning in his defense of deceiving patients. In the first place, he argues, it is impossible for patients to be told the "whole truth" because they do not have the medical expertise to understand it. One result of this is that patients can actually be misinformed by information. "Cancer" or "heart trouble," which can range from curable and nonserious to incurable and fatal, can mean a vastly different array of things to different patients. Because patients are already anxiety ridden and vulnerable, they are likely to misunderstand the doctor's diagnosis and prognosis, and indeed, "[t] he news of serious illness drives some patients to irrational and destructive behavior" (Lipkin, 1979). This fact, he says, justifies withholding-information. The fact that there is a demonstrable-placebo effect that can have incredibly positive effects on a patient's health justifies a great deal of what society would now call spin, if not outright lying (Lipkin, 1979). As Cullen and Klein (2000) point out, however, there are several problems with Lipkin's argument. First, his claim about patients not possessing sufficient medical knowledge to understand fully their diagnosis and prognosis is disingenuous. There are many occupations where the expert has much more knowledge than the customer—from lawyers to car mechanics and computer software specialists. This does not give these professionals the right to withhold information or "spin" the truth: rather, it obligates them to be honest and to find ways in which to articulate points in ways the individual will understand. Most importantly, however, in attempting to promote a patient's good, arguments such as Lipkin's completely ignore the respect owed to patients as autonomous beings. People faced with imminent death have a right to that information, and, moreover, it may actually be good for them to know their prognosis in at least two ways. First, "most treatments for serious diseases require the full cooperation of the patient" (Cullen & Klein, 2000). Second, in cases where no cure is at all likely, "good" for a patient may consist in planning for his or her death. This might include a variety of activities from entering palliative care, securing child care for dying patients who are also young parents, reconnecting with lost loved ones, to making out a final will. Hence, Cullen and Klein argue, telling patients the truth ought to be considered the "default position," which

I I !III

llllll

3 Management of Medical Information

can be overridden only in special cases and/or for a limited amount of time. cAt times, patients themselves may explicitly request thdt they not be told the truth about their condition if it is very dire. Withholding the truth in such a case as this is actually an instance of respecting the patient, not overriding it for some other purpose. At other times, deceiving a patient may be justified, in the short term, in order to have the best chance to restore a patient to health. This would require that the physician know his/her patient's personality well enough that he or she know that the patient cannot handle bad news and that telling him or her such bad news will lessen the chances of patient recovery Physicians ought to be cautious in using such a justification for failing to tell the truth. In the first place, health professionals rarely know other people that well, and, in general, they tend to underestimate's people's ability to handle bad news. Secondly, this argument is at best a defense of telling the truth in a process over a period of time, not for withholding information indefinitely or lying. Indeed, Freedman (1993) refers to such a process as "offering truth." Unlike the more typical situation where the physician, having ascertained that the patient is competent, proceeds to tell the patient everything about their diagnosis and prognosis in one sitting, "offering truth" conceives of truth telling to patients as a process over a period of time. As Freedman says: "A patient's knowledge of diagnosis and prognosis is not all-or-nothing. It exists on a continuum, anchored at one end by the purely theoretical `absolute ignorance' and at the other by the unattainable `total enlightenment.' Actual patients are to be found along this continuum that vary in response to external factors (verbal information, non-verbal clues, etc.) as well as internal dynamics such as denial" (Freedman, 1993). The best way to begin this process, Freedman suggests, is to find out where the patient is on the continuum by hearing "from the patient himself or herself, so that [the health care team] can confirm what he or she knows or clear up any misunderstanding that may have arisen" (Freedman, 1993). Beginning here, the health care team will not only know what the patient him/herself knows (which is often much more than the family thinks the patient knows), they can also get a sense of how much the patient wants to know. Surely knowing this can only help in promoting the patient's good.

Kindness Thomasma (1994) looks upon kindness as a kind of virtue, and virtue ethics are typically thought to differ in an important respect from deontology and utilitarianism discussed above under the rubric of respect for persons and utility. In virtue ethics, the

focus of concern is on the character of the persons involved and of the relationship between various actors (as opposed to deontology and utilitarianism where the focus is on specific judgments such as does this action produce the most utility or would this action promote respect for persons). In virtue ethics, then, individuals want to consider truth telling within the context of relationships. Clearly, it would be hard to develop a real friendship, for example, with a person who often deceived the other person in one way or another. Similarly, it would hard to imagine a trusting relationship being built and sustained between a health care worker and a patient if the patient could not trust the health care worker to tell him or her the truth. Of course, this leaves open the possibility that the relationship would not be negatively affected if the truth never came to light. Thomasma offers an example of such a case where a dying victim of a car crash is told in his last minutes of life that his family in the car with him are being taken care of even though all of them are in fact dead (Thomasma, 1994). Typically, however, the truth of the situation will come out, even in the typical palliative care case. The prominent and extenuating features of the car crash case are that the person came to the hospital as an emergency patient and will die almost immediately. These are important because (1) emergency cases do not typically demand the sort of informed consent that competent patients do and (2) the patient will die before any further measures have to be taken that would demand his consent. Even palliative patients typically do not present in this fashion. That is, it would be quite uncommon that an undiagnosed cancer patient arrived at the hospital minutes or hours before his/her death from his/her cancer. There is almost always a much longer period of time between diagnosis and death. Even if that period is only a few days or weeks, there will be decisions that have to be made—such as the one whether to transfer the patient into a palliative care unit. In such cases, it is very unlikely that the patient will not somehow, perhaps unintentionally, discover their diagnosis and prognosis. In such cases as these, health professionals are presented with a worst case scenario. Whatever benefit was to come of concealment is not met, and they have hence disrespected the patient for no compensating benefit. Moreover, the health professionals have been unkind and jeopardized the possibility of a fruitful and wholesome relationship between health care team members and the patient. Moreover, a further point can be made. Asking health care professionals to withhold information or lie to patients puts them in a moral dilemma and can lead to moral distress and residue. Over time, this can even threaten the very moral identity of health

Stewart: Telling Patients the Truth

care workers and cause them to leave their profession or to stop caring for patients in the way they had previously (Kelly, 1998; Rubin & Baylis, 2000). Conclusion

This article has argued that the default position for physicians and other health care workers is to tell their competent patients the truth. The basis for this are: respect for persons, utility, and kindness. Telling competent patients the truth need not be done, however, "all at once." Indeed, in many situations, it is better to think of telling the truth to patients as a process, which typically

begins by getting the competent patient to tell his/her own account of what he/she understand about his/her condition and how much he/she wants to be told. Given that patients have a right rather than a duty to be told the truth, patients can decide for themselves that they do not want to be told the truth and that their proxies should make decisions for them. This may be especially true for those cultures that do not value truth telling and autonomy as much as is currently the case for the majority in North America (Buken, 2003; Asai, 1995). Cases where the truth will be withheld for long periods of time will be extraordinarily rare, and the burden of proof will fall squarely on those who want to withhold the truth.

References Asai, A. (1995). "Should physicians tell patients the truth." The Western Journal of Medicine 163, 36-39 Buken, 0. (2003). 'Truth-telling information and communication with cancer patients in Turkey."Journal of the International Society for the History of Islamic Medicine 3: 31-36 Beauchamp, T & Childress, J. (2001). Principles of biomedical ethics. 5th ed. Oxford: Oxford University Press. Bentham, J. (1789/2007). Introduction to the principles of morals and legislation. Dover. Braddock, C.H. (1998). "Truth-telling and withholding information." Ethics in Medicine. http://depts.washington.edu/ bioethx/topics/truth.html. Cullen, S & Klein, M. (2000). "Respect for patients, physicians, and the truth." In R. Munson, ed. (2004). Intervention and reflection: Basic issues in medical ethics, 7th ed. (Belmont, CA: Wadsworth), 155-163. Ells, C. (2006). "Healthcare ethics committees: Contribution to review of institutional policy" HEC Forum 18 265-275. Emanuel, EJ & Emanuel, L. (1992). "Four models of the physician-patient relationship." Journal of the American Medical Association 267, 16: 2212-2226. Freedman, B. (1993). "Offering truth." Archives of Internal Medicine 153, 3: 572-576.

Medical Secrecy: Patients' Right of Access to Medical Records Arthur Schafer

In Moscow, recently, a woman dying of a brain tumour was thrown out of a Soviet hospital. Her offence? She looked, without authorization, at her medical record.

Kelly, B. (1998). "Moral identity: A follow-up study of English graduate nurses. Journal of Advanced Nursing 24: 1063-1069. Kenny, N. (2004). "What's fair? Ethical decision-making in an aging society" Canadian Policy Research Networks Research

Report. Lipkin, M. (1979, 4 June). "On telling patients the truth." Newsweek 13. In R Munson, ed. (2004). Intervention and reflection: Basic issues in medical ethics, 7th ed. (Belmont, CA: Wadsworth), 154-156. McDonald, M. (2001). A framework for ethical decisionmaking: version 6.0-Ethics Shareware January www .ethics.ubc.ca/people/mcdonald/decisions.htm. Mill, JS. (1863/2007). Utilitarianism. Dover. Randall, F & Downie, RS. (2006). The philosophy of palliative care. Oxford: Oxford University Press. Thomasma, D. (1994). Telling the truth to patients: A clinical ethics exploration." Cambridge Quarterly of Healthcare Ethics 3: 375-382. In T. Beauchamp & L Walters, eds. (2003). Contemporary issues in bioethics, 6th ed, 128-132. Thomson/Wadsworth. Webster, GC & Baylis, E (2000). "Moral residue." In Mar gin of Error: The ethics of mistakes in the practice of medicine (Hagerstown, MD: University Publishing Group), 172-175.

Subsequently, the arrogance of the physician who took this action was much criticized in the Soviet press. At almost the same time, in Toronto, a young woman was forced to go to court to obtain access to her father's medical records. The family wished to determine the quality of the medical care he received before he died. Both of these cases raise interesting and important questions: why are physicians, in general, so reluctant

Source: From Humane Medicine 1, 2 (October 1985), 91-9. Reprinted with permission of the author.

LE

114

I I ILII

111111

3 Management of Medical Information

to allow the individual's access to their own records? And, why are patients increasingly insistent upon their right of access?

nursing, respiratory therapy, and clinical pharmacy; four unit secretaries; four hospital financial officers; and four chart reviewers. It seems obvious that the risk of serious errors and omissions is significant when the information in a paThe Dignity of the Patient tient's medical file is generated by such a large number Many people will have had an experience somewhat of persons. Partly, this is because physicians and others like the following You are in the doctor's office. He has vary in the level of skill and care they exercise in rebeen taking notes as you describe your problems and cording the information. The 1980 Krever Commission symptoms. Then he is called from the room. The med- on confidentiality of health records cited many cases in ical file is left, open, on the desk. which incorrect information was entered into patient Who has not felt tempted, in such a situation, to files, with disastrous consequences. If patients have take a surreptitious look at what the doctor has written no right of access to the information in their files, they in that file? Whether you succumb to this temptation or may be denied the opportunity to correct errors and resist, you may feel belittled by the lack of open access omissions. to your record. As one patient put it, "It is not that Moreover, in a variety of circumstances, people everyone would want to know, but it is the assumption will be asked to waive the confidentiality of their health that you are not one of the team that is so undignified." records. When you apply for life insurance, or welfare, It is important to stress that the issue is not simply or worker's compensation, or credit, or employment, one of indulging idle curiosity. The issue, rather, con- you may be asked to release such information. How cerns the basic nature of the doctor—patient relation- can patients make intelligent, informed decisions as to ship. When patients' rights advocates insist upon the whether to give such permission for release of informaneed for legislation to guarantee a patient's right of tion unless they have access to their files and an opporaccess to information in the medical record, they are tunity to ensure that the information contained therein rejecting, implicitly or explicitly, the physician's trad- is accurate and non-prejudicial? itional paternalistic role as benevolent sovereign. Given the large number of persons who have authorized access to the patient's personal medical record, and given the number of persons who may seek and Correcl:ing Mistakes and Protecting gain illegitimate access to the computers in which Confidentiality such information is stored (not always very securely), Most patients are aware only vaguely that many people it seems ironic that the one person who lacks access have access to their health records; and even fewer are is often the patient. You may not agree entirely with aware that their record may contain serious errors and Siegler when he declares that "medical confidentiality, omissions. as traditionally understood by patients and doctors, no Prompted by a patient's complaint, Siegler, an• longer exists,' but doubtless, there is some cause for American physician, took the trouble to enumerate the concern. number of persons who had legitimate access to the patient's hospital record.' He was surprised to find that The Doctors' Dilemma "at least 25 and possibly as many as 100 health professionals and administrative personnel at (his) university The case in favor of patient access to personal medical hospital had access to the patient's record and that all records is so strong that one may wonder that so many of them had a legitimate need . . . to open and use physicians continue to resist. Some of this resistance that chart." The patient who complained had a med- may be attributed to the physician's desire to retain ical condition—mild chronic obstructive pulmonary power over the patients. Monopoly of knowledge condisease—that was relatively straightforward and rou- fers power. Those physicians who enjoy the status of tine. Despite this, those with access to his file included "priest" and expect their patients to accept, without six attending physicians—the primary physician, the question, decisions made on their behalf, naturally will surgeon, the pulmonary consultant, and others; 12 hour wish to retain exclusive control of access to medical officers--medical, surgical, intensive-care unit, and records. This monopoly provides the ancillary "benefit" "covering" house staff; 20 nurses (on three shifts); six of making it difficult (if not impossible) for patients to respiratory therapists; three nutritionists; two clin- use the legal system against incompetent or unethical ical pharmacologists; 15 students—from medicine, physicians.

Schafer: Medical Secrecy: Patients' Right of Access to Medical Records

It would be a mistake, however, to view all opposition to patient access to medical records as a conspiracy of professionals against the public. As the American Medical Association has declared: "It is our position that the right of a patient to medical information from his physician is based upon the fiduciary relationship which imposes a duty to act in the best interest of the patient." It is frequently claimed that physicians have both a right and a duty to withhold information from patients when they judge this to be in the patients' best interests. Complete disclosure is opposed on the grounds that it is likely to alarm and upset patients needlessly. (Imagine the anxiety of the patient who reads his physician's tentative speculations, e.g., "query, possibility of cancer.") According to this view of the doctor—patient relationship, patients should be willing to trust the good will and sound judgement of their physicians. In an editorial published by the Journal of the Royal Society of Medicine,' J.M.A. Northover, a British consultant surgeon, argues that patient trust in doctors would be undermined by giving patients access to their medical record: "Part of that trust is based on the understanding between the parties that the doctor regards helping the patient to come to terms with the disease and its implications as an important part of his or her job; many doctors feel that this responsibility requires the careful 'rationing' of information, the best interests of the patient determining just how that information is divulged." Northover concedes that patients frequently are underinformed, but he insists that doctors should "resist the endeavours of those who advocate patient access to medical files," principally because "information management is an important aspect of proper medical care." Northover buttresses his argument against giving patients the right of access to their medical files with the argument that these records usually are written in a language that for technical and other reasons would be of little practical use to patients. "Every doctor," he declares, ". . . would be horrified by the suggestion that the best way to inform a patient is simply to hand over a fairly incomprehensible, untidy and potentially frightening document." Patients are likely to be misled, and alarmed unnecessarily by the "dispassionately pessimistic" style preferred by many physicians in their notetaking, and by the sinister diagnostic possibilities included as part of a conscientious differential diagnosis. Northover confers the label "filophile" on those who favor giving patients a legal right of access to their files and accuses them of favoring "do-it-yourself' communication. Another potentially harmful consequence of giving patients the right to see their files would be that many physicians might change deleteriously their

recordkeeping practices; for example, they might attempt to remember potentially frightening diagnoses rather than entering them in the record. Or they might become reluctant to enter into the patient's record such unflattering comments as "working class background-?? Poor self-image." The physician would omit such comments in order to avoid mutual embarrassment and to preserve the patient's confidence, but such omissions might lead to less efficient clinical management. Resolving the Dilemma Fortunately, there are ways of resolving or "dissolving" most of these objections. If medical records were written on the assumption that patients might see them, the physician would be obliged to replace technical jargon by good clear English, without prejudice to medical efficiency. It is conceded that, in the absence of supplementary explanation, the patient would find the information in the file often more confusing than edifying. However, the conclusion to be drawn from this is not that patients should be denied access to their files, but that the physician should be present when the patient receives the file, to provide necessary explanation and, where necessary, reassurance. Patients are entitled to know the truth about their medical status, and to have an opportunity to correct errors and omissions. It seems reasonable to assume that those who wish to protect themselves from such information generally will not request access. The process of explaining and interpreting medical records is likely to require some additional investment of the physician's time, but this would not be time wasted. Surely patients are entitled to receive both information and emotional reassurance. The experience of having the physician explain the significance of the information in his history is likely to enhance the patient's trust and thereby contribute to good therapy. A trust based on shared knowledge is more valuable (because better founded) than one based on blind faith. As for the mental "distress" or "alarm" some physicians fear will result from free patient access, we need to see some evidence of this. Such fears may be exaggerated or misplaced. A randomized clinical trial to assess the comparative therapeutic effects of access and secrecy might show that patients experience greater fear and anxiety when they are denied access to their files than when they gain access in propitious circumstances. Ignorance has a way of generating unwarranted anxieties and fears that can be more corrosive of patient morale and trust than would be produced by genuine knowledge.3

I I al

111111

3 Management of Medical Information

As For the danger that the patients morale may be undermined when he discovers in the record comments which are unflattering, we can respond that such comments have no place in the record in any event, because they have no medical justification. Thus, instead of writing "hypochondriac," the doctor would show a more professional attitude by recording that the patient seems unduly worried. This problem then could be discussed openly. [f the physician cannot persuade his patient that this assessment is appropriate, the patient may choose to seek other advice. This outcome is consistent with a patient's right to autonomy and should not necessarily be viewed as an unfortunate consequence. Northover concedes that "it is a sad reflection on the medical profession that many members of the public are sufficiently dissatisfied by their doctor's inclination and ability to talk to them that they feel a need to seek access to their medical records to find out what is happening." But his opposition to legislation that would open medical files at the patient's request rests on his belief that Information management is an important aspect of proper medical care" and his belief that the Hippocratic principle of non-maleficence (not doing harm) requires that physicians carefully ration the information they give their patients. Dr. Northover illustrates his position with the following example: "In a state of frightened desperation, a patient might ask 'Have I got cancer?', whilst wanting to hear only one answer: 'No.' The good doctor will recognize the situation and will choose his or her words carefully and compassionately." The crucial question (ignored, however, by Northover) is this: "How can the doctor be sure that the patient does not want to know the truth?" Perhaps the doctor has misinterpreted the patient's real wishes; perhaps the doctor has projected his/her anxieties about cancer onto the patient. Failure to inform, however well motivated, will often produce more harm than good. Whether a patient suspects deception or is successfully kept in a state of complete ignorance, the lack of knowledge will often produce more harm than would have been produced by the "painful" truth. I believe the risks associated with medical paternalism are too great to be brushed aside. This is not to deny that physicians ought to communicate painful information to patients with tact and sensitivity. The communication skills Northover calls for (pp. 93 and 95) will be no less important once patients have won the legal right to access their files. This position does not deny that exceptional cases may require the withholding of certain specific information. But those of us who are not physicians will find it difficult to accept that many patients who genuinely wish to remain in ignorance of their diagnosis or prognosis will insist upon seeing their record.

In the rare cases where this does occur and a patient is harmed by the policy of openness, one may regret the harm and yet still insist that patients have a right to know what is in their records and, in any event, that more harm is caused by the present policy of secrecy. Records of Psychiatric Patients

The case for complete openness and disclosure is least strong with respect to some kinds of psychiatric patient. In Sweden, where patients are entitled to ask for their medical records, doctors have the right to refuse access if they believe that such access might endanger a patient's life. This seems to be a reasonable compromise, although some patients' rights advocates fear that physicians, including psychiatrists and other mental health professionals, often misuse their discretionary power over mental patients. Sociologists have frequently noted that once a patient has been labelled as "mentally ill," mental health professionals tend to interpret all the patient's subsequent behaviour as evidence confirming the original diagnosis.4 Thus, innocent behaviour may be interpreted as pathological because of the context in which it occurs. Once a patient has been so labeled, and the label is recorded in the patient's file, it exerts a powerful influence upon all future decisions made with respect to that patient. If the patient has been mislabeled, much harm may result. However, if mental patients (and/or their family or a court-appointed guardian) had a right of access to their records, psychiatrists might be encouraged to develop a more careful and scrupulous attitude towards recordkeeping. Cohen has offered some evidence that, for many psychiatric patients, access to information about themselves can be therapeutically beneficial.5 Of course, there are some instances in which the risk of serious harm to the patient from access is so great that access ought to be denied. It seems to me that such cases are likely to be uncommon. Hence, a sensible policy would be to legislate a right to access but at the same time to incorporate in the law a set of criteria for non-disclosure. This would provide a safety buffer for such exceptiona cases. Canadian Law

At present, only a few Canadian provinces have enacted legislation which confers upon patients the legal right of access to the information contained in their medical files. In Alberta, for example, the medical record in the doctor's office belongs to the physician, and in the hospital it belongs to the hospital The patient is entitled, upon request,

Schafer: Medical Secrecy: Patients' Right of Access to Medical Records

to have access to the information contained in the record but is not entitled to physical possession of the records. In Quebec, medical records are regarded asii6perty to be shared between physician and patients. Hospital records are shared between the institution, the physician, and the patient. In neither case is the patient entitled to take the original record and amend it. Patients may be denied access to their records only when such access is deemed to be prejudicial to the patients' health. Section 4.02 of the regulations under the Quebec Medical Act reads: "Except when it is prejudicial to the patient's health, the physician must respect the patientb right to information in the records which concern him and to obtain a copy of such records." With respect to hospital records, the Quebec legislation reads: "A recipient. to whom an establishment refuses access to his record or ref uses to give written or verbal communication of it may, on summary motion, apply to a judge of the Superior Court, Provincial Court, Court of the Sessions or Youth Court or to the Commission, to obtain access to or communication of it as the case may be. The judge shall order such establishment to give such recipient access to his record, or communication of it, as the case may be, unless he is of the opinion that it would be seriously prejudicial to the health of such recipient to examine his record."

Conclusion

Evidence from jurisdictions in which patients have gained the legal right of access to the information in their medical files suggests that only a minority will choose to exercise their rights. Does this mean that patients regard the right of access as of little importance? The minority who seek such access seem to see the right as highly important. It is probable that the consumer's rights philosophy which has increased in strength during the last few decades will continue to spread and to influence "consumers" of health care, that is, patients. Whether or not the number of patients demanding access to their medical files continues to grow, it should be emphasized that the right of access is important even to those who do not wish to see their files. That is, it is important that patients know that they have the right to see what is in their files, even if they do not choose to exercise this right. The attitudes and values of both physicians and patients have changed significantly over the past decade. Patients now expect to take more responsibility for their own health. Physicians are significantly less paternalistic than they used to be. This change in attitude has been beneficial for both physicians and patients. It seems likely that when patients across Canada win the legal right to see their medical files, this change will enhance the values of patient autonomy and patient responsibility.

References 1. 2. 3.

Siegler M. NE JM 1982;307: 1518-1521. Ibid. 1985. pp. 93-95. See, e.g., Aphidi R. "Informed consent: A study of patient reaction." JAMA 1971;216: 1325-1329; Mcintosh J. "Patient's awareness and desire for information about

4. 5.

diagnosed but undisclosed malignant disease." The Lancet 1976;1: 300-303. Garman E. Asylums. Penguin Books, Harmondsworth, 1968. Cohen RN. Whose file is it, anyway? National Council for Civil Liberties, 1983.

3.3 Confidentiality of Information

A Defense of Unqualified Medical Confidentiality Kenneth Kipnis The Case of the Infected Spouse

The following fictionalized case is based on an actual incident.

1982: After moving to Honolulu, Wilma and Andrew Long visit your office and ask you to be their family physician. They have been your patients ever since. 1988: Six years later the two decide to separate. Wilma leaves for the Mainland, occasionally sending you a postcard. Though you do not see her professionally, you still think of yourself as her doctor.

Source: From American Journal of Bioethics 6 (2006) (2), 7-18. Reprinted with permission of Taylor & Francis Journals.

II

111111

3 Management of Medical Information

1990: Andrew comes in and says that he has embarked upon a more sophisticated social life. He has been hearing about some new sexually transmitted diseases and wants to be tested. Testing reveals that he is positive for the AIDS virus, and he receives appropriate counseling. 1991: Visiting your office for a checkup, Andrew tells you Wilma is returning to Hawaii for reconciliation with him. She arrives that afternoon and will be staying at the Moana Hotel. Despite your best efforts to persuade him, Andrew leaves without giving you assurance that he will tell Wilma about his infection or protect her against becoming infected.

Do you take steps to see that Wilma is warned? If you decide to warn Wilma, what do you say to Andrew when, two days later; he shows up at your office asking how you could reveal his confidential test results? If you decide not to warn Wilma, what do you say to her when, two years later in 1993, she shows up at your office asking how you, her doctor; could possibly stand idly by as her husband infected her with a deadly virus. She now knows she is positive for the virus, that she was infected by her husband, and that you—her doctor—knew, before they reconciled, that her husband would probably infect her.

It is not enough that a warning would discharge a duty to Wilma, merely so she could make an informed choice. Plainly, the paramount concern has to be to save Wilma's life. Finally, Wilma is not a mere stranger. Instead she has an important relationship with you—her doctor—that serves as a basis for special obligations: You have a special duty to look out for her health. In the light of these five considerations, it should not be a surprise that the conventional wisdom in medical ethics overwhelmingly supports either an ethical obligation to breach confidentiality in cases like this one or, occasionally and less stringently, the ethical permissibility of doing so (Lo 1995). Notwithstanding this consensus, it is my intention to challenge the received view. I will argue in what follows that confidentiality in clinical medicine is far closer to an absolute obligation than it has generally been taken to be; doctors should honor confidentiality even in cases like this Although the focus here is on the Case of the Infected Spouse, the background idea is that, if it can be demonstrated that confidentiality should be scrupulously honored in this one case where so many considerations support breaching it, the duty of confidentiality should be taken as unqualified in virtually all other cases as well (Kottow 1986). I shall not, however, defend that broader conclusion here. Although this essay specifically addresses the obligations of doctors, its approach applies more broadly to all professions that take seriously the responsibility to provide distressed practitioners with authoritative guidance (Kipnis 1986, 63-79; Wicclair 1986). With its focus narrowly on "professional obligations," the methodology used below also represents something of a challenge to much of the conventional thinking in medical ethics.

The ethical challenges here emerge from an apparent head-on collision between medical confidentiality and the duty to protect imperiled third parties. Notwithstanding Andrew's expectation of privacy and the professional duty to remain silent, it can seem unforgivable for anyone to withhold vital assistance in such a crisis, let alone a doctor. The case for breaching confidentiality is supported by at least five considerations: First, the doctor knows, to a medical certainty, that Andrew is both infected with HIV and infectious. Clearing the Ground: What Professional Second, knowing Wilma as a patient, let us suppose Obligations Are Not the doctor reasonably believes that she is not infected. (Wilma cannot be at risk of contracting the disease if Among philosophers, it is commonplace that if people she is infected already.) Third, Wilma's vulnerability is are not asking the same questions, they are unlikely both serious and real. HIV infection is both debilitating to arrive at the same answers. It may be that the main and, during those years, invariably fatal: The couple's reason doctors have difficulty reaching consensus in sexuality makes eventual infection highly likely. Fourth, ethics is that, in general, systematic discussion about assuming that preventing Wilma's death is the goal, it professional responsibility is commonly confused with is probable that, were Wilma to be told of Andrew's at least three other types of conversation. When one infection, she would avoid exposing herself to the risk. asks whether one should call the hotel to warn Wilma, This is not a trivial condition: many people knowingly one can be asking: 1) what the law requires 2) what risk illness and injury out of love and other honorable one's personal morality requires (e.g., as an Orthodox motivations. Molokai's Father Damien contracted and Jew, a Roman Catholic, etc.); or 3) what is required by died from Hansen's disease while caring for patients one's most deeply held personal values (e.g., preventing he knew might infect him. Soldiers, police, and fire- deaths or scrupulously honoring other obligations). fighters commonly expose themselves to grave risk. Discussions can meander mindlessly over all three

Kipnis: A Defense of Unqualified Medical Confidentiality

areas without attending to boundary crossings. More to the point, effective deliberation about professional obligations, as I will try to show, differs importantly from all three of these discussions. Accordingly, it is necessary to identify and bracket these other perspectives in order to mark off the intellectual space within which practitioners can productively reflect on questions of professional responsibility. Let us examine these different conversations. Law The conventional wisdom on the ethics of medical confidentiality has been largely shaped by a single legal case: Tarasoff v. Regents of the University of California (Supreme Court of California; 529 p. 2d 553, Cal. 1974). In 1969, Prosenjit Poddar, a student at U.C. Berkeley told a university psychologist he intended to kill a Ms. Tatiana Tarasoff, a young woman who had spurned his affections. The psychologist dutifully reported him to the campus police, who held him briefly and then set him free. Shortly afterwards, Poddar did as he said he would, stabbing the young woman to death. The Tarasoff family sued the University of California for their daughter's death, finally prevailing in their contention that the psychologist (and, by implication, the University) had failed in their duty to protect, since neither Tatiana nor those able to apprise her of danger were warned. The University was found liable and had to compensate the family for its loss. Today it is hard to find discussions of the ethics of confidentiality that do not appeal to this legal parable and, occasionally, to its California Supreme Court moral: "The protective privilege ends where the public peril begins." Taking its cue from Tarasoff, the prevailing standard in medical ethics now holds that the obligation of confidentiality will give way when a doctor is aware that a patient will seriously injure some identified other person. (One might ask why disclosure is not required when a patient will seriously injure many unidentified persons. Under the narrower standard, there is no duty to alert others about an wv-infected prostitute who neither informs nor protects a large number of anonymous at-risk clients.) We assume that the physician knows Andrew is seropositive, that Wilma is likely seronegative, that the two will likely engage in activities that transmit the virus, and that breaching confidentiality will probably result in those activities not occurring and Wilma's not becoming infected. Thus, a physician's warning in the Case of the Infected Spouse will mean that Wilma is very likely to remain infection-free, and a failure to warn her is very likely to result in her eventual death from AIDS.

Focusing on the legal standard, it is useful to distinguish between "special" and "general" legal duties. Special duties can apply to individuals occupying certain roles. A parent, but not a bystander, has a special duty to rescue a drowning daughter, firefighters and police officers have special duties to take certain occupational risks, and doctors have many special duties toward their patients: confidentiality is a good example. In contrast, virtually everyone has a general duty to be scrupulously careful when handling explosives, to pay taxes on income, to respect others' property, and so on. It is notable that the duty to warn in Tarasoff is a special duty, applicable only to those occupying roles. So if my neighbor casually assures me he is going to kill his girlfriend tomorrow, the Tarasoff ruling does not require me to warn her: It is surprising to many that the default standard in Anglo-American jurisprudence is that there is no general duty to improve the prospects of the precar- , iously placed, no legal obligation to undertake even an easy rescue. As first-year law students discover, one can stand on a pier with a lifeline in hand and, with complete impunity, allow a stranger to drown nearby. Although we will pass over it, it is notable that, in general, the parties who are legally obligated to warn are those who are otherwise ethically obligated not to disclose. One should reflect on the absence of a general duty to warn. The easy transition from law to ethics reflects a common error. The mistake is to move from the premise that some action is legally required (what the Tarasoff opinion establishes in the jurisdictions that have followed it) to the conclusion that the same action is ethically required. But ethical obligations can conflict with legal ones. Journalists, for example, are sometimes ordered by the courts to reveal the identities of their confidential sources. Although law demands disclosure, professional ethics requires silence. Reporters famously go to jail rather than betray sources. Journalists can find themselves in a quandary: while good citizens obey the law and good professionals honor their professional codes, laws requiring journalists to violate their duties to confidential sources force a tragic choice between acting illegally and acting unethically. Conscientious persons should not have to face such decisions. Similarly in pediatrics, statutes may require doctors to report suspicions of child abuse. But where protective agencies are inept and overworked and foster care is dangerous or unavailable, a doctor's report is more likely to result in termination of therapy and further injury to the child instead of protection and care. To obey the law under these appalling, but too common,

II

al

III Lid

3 Management of Medical Information

circumstances is most likely to abandon and even cause harm to the minor patient, both of which are ethically prohibited in medicine. i To assume that legal obligations al ways trump or settle ethical,, ones is to blind oneself to the possibility of conflict. Professions have to face these dilemmas head-on instead of masking them with language that conflates legal standards and ethical ones. They must conceive professional ethics as separate from the law's mandate. When law requires what professional responsibility prohibits (or prohibits what professional responsibility requires), professional organizations must press the public, legislatures, and the courts to cease demanding that conscientious practitioners dishonor the duties of their craft. This is an important responsibility of professional organizations. It is a mistake to configure professional , obligations merely to mirror the law's requirements. Rather, the law's requirements must be configured so that they do not cpnflict with well-considered professional obligations! Law is a human artifact that can be crafted well or badly In a well-ordered society no one will have to choose between illegality and immorality. Since the law can require conduct that violates ethical standards (and ethical standards can require conduct that violates the law), it cannot be the case that legal obligations automatically create ethical obligations. As the tradition of civil disobedience shows, it can be ethically permissible or obligatory (though not legal) to violate an unjust law. Even though laws cannot create ethical obligations by fiat, professions need to distinguish between the state's reasonable interests in the work of doctors (e.g., preventing serious harm to children) and the specific legal mandates a state imposes (e.g., requiring doctors to report suspicion of child abuse to an incompetent state agency). Just as patients can make ill-considered demands that should not be satisfied, so too can the state and its courts. Accordingly, it is assumed that the state has a legitimate interest in preventing harm to people and that doctors have an ethical obligation to further that important public objective. The focus in this essay is on itfie shape of the resulting ethical obligation as it applies I narrowly to cases like those involving Wilma Long and Tatiana Tarasoff. Because they introduce complexities that will carry us far afield, we set aside cases involving: (a) children brought in by parents (Kipnis 2004); (b) patients referred for independent medical evaluation; (c) mentally ill or retarded patients in the custody of health care institutions; (d) health care that is the subject of litigation; (e) gunshot, knife wounds, and the like; (f) workers' compensation cases; and a few others. While a much longer discussion could cover these areas, many readers can extend the analysis

offered here to discern much of what I would want to say about those other cases. Though I will not discuss them, institutional policies (hospital rules, for example) function very much like laws. Both involve standards that can be imposed externally upon practitioners. Both can be formulated knowledgeably and wisely or with a disregard for essential professional responsibilities. Personal Morality

We will understand a "morality" as a set of beliefs about obligations. There are plainly many such sets of beliefs: the morality of Confucius has little in common with the moralities of George W. Bush and Thomas Aquinas. For most of us, morality is uncritically absorbed in childhood, coming to consciousness when we encounter others whose moral beliefs differ. There are still parts of the world in which virtually all members of a community are participants in a common morality. But moral pluralism now seems a permanent part of the social order. Consider a Jehovah's Witness physician who is opposed, on religious grounds, to administering blood transfusions. If this doctor were the only physician on duty when his patient needed an immediate transfusion, a choice would have to be made between being a good Jehovah's Witness and being a good doctor. The doctor's personal moral convictions are here inconsistent with professional obligations. It follows that clarity about personal morality is not the same as clarity about medical ethics. Professionalism can require that one set aside one's personal morality or carefully limit one's exposure to certain professional responsibilities. Here the rule has to be that doctors will not take on responsibilities that might conflict with their personal morality. Problems could be sidestepped if the Jehovah's Witness doctor specialized in a field that didn't involve transfusion (e.g., dermatology) or always worked with colleagues who could administer them. If I am morally against the death penalty, I shouldn't take on work as an executioner. If I am deeply opposed to the morning-after pill, I shouldn't counsel patients at a rape treatment center. To teach medical ethics in a pluralistic professional community is to try to create an intellectual space within which persons from varied backgrounds can agree upon responsible standards for professional conduct. Participants in such a conversation may have to leave personal morality at the door. For some, it may be a mistake to choose a career in medicine. If ethics is a critical reflection on our moralities, then the hope implicit in the field of medical ethics is that we might some day reach a responsible consensus

Kipnis: A Defense of Unqualified Medical Confidentiality

on doctors' obligations. While medicine has dozens of codes, it is not hard to observe commonalities: the standards for informed consent, for example. At a deeper level, there can also be consensus on the justifications for those standards. One role for the philosopher is, as in this essay, to assess carefully the soundness of those arguments. A major task for professions is to move beyond the various personal moralities embraced by practitioners and to reach a responsible consensus on common professional standards. Personal Values

Values are commonly a part of an explanation of personal conduct. It is always reasonable to ask of any rational action: what good was it intended to promote? While some wear shoes to avoid hurting their feet (embracing the value of comfort), others think they look better in shoes (embracing aesthetic values). Where we have to make personal decisions, often we consider how each option can further or frustrate our values and try to decide among the good and bad consequences. This strategy can serve when the question is "What should I do?" But the question "What should a good doctor do?" calls for a different type of inquiry. For while I have many personal values, the "good doctor" is an abstraction. She is neither Protestant nor Buddhist, doesn't prefer chocolate to vanilla, and doesn't care about money more than leisure time. questions about professional ethics cannot be answered in terms of personal values. A second difficulty appears when we consider that one can give perfect expression to one's most deeply held personal values and still act unethically. Hannibal Lecter in Silence of the Lambs and Mozart's Don Giovanni are despicable villains who give vigorous effect to deeply held if contemptible personal values. While personal values can determine action, they do not guarantee that the favored actions are ethical. Accordingly, we cannot appeal to our personal values to inquire about what physicians in general ought to do. Medicine has no personal values, only individual physicians do. When a physician must decide whether or not to resuscitate a patient, personal values should have nothing to do with the issue. Whether you like the patient or detest him, whether you are an athe: ist or a fundamentalist believer in a joyous hereafter, should not weigh in the balance. A key part of professionalism involves being able to set personal values aside. While medical students have much to gain by becoming clear about their personal values, that clarity is not the same as responsible certainty about professional obligations.

To summarize the argument so far, discussion about professional obligations in medicine is not the same as discussion about legal and institutional obligations, personal morality, or personal values. If a responsible ethical consensus is to be achieved by a profession, it is necessary for physicians to learn to bracket their personal moral and value commitments and to set aside, at least temporarily, their consideration of legal or institutional rules and policies. The practical task is to create an intellectual space within which responsible consensus can be achieved on how physicians, as professionals, ought to act. I will now describe one way in which this might be done. The Concept of a Professional Obligation

Professional ethics involves disciplined discussion about the obligations of professionals. One place to begin is with a distinction between personal values, already discussed, and what can be called "core professional values." A physician can prefer (1) pistachios to Brazil nuts and (2) confidentiality to universal canaoi.) While the preference for pistachios is merely personal; the preference for confidentiality is a value all doctors ought tii_possess. The distinction between personal values and "core professional values" is critical here. There is what this flesh-and-blood doctor happens to care about personally and what the good doctor ought to care about. This idea of a "good doctor" is a social construction, an aspect of a determinate social role, an integral element of medical professionalism. Our idea of a good doctor includes a certain technical/intellectual mastery coupled with a certain commitment to specific professional values. As with the Jehovah's Witness doctor, personal and professional values may be in conflict. As part of an appreciation of the ethical claims of professionalism, physicians must be prepared to set aside their personal values and morality, to set aside what the legal system and their employers want them to care about, and to take up instead the question of what the responsible physician ought to care about. The profession's core values inform those purposes that each medical professional should have in common with colleagues. In discussing the professionally favored resolution of ethically problematic cases (the Case of the Infected Spouse, for example), physicians can ask—together—how medicine's core professional values ought to be respected in those circumstances. We have alluded to some of these core professional values. Trustworthiness needs to be on the list. Beneficence toward the patient's health needs is essential. Respect for patient autonomy is a third. Others might be collegiality (duties to colleagues) and perhaps a few

11 I

II

3

111111

3 Management of Medical Information

others: nondiscrimination and a certain deference to families are among the most commonly mentioned candidates. If we were to leave out that doctors should care about the well-being of the public, the argument for confidentiality would be easy. But it too properly goes on the list. Anyone seeing no point in furthering and securing these values would be ill-suited for the practice of medicine. Each of these professional values has two dimensions. Along one vector, they define the shared aspiration of a profession. At any time, medicine's ability to benefit patients will be limited. But it is a part of the profession's commitment to push its envelope, to enlarge its collective competency and draw upon its knowledge and skill. Those who master and extend the profession's broadest capabilities are exemplary contributors, but practitioners do not discredit themselves by failing to serve in this estimable way Along the second vector, values define a bottom line beneath which practitioners shall not sink. Paraphrasing Hippocrates, although you may not always be able to benefit your patients, it is far more important that you take care not to harm them. Knowingly to harm a patient (on balance) is not merely a failure to realize the value of beneficence. It is a culpable betrayal of that value, a far more serious matter. . All the values above can be understood in this second way. Trustworthiness entails that I not lie to patients or deliberately withhold information they have an interest in knowing. Respect for patient autonomy can require that I not use force or fraud upon them. And the concern for the well-being of the public requires that that interest somehow appear prominently upon every practitioner's radar screen, that doctors not stand idly by in the face of perils the profession can help to avert and, as a lower limit, that they not do anything to increase public peril. Consider that the overutilization of antibiotics, resulting in drug-resistant infectious agents, is professional misconduct that increases public peril. Ethical problems can arise, first, when core values appear to be in conflict, as with the Case of the Infected Spouse. At issue are trustworthiness toward Andrew on one side and beneficence toward Wilma and a concern for the well-being of the public on the other. If the conflict is real, what is required is a priority rule. For example, the concept of decisional capacity is part of a priority rule resolving the well-studied conflict between beneficence and autonomy when do physicians have to respect a patient's refusal of life-saving treatment? There is what the patient wants and what the patient needs. But when a patient is decisionally capacitated and informed, his or her refusal trumps the doctor's recommendation.

Second, ethical problems can also arise when it is unclear what some core professional value requires one to do. Though we can all agree that doctors should avoid harming their patients, there is no professional consensus on whether deliberately causing the deaths of certain unfortunate patients—those experiencing irremediable and intense suffering—is always a betrayal of beneficence. Likewise, although doctors may be in a position to prevent harm to third parties, it is not well understood what they must do out of respect for that value. When core values conflict, what is required is a priority rule. When they are unclear, what is required is removal of ambiguity: what philosophers call "disambiguation." These two tasks—prioritizing and disambiguating core professional values—need to be carried out with a high degree of intellectual responsibility. The above list of medicine's core values is not controversial. Propose a toast to them at an assemblage of physicians, and all can likely drink with enthusiasm. What is less clear is why such a consensus should obligate professionals. A criminal organization can celebrate its shared commitment to the oath of silence. But it doesn't follow that those who cooperate with the police are unethical. In addition to organizational "celebratability," three additional elements are required to establish a professional obligation. The first element is that attention to core values has to be a part of professional education. Most medical education is aimed at beneficence. The procedures used in informed consent express a commitment of respect for patient autonomy and trustworthiness. If the profession wholly fails to equip its novices to further its core values, it can be argued that it is not serious about those professed values. Its public commitments will begin to look like they are intended to convey an illusion of concerned attention. In replicating itself, a profession must replicate its commitment. Students of medicine must come to care about the goods that doctors ought to care about. Because justice is rarely explored as a topic in medical education, I do not think it can be counted as a core professional value. However some parts of justice—nondiscrimination, for example—are routinely covered. The second element is critical. The core values are not just goods that doctors care about and that doctors want other doctors to care about. They are also goods that the rest of us want our doctors to care about. I want my doctor to be trustworthy, to be intent on benefiting my health, to take my informed refusals seriously, and so on And we want our doctors to look out for the well-being of the public. The core professional values are also social values. (Consider that it is not reasonable to want our mobsters to respect their oaths of silence.)

Kipnis: A Defense of Unqualified Medical Confidentiality 123

The third element flows from the second: an exclusive social reliance upon the profession as the means by which certain matters are to receive due attention. We mostly respect medical competence. But it is precisely because, as a community, we have also come to accept that doctors are reliably committed to their values (our values) that we have, through state legislatures, granted the medical profession an exclusive monopoly on the delivery of medical services. The unauthorized practice of medicine is a punishable crime. If, like the medical profession, one were to make a public claim that, because of unique skills and dedication, some important social concern ought to be exclusively entrusted to you, and the public believes you and entrusts those important matters to you, incidentally prohibiting all others from encroachment upon what is now your privilege, you would have thereby assumed an ethical obligation to give those important matters due attention. Collectively, the medical profession has done exactly this in securing its monopoly on the delivery of certain types of health care. Accordingly, the profession has a collective obligation to organize itself so that the shared responsibilities it has assumed in the political process of professionalization are properly discharged by its membership. A sound code of ethics consists of a set of standards that, if adhered to broadly by the profession's membership, will result in the profession as a whole discharging its responsibilities. Where physician behavior brings about a public loss of that essential trust, society may have to withdraw the monopolistic privilege and seek a better way of organizing health care. Professionalization is but one way of organizing an essential service. There are others. In summary, the medical profession has ethical obligations toward patients, families, and the community because of its public commitment to secure and further certain critical social values and because of society's exclusive reliance on the profession as its means of delivering certain forms of health care. With the professional privilege comes a reciprocal collective responsibility (Kipnis 1986, 1-14). We can now turn our attention to medicine's responsibility to diminish public perils.

The Duty to Diminish Risks to Third Parties There is an implication for the way in which we must now understand the problem in the Case of the Infected Spouse. The opening question "Do you take steps to warn Wilma?" has to be understood as a question about medical ethics and not about "you." We want to know what the "good doctor" should do under those

circumstances? Each doctor is ethically required to do what a responsible doctor ought to do: in order to properly respect the core values of the profession. To become a doctor without a proper commitment to respect the profession's values is to be unfit for the practice of medicine. So how are trustworthiness and confidentiality to be understood in relationship to medicine's commitment to diminish risks to third parties? In the Case of the Infected Spouse, the ethical question is posed in 1991, after the doctor—family relationship has been in place for a decade. The dilemma arises during and immediately after a single office visit, forcing a choice between calling Wilma either you will have to explain to Andrew, in two days, why you disclosed his infection to his wife or you will have to explain to Wilma, in two years, why you did not disclose his infection to her. Each option has a bad outcome: the betrayal of Andrew's trust or the fatal infection of Wilma. Either way, you will need to account for yourself. Infection seems a far worse consequence for Wilma than betrayal is for Andrew. Much of the literature on confidentiality has been shaped by this fact, and perhaps the standard strategy for resolving the problem calls attention to the magnitude and probability of the bad outcomes associated with each option. While predictions of harm can sometimes be wrong, it can be evident that Tatiana Tarasoff and Wilma Long are at grave risk and, accordingly, it can seem honorable to diminish the danger to vulnerable parties like them. Justice Tobriner appeals to a version of this consequentialist argument in Tarasoff: Weighing the uncertain and conjectural character of the alleged damage done the patient by such a warning against the peril to the victim's life, we conclude that professional inaccuracy in predicting violence [or deadly infection] cannot negate the therapist's duty to protect the threatened victim. Beauchamp and Childress, in their widely read Principles of Biomedical Ethics (2001, 309), urge clinicians to take into account "the probability that a harm will materialize and the magnitude of that harm" in any decision to breach confidentiality. (While they also urge that clinicians take into account the potential impact of disclosure on policies and laws regarding confidentiality, they are not very clear about how this assessment is to be carried out.) In brief, the very bad consequences of not disclosing risk to Wilma—disease and death and the betrayal of her trust—outweigh the not-all-that bad consequence of breached confidentiality to Andrew. Your explanation to Andrew could cover those points.

II III I

ai

II

IA

111111

3 Management of Medical Information

The preferred argument would go something like this: The state's interest in preventing harm is weighty. Medicine has an obligation to protect the well-being of the community. Because the seriousness of threatened grave injury to another outweighs the damage done to a patier t by breaching confidentiality, the obligation of confidentiality must give way to a duty to prevent serious -num to others. Accordingly, despite confidentiality, warning or reporting is obligatory when it will likely avert very bad outcomes in this way. Of course clinicians should try to obtain waivers of confidentiality before disclosure, thereby avoiding the need to breach a duty. But the failure to obtain a waiver does not, on this argument, affect the overriding obligation to report. A Defense of Unqualified Confidentiality As powerful as the above justification is, there are problems with it. Go back to 1990, when Andrew comes in to be tested for sexually transmitted diseases. Suppose he asks: "If I am infected, can I trust you not to disclose this to others?" If, following the arguments set out in the previous paragraphs, we are clear that confidentiality must be breached to protect third parties like Wilma, then the only truthful answer to Andrew's question is "No. You cant trust me." If the profession accepts that its broad promise of confidentiality must sometimes be broken, then any unqualified assurances are fraudulent, and the profession should stop making them. If there are exceptions, clinicians have a duty to be forthcoming about what they are and how they work Patients should know up front when they can trust doctors and when they can't. To withhold this important information is tc betray the value of trustworthiness. Accordingly, the argument for breaching confidentiality has to be modified to support a qualified confidentiality rule, one that carves out an exception from the very beginning, acknowledging an overriding duty to report under defined circumstances. (In contrast, an unqualified confidentiality rule contemplates no exceptions.) Instead of undertaking duties of confidentiality and then violating them, doctors must qualify their expressed obligations so they will be able to honor them. Commentators who have walked through the issues surrounding confidentiality have long understood the ethical r_ecessity of "Miranda warnings" (Bok 1983; Goldman 1980): A clinician would have to say early on, "Certain things that I learn from you may have to be disclosed to . . . under the following circumstances . . .; and the following things might occur to you as a result of my disclosure: . . ." If doctors are ethically obligated to report, they need to say in advance what will be

passed along, when, to whom, and what could happen then. They should never encourage or accept trust only to betray their patients afterwards. To do so is to betray the value of trustworthiness. But now a second problem emerges. If prospective patients must understand in advance that a doctor will report evidence of a threat to others, they will only be willing to disclose such evidence to the doctor if they are willing to accept that those others will come to know. If it is important to them that the evidence not be reported, they will have a weighty reason not to disclose it to those who are obligated to report it. Some have questioned this proposition, arguing that there is no empirical evidence that prospective patients will avoid or delay seeking medical attention or conceal medically relevant information if confidentiality is qualified in this way. Despite widespread reporting practices, waiting rooms have not emptied, and no one really knows if people stop talking openly to their doctors when confidentiality is breached. Three responses are possible regarding this claim. First, there is a serious difficulty doing empirical research in this area. How, for example, do we determine the number of abusive parents who have not brought their injured children to doctors out of a fear that they will get into trouble with the authorities? How many HIV+ patients avoid telling their doctors all about their unsafe sexual practices? How many of us would volunteer unflattering truthful answers to direct questions on these and other shameful matters? It is notoriously difficult to gather reliable data on the embarrassing, criminal, irresponsible things people do and the steps they take to avoid exposure, especially if those are wrongful too. I don't want to suggest that these problems are insurmountable (Reddy et al. 2002), but they are decidedly there and they often make it hard to study the effects of these betrayals. Second, despite the problems, certain types of indirect evidence can occasionally emerge. Here are two anecdotal examples from Honolulu. There was a time, not long ago, when military enlistees who were troubled by their sexual orientation knew that military doctors and psychologists would report these problems to their officers. Many of these troubled soldiers therefore obtained the services of private psychologists and psychiatrists in Honolulu, despite the fact that free services were available in military clinics. The second example emerged from the failure of the Japanese medical system to keep diagnoses of HIv infection confidential. Many Japanese who could afford it traveled to Honolulu for diagnosis and treatment, avoiding clinics in Japan. At the same time, Japanese data on the prevalence of HIv infection were unrealistically

Kipnis: A Defense of Unqualified Medical Confidentiality

low, especially considering the popularity of Japanese sex tours to the wv-infected brothels of Thailand. Evidence of this sort can confirm that the failure to respect Confidentiality can impair the ability of doctors to do their job. And third, there is an argument based on the motivational principle that if one strongly desires that event E does not occur and one knows that doing act A will bring about event E, then one has a weighty reason not to do act A. The criminal justice system is based on this idea. We attach artificial and broadly unwelcome consequences (imprisonment and other forms of punishment) to wrongful, harmful conduct with the expectation that, even if inclined, most people will decide against the conduct in order to avoid the unwelcome consequence. If I don't want to go to prison and a career in burglary will likely result in my going to prison, then I have a weighty reason to choose a different career. Likewise, if I don't want my marriage to be destroyed by my wife's discovery that I am HIV+ and I know that telling my doctor about reconciliation will result in her discovering just that, then I have a weighty reason not to tell my doctor. The presumption must be in favor of the truth of this seemingly self-evident principle. If critics allege that it is false or otherwise unworthy of endorsement, it seems the burden of disproof belongs to them. It is their responsibility to come up with disconfirming evidence. It can be argued, in rebuttal, that people still commit burglary and, despite reporting laws, people still go to doctors for HIV testing, even knowing that confidentiality has its limits. But no one would maintain that punishing convicted criminals totally prevents crime and that breaching confidentiality results in all people avoiding or delaying medical treatment or concealing aspects of their lives. The situation is more complicated. Consider that Andrew belongs to one of two groups of prospective patients. Members of the first group are willing enough to have reports made to others. Members of the second are deterred from disclosure by the fear of a report. Of course we can't know in advance which type of patient Andrew is, but if both groups are treated alike, uncertainty will not be a problem. (While this division into two groups may be oversimplified, working through the qualifications would take us too far afield.) Consider the first group: patients who would be willing to have a report made. Recall that the physician in the Case of the Infected Spouse tried to obtain assurance that Wilma would be protected. Under an unqualified confidentiality rule—no exceptions—if the patient were willing to have reports made to others,

the doctor should be able to obtain a waiver of confidentiality, and Wilma could then be informed. Once permission .to report is given, the ethical dilemma disappears. 'Notice that for this group of patients, an exceptionless confidentiality rule works j ust as well as a rule requiring doctors to override confidentiality when necessary to protect endangered third parties. At-risk parties will be warned just the same but with appropriate permission from patients. In these cases there is no need to trim back the obligation of confidentiality since patients in this first group are, by definition, willing to have a report made. Difficulties arise with the second type of patient: those who will not want credible threats reported. Notice that these prospective patients are in control of the evidence doctors need to secure protection for parties at risk. If a patient cannot be drawn into a therapeutic alliance—a relationship of trust and confidence—then doctors will not receive the information they need to protect imperiled third parties (at least so long as patients have options). As a result, doctors will not be able to mobilize protection. When one traces out the implications of a reporting rule on what needs to be said in 1990 (when Andrew asked to be tested and the doctor disclosed the limits to confidentiality), it becomes evident that Wilma will not be protected if Andrew (a) does not want her to know and (b) understands that disclosure to his doctor will result in her knowing. Depending on his options and the strength of his preferences, he will be careful about what he discloses to his doctor, or will go without medical advice and care, or will find another physician who can be kept in ignorance about his personal life. We began by characterizing the Case of the Infected Spouse as an apparent head-on collision between the doctor's duty of confidentiality and the duty to protect imperiled third parties. But if the argument above is sound, there is no collision. The obligation to warn third parties does not provide added protection to those at-risk In particular, a no-exceptions confidentiality rule has a better chance of getting the facts on the table, at least to the extent that honest promises of confidentiality can make it so. To be sure, clinicians would have to set aside the vexing "Should I report?' conundrum and search for creative solutions instead. These strategies will not always prevent harm, but they will sometimes. The nub of the matter is that these strategies can never work if they can't be implemented. And they can't be implemented if the fear of reporting deters patients from disclosure. Accordingly, there is no justification for trimming back the obligation of confidentiality since doing so actually reduces protection to endangered third parties, increasing public peril.

II

al

liIIlI

3 Management of Medical Information

The argument advanced here is that—paradoxically— ethical and legal duties to report make it less likely that endangered parties will be protected. Depending on the prospective patient, these duties are either unnecessary (when waivers can be obtained) or counterproductive (when disclosure to the doctor is deterred and interventions other than disclosure are prevented). In part, the conventional wisdom on confidentiality errs in focusing on the decision of the individual clinician at the point when the choice has to be made to disclose or not. The decision to violate confidentiality reaches backwards to the HIv test administered years earlier and, as we shall see, even before. Perhaps little will be lost if one doctor betrays a single patient one time or if betrayals are extremely rare. But medical ethics is not about a single decision by an individual clinician. The consequences and implications of a rule governing professional practice may be quite different from those of a single act. Better to ask, what if every doctor did that? While it is accepted here that doctors have an overriding obligation to prevent public peril, it has been argued that they do not honor that obligation by breaching or chipping away at confidentiality. This is because the protective purpose to be furthered by reporting is defeated by the practice of reporting. The best public protection is achieved where doctors do their best work and, there, trustworthiness is probably the most important prerequisite. Physicians damage both their professional capabilities and their communities when they compromise their trustworthiness. If the argument above is sound and confidentiality must be respected in this case, we must now return to the question of what the doctor must say to Wilma when, now infected, she returns to the office two years after the reconciliation. Though this question has finally to be faced in 1993, it is on the table before her return to Honolulu. It is there even before Andrew asks to be tested in 1990, and you then have to decide whether to live out the trust he has placed in you or disabuse him of it. In fact, the problem is on the table in 1982, when the couple first enters your office and asks you to be their physician. As a doctor, you have obligations of beneficence and confidentiality, and you owe both to each. But now—having read this far—you are aware that something can happen that you cannot control; and, if it does happen, you will face those apparently conflicting obligations. You can only provide what you owe to one if you betray your obligation to the other. That is the choice you will have to make in 1993, unless you (and the medical profession) contour professional responsibilities now. in choosing a governing ethical principle, the end-in-view is to protect vulnerable third parties, and

if this can be done best, as I have tried to show, by honoring confidentiality and doing one's best to protect imperiled third parties within that framework, then what you must say to both Wilma and Andrew, when they enter your office in 1982, should be something like this: There is an ethical problem physicians sometimes face in taking on a married couple as patients. It can happen that one partner becomes infected with a transmissible disease, potentially endangering the other. If the infected partner won't share information with me because he or she fears I will warn the other, there will be no protection at all for the partner at risk There may, however, be things I can do if I can talk with the infected partner. What I promise both of you is, if that were to happen, I will do everything I possibly can to protect the endangered partner, except for violating confidentiality, which I will not do. You both need to remember that you should not count on me to guarantee the wholesomeness of your spouse, if doing this means betrayal. It is in these words that the final explanation to Wilma can be found. If Wilma understands from the beginning that medical confidentiality will not be breached, if she (and the public generally) understand that the precariously placed are safer under unqualified confidentiality, she will understand she has final responsibility for her choices. If you are clear enough about it, she will grasp that she can't depend on you to protect her at the cost of betrayal and that she is better off because of that. Both the doctor and the medical profession collectively need to work through these issues and fully disclose the favored standard to prospective patients before the occasion arises when a doctor must appeal to it. The view defended here is that the profession should continue to make an unqualified pledge of confidentiality and mean it. It is also appropriate to consider what should be said to Andrew as he is about to leave your office in 1991 to prepare for a romantic dinner with Wilma I once spent part of an afternoon with a healthcare professional who had served in Vietnam. He had counseled married enlistees who had returned from visits with their wives and had been diagnosed with a venereal disease that was probably contracted before they left Vietnam. It is likely that these men may have infected their wives. This clinician had learned how to persuade these men to agree to disclosure. He stressed that their wives would likely find out eventually and that the emotional and medical consequences would be far more severe because of the delay. More importantly—given

Kipnis: A Defense of Unqualified Medical Confidentiality

the soldiers' tentative decisions not to let their at-risk spouses know—he would ask whether this was a marriage they really wanted to preserve? I recall that he claimed a near perfect record in obtaining permission to notify the at-risk spouses. It would be useful if there were skilled allied caregivers, bound by confidentiality, who could routinely conduct these specialized counseling sessions. While this is not the place to set out the full range of options for a profession reliably committed to trustworthiness, it will suffice to point out a direction for professional and institutional development. Concluding Remarks Even if the forgoing is accepted, what may trouble doctors still is a fear that they will learn about an endangered person and be barred by this no-exceptions confidentiality rule from doing anything. Actually, there is only one thing they cannot do: disclose. All other paths remain open. Even if a reporting rule keeps many prospective patients out of the office or silences them while they are there, the rule protects doctors from the moral risk of having to allow injury to third _ parties when a simple disclosure would prevent it. This distress is significant and has to be faced. Here we must return to an error discussed earlier: the conflation of personal morality and professional ethics. Like law, personal morality can also conflict with professional responsibility. We considered a Jehovah's Witness surgeon, morally prohibited from administering blood transfusions to patients needing them. Likewise a Catholic doctor may be unable to discuss certain reproduction-related options. And despite understandable moral misgivings, doctors everywhere must be prepared to administer high-risk treatments they know will cause the deaths of some of their patients. Paradoxically, a personal inability

to risk killing patients can disqualify one for the practice of medicine. While personal morality can play a decisive role in career choice, it shouldn't play a decisive role within medical ethics. Many enter medicine believing that good citizens must prevent serious injury to others, even if that means violating other obligations, But the task of professional ethics in medicine is to set out principles that, if broadly followed, will allow the profession to disc charge its collective responsibilities to patients and §-0-ciaY, Confidentiality, I have argued, is effective at getting more patients into therapeutic alliances more quickly, it is more effective in bringing about better outcomes for more of them, and—counterintuitively—it is most likely to prevent serious harm to the largest number of at-risk third parties. Now it is ethically praiseworthy for honorable people to belong to a profession that, on balance, diminishes the amount of harm to others, even though these same professionals must sometimes knowingly allow (and sometimes even cause) harm to occur. Although doctors may feel guilty about these foreseeable consequences of their actions and inactions, they are not guilty of anything. They are acting exactly as it is reasonable to want doctors to act. It is hard enough to create therapeutic alliances that meet patients' needs. But if doctors take on the added duty to mobilize protective responses without waivers of confidentiality, their work may become impossible in too many important cases. And all of us will be the worse for that. The thinking that places the moral comfort of clinicians above the well-being of patients and their victims is in conflict with the requirements of professional responsibility properly understood. While it will be a challenge for many honorable physicians to measure up to this standard, no one ever said it was easy to be a good doctor.

References Beauchamp, T. L., and J. E Childress. 2001. Principles of biomedical ethics. New York: Oxford University Press. Bok, S. 1983. Secrets. New York: Pantheon Books. Goldman, A. 1980. The philosophical foundations of professional ethics. Totowa, NJ: Rowman & Littlefield. Kipnis, K.1986. Legal ethics. Englewood Cliffs, NJ: Prentice-HalL Kipnis, K. 2004. Gender, sex, and professional ethics in child and adolescent psychiatry. Child and Adolescent Psychiatric Clinics of North America. 13(3): 695-708. Kottow, M. 1986. Medical confidentiality. An intransigent and absolute obligation. Journal of Medical Ethics. 12: 117-22.

Lo, B. 1995. Resolving ethical dilemmas: A guide for clinicians. Baltimore Williams and Wilkins. Reddy, D. M., R. Fleming, and C. Swain. 2002. Effect of mandatory parental notification on adolescent girls' use of sexual health care services. Journal of the American Medical Association. 288: 710-714. Wicclair, M. 1986. A shield right for reporters vs. the administration of justice and the right to a fair trial: Is there a conflict? Business & Professional Ethics Journal. 4(2).

II

,111

1111.111

3 Management of Medical Information

Breaching Confidentiality Dave Unger

E. . .1 Information that is revealed to someone such as a health care provider can be potentially compromising (embarrassing, financially compromising, legally compromising, etc.) to that individual. Confidentiality is the concept that this information can be kept safe and only be revealed to authorized persons—persons authorized by the patient. In general, the scope of that authority is very narrow and very carefully defined and scrupulously defended. This is fundamental to a notion of respect . . . for autonomy. Put respect for autonomy is only one of the principles in bioethics, and where autonomy competes with the beneficence (and or non-maleficence) of the greater society or of another individual, that is the place where contemplating breaking confidentiality becomes a necessary consideration. .1. . [C] onfidentiality (as an individual right) is carefully protected, but in the interest of the competing collective concerns of the population or society, it is sometimes necessary to breach it. What follows is a discussion of some of the situations in which this can and must occur, the ethical rationale for warranting these occurrences, and an examination of the relevant guidelines policy and law in this arena. The discussion starts with a look at notification: the mandatory notification and reporting of such things as child abuse and communicable diseases. Next the discussion turns to the notion of a duty to self report, with a focus on the archetypal example, HIV infection. Keeping with the HIV theme, the notion of a "duty to warn" is introduced, that is, the duty of a health care professional to warn a third party who is at risk of harm from a patient under the care of that professional. This concept is further developed with a look at genetic diseases and finally this discussion funnels into a discussion of duty to warn in general and the relevant law and policy that guides health care providers in this difficult area. 1) Reporting and Notification

First, to a relatively non-contentious topic: the duty to report child abuse. A just society provides for its vulnerable and its voiceless members. Sanctity of the family is acknowledged in Canada. The respect for the autonomy of the family in governing its own affairs and directing the course of childrearing according to family (or cultural) values is fundamental to Canadian society.

However children suffering abuse or neglect at the hands of their parents, or when the parents are making unwise decisions for them, each province has a mechanism by which children can be protected. Discussion in this section is brief not because this topic is unimportant but rather the contrary, it is so important there is little disagreement (in principle) about the proper governance and enforcement of this area: there is a duty to breach privacy and confidentiality considerations in favour of the welfare of children in every region of Canada. This is one area where "Good Samaritanism" is not only morally required, but legally required. Section 215 of the, Canada Criminal Code outlines the duty parents or guardians owe to children: Everyone is under a legal duty as a parent, foster parent, guardian or head of a family to provide necessaries of life for a child under the age of sixteen years, to provide necessaries of life to their spouse or common-law partner, and to provide necessaries of life to a person under his charge if that person is unable, by reason of detention, age, illness, mental disorder or other cause, to withdraw himself from that charge, and is unable to provide himself with necessaries of life.' Breach of this duty is an offence under the Criminal Code, and hence, is a federal concern. But in child protection the prosecution of errant parents under criminal law is not the issue of primacy, child protection is carried out under civil law. The courts have inherent jurisdiction over incapacitated persons including children, under the general and ancient common law provision of parens patriae—the legal concept by which the state is given power to intervene and protect those who are being abused or injured and are unable to protect themselves. In effect, parens patriae is the legal instrument by which the state can effectively become the "parent" when a real parent or caregiver is effectively lacking or negligent. This common law mechanism is not often invoked as there is almost universal statutory law across Canada that makes the means of child protection more explicit. Each province has a child protection agency that is run by the ministry responsible for family and child development. These ministerial departments have different names in different provinces (it is called the Ministry of Children and Youth Services in Ontario, and the Ministry of Children and Family Development in BC) and they refer to this issue variably as child welfare or child protection. Their mandate is established by provincial legislation:

Source: From The Canadian Bioethics Companion, Chapter 2 (2011). Reprinted with permission of the author.

Unger: Breaching Confidentiality

in BC, this is the Child, Family and Community Service Act; in Ontario, thkChild and Family Services Act; and in Alberta, the Child, Youth and Family Enhancement Act. It is mandatory in Canada to report abuse or neglect of children to the ministry. It is important to note that it is mandatory for anyone to report abuse that is witnessed or suspected, not just health care workers. Children are deemed to need protection if they are physically or sexually abused, emotionally harmed, deprived of necessary health care,2 or if parents are absent from home or have otherwise abandoned their children and are unwilling to provide parenting. The legislation typically protects the identity of the reporting persons as well. Children can be apprehended by law enforcement, hospital personnel etc., and are usually given over to social workers or similar agents. Immediate health concerns are managed before the children are ultimately found safe living conditions—usually this is in the home of another family member or a foster home. The case of a child needing protection will be decided in a court. Sometimes this means several hearings and navigating a complicated legal process before a definitive outcome is found. Ultimately, the goal is to return children to their home of origin or their family once the potential for abuse is mitigated or the situation has been dealt with. The child protection laws give authority to child protection workers to follow up on the safety of the children long after they are returned to their parents or families and sometimes the children are supervised or returned gradually to live with their parents.3 Moving from the concerns of the vulnerable individuals to the public health concerns of a population, there are other forms of mandatory reporting that are not predicated on "good Samaritan" principles but are nonetheless good for society and overstep confidentiality of patient information. In Canada there are a whole array of situations that require reporting under the various health acts, motor vehicle acts, and vital statistics acts (etc.) of the provinces and territories. These include unexpected deaths and births and many, many matters in between. All deaths must be recorded as vital statistics. Deaths must be reported to the coroner's office if they are unexpected, self-induced or accidental, or if they occur outside a hospital for a patient that is not known to be terminally ill. Births are also required to be recorded. Accidents that occur at a worksite are required to be reported and managed through the appropriate Worker's Compensation Board (or similar agency) in each province. In addition most provinces have registries of certain medical conditions, such as cancer registries that compile statistics on and monitor follow-up of cancer patients. There_are_times __when confidentiality can be breached to protect members of the public from harm.

There is a duty to report impaired physicians and a duty to report physicians suspected of sexually assaulting patients and to report impaired flight personnel— these serve also as a means of professional regulation. There is a duty to report impaired or unfit drivers (such as those with seizures or dementia). This list is not comprehensive but serves to point out the extent of the paternalism and the range of harm reduction tactics that are invoked in the name of the greater good. It also highlights the degree to which breaches of confidentiality occur within a nation that holds autonomy so dear. With an appeal to a (legitimate) public health agenda, Canada has several mechanisms in place for the protection of its citizens against certain communicable diseases. This is one of the few legislated (albeit very limited) breaches of confidential information of patients per se. It is mandatory to report diseases (both new and emerging, as well as old and now rare) that pose the threat of rapid spread, difficult containment, or that pose threat of significant morbidity. Reports are made to provincial health authorities (usually to a "Provincial Health Officer" or "Chief Medical Officer" of the various provinces), but these authorities differ in their scope. For the most part, tracking and reporting is automatic and there are robust federally or provincially mandated agencies at the helm. Most of this reporting is not done with the immediate knowledge or consent of the patients involved; and for the most part this is understood as acceptable by citizens in the society. Communicable diseases which are reportable, or notifiable, are listed in the Notifiable Diseases Summary at the Public Health Agency of Canada (PHAC).4 There are provincial and territorial statutes in each region stipulating the reporting of these diseases provincially and ultimately nationally and a report is published as the Canada Communicable Disease Report (ccDR) to monitor trends and keep surveillance.5 These diseases include sexually transmitted diseases such as chlamydia and gonorrhoea and the re-emerging syphilis. Reporting also includes now rare diseases that were once common but are carefully monitored in light of attempts at vaccine eradication. Examples include such things as measles, mumps and rubella. Significant chronic infections such as Hepatitis C and HIV are reportable. Syphilis is an example of why it is important to report infections inasmuch as monitoring has revealed new outbreaks and vulnerable populations that are affected by this new visitation of an old scourge.' Other infections are reportable not only to monitor demographics and aid prevention, but also to monitor treatment and follow-up; the example of tuberculosis (TB) will be discussed in the next section It should be noted that infectious disease reporting while effective

uW I

II

11

111111

3 Management of Medical Information

within a province, and while effective in conveying statistics to national agencies, are not perfect inasmuch as contact tracing (wherein the public health agency responsible will track down contacts of the index case or infected patient) between provinces often breaks down. Patients have no choice over the reporting of these diseases; their confidentiality is implicitly waived. It is the responsibility of health care providers to report this fact to patients before the testing is ordered so that patients can make an informed decision about their rights and, in effect, voluntarily waive their confidentiality. All this is not to say that their confidentiality is entirely waived, while these diseases are reportable, it simply means that persons other than the health care provider and the patient have access to the results. These results do not go beyond the disease control authorities and any follow-up contact tracing done by these authorities is anonymous— patients' information is not public, the confidentiality seal is broken and the envelope is simply expanded. There are diseases that lie in ethical grey zones. It is mandatory to report HIV in all provinces and anonymous contact tracing is provided by the various disease control agencies in the provinces (again, this is sometimes ineffectual between provinces). This arrangement has driven the emergence of a new issue and a hot topic among those interested in public health ethics, one that is especially acute in the management of HIV. Breaches of confidentiality are allowed in the reporting of HIV so it has been questioned whether people will forego testing on these grounds alone. That is, it has been a growing fear that people will remain intentionally ignorant of their infection to spare themselves the stigma and embarrassment of notification and contact tracing, and this situation, of course, puts the broader society in harm's way as possible infections go unchecked and unreported. One theoretical way to resolve this issue would be to allow anonymous testing of individuals, they (theoretically) could spare themselves the pain of shame and stigma, and society could then rely on them to be good stewards of this personal information and use appropriate discretion in their sexual and drug use habits. A complete discussion of this cannot be had here, nor are there agreed upon resolutions of the argument in either the empirical or theoretical literature. Nonetheless, there remains a loud cry for the availability of anonymous testing in Canada (and the US). There is no true means of anonymous testing at hospitals and labs in most provinces as positive tests will be traced back to identify and record the patient and the disease and contact tracing will be attempted. There are, however, specialized facilities in many Canadian cities that attempt to provide anonymous (so called "non-nominal") testing whereby a patient's identifiable information (such as name) is not attached to the lab

result.' If the US is any indication, this is a trend that will likely continue to grow: in the US as of 2005 there were 45 states that offered anonymous testing.8 2) Disclosing and Warning

Reporting is akin to disclosing when the information is transmitted, not to a government agency, but to other personal stakeholders. In discussing the obligation to self report (disclose), HIV is a prime example. There are several aspects to this issue. First: consider the obligations that a health care provider has to his or her patients. Much of the above discussion involves the respect for the autonomy and the confidentiality of patients, but one other way in which the interests of the autonomous person must yield to the needs of the many is in the doctor patient relationship where there exists the potential for harm to patients. In this respect, the old adage, "first do no harm" is immediately apparent as operational—non-maleficence must come before autonomy. As discussed above, in every province there are professional colleges responsible for the licensing and the discipline of physicians, dentists, midwives, nurses, podiatrists and other specialties that provide patient care (for example the College of Physicians and Surgeons of each province). Health care providers that deliver this care have an ethical responsibility to report to their professional college-the presence of a diagnosis that could compromise their patients—the classic example is the surgeon who has HIV. The college can then, at its discretion, restrict or alter the licensing of these professionals, but it is usually left to the professional to self report in the first place, and hence restrict his or her own practice so as to not compromise his or her patients. The next aspect of the duty to report is the duty a person has to report his or her own HIV status to a sexual partner. (This can be viewed as a transition to the issue of "duty to warn" dealt with immediately below) While there are several specific American laws on this count, Canadians rely on sections of the Criminal Code for sufficient jurisprudence in this real& Although not strictly speaking the first case, the most prominent case and the one that has set the stage for future precedents was that of R v Cuerrier wherein Henry Cuerrier was charged and ultimately convicted of assault for having sex with two separate partners.w Here it was found that because, although the women consented, they were not told of his HIV status and therefore (taking a page from the bioethics playbook) could not be considered to have given proper informed consent. Since then, there have been over 70 such cases in Canada, and some HIV advocates would suggest that they are tendering an ever increasingly pernicious attitude in law against persons with HiwAiDs.11 In April 2009, Johnson Aziga was convicted in Ontario for

Unger: Breaching Confidentiality

two counts of first degree murder (the first conviction on this charge in Canada) for infecting two women who died and several counts of aggravated assault.12 3) Duty to Warn As was mentioned above, there are duties to report many diseases and circumstances, but where this crosses over to a duty to warn the public or a specific individual is a significant departure from the confidentiality issues that have been discussed so far, and a significant area of contention in the study of public health ethics. Where a duty to report becomes a Dutyto-Warn is another area in which confidentiality is waived and represents an ethical minefield that has not been entirely mapped(The basic logic here follows the "Harm Principle": that. individual rights and autonomy must give way when there are risks of harm to other Carrying on in the theme of HIV, there is an obligation to report HIV to provincial health officers, and then a duty for said officers to attempt contact tracing of possible sexual contacts and give information to contacts about the need to be tested. (This is standard with any STD or reportable disease, and the health officers' duty is usually dispatched through its mandated delegate, the provincial centre for disease control.) All of this is done while maintaining the anonymity of the index case. But precisely when it becomes a duty to broadcast the status and identity of a potentially dangerous infectious person to the general public is a topic which is long on speculation, yet short on empirical evidence and legal backing. Police forces have been criticized for releasing media advisories on known HIV positive persons, and on the other end, governments have been criticized for not warning about potentially dangerous HIV positive persons. Thus far, there have been no legal cases specifically involving either a duty to warn, or the breach of duty to warn, for a health care provider to warn about an HIV infection or HIV infected person.1" The years to come will be fraught with challenging cases and evocative new policy as this disease pushes these ethical issues to new extremes and with a new urgency." While much of the foregoing discussion surrounds HIV—as it exemplifies the ethical problems and defines the contour of the law and policy of Canada in these ethical realms—there are further, broader concerns about a duty to warn that move beyond this archetypal disease situation. Before moving to a general discussion on duty to warn and protect, a digression is warranted into the realm of genetic diseases and the obligations that flow from their diagnoses. Just as serious socially transmissible diseases evoke questions of duty and obligation, so too do serious genetically transmissible diseases evoke such questions.

Much is made in the bioethical literature on defining the full extent of this issue, and much ink spilled also on the implications of legislation and policy in this realm. Genetic issues are simultaneously individual and familial in nature, the one facet merges into the other, and obligations to the patient sometimes collide with obligations to patients' families, and this, in turn, places a health care provider on very thin ice. Part of the problem in this realm is in the determination of risk and imminence, and whether this information would be certain to benefit the third party—these issues are considered below. The uncertainty of these factors gives duty to warn a dubious legitimacy, yet the magnitude of the diseases considered strikes fear that failing to warn could be catastrophic. As the risk of injury stemming from ignorance is amplified, the case for a duty to warn becomes more compelling, In Canada there is no specific statute or law which governs this problem, nor has there been a case to test the matter of whether it is ethically justifiable or obligatory to inform family members of a serious genetic illness diagnosed in one's patient.16,17 Canada can look to significant cases internationally for guidance, as these issues have been forced upon other countries and where they have been decided in courts there are precedents for a duty to warn, and where professional advisory bodies have been forced to scrutinize the issue, policy has condensed into guidelines that favour a duty to warn under specific circumstances." In the US there have been three cases that have found a physician has a duty to warn the relatives of a patient with an identifiable genetic disease, or at the very least, warn the patient that he or she has the duty to warn the family members.19-21 This area remains unclear but it is possible that some of the law and policy recommendations discussed below may, in theory or in practice, come to bear on this issue. What is certain is that in the rapid advance of medical and genetic technology, this is sure to be an ethically charged area that will continue to be confronted and shaped in the years to come. At present, the current ethical guidelines seem to accommodate these issues at least to some extent. Current good clinical practice guidelines of any Canadian professional body would certainly accommodate general recommendations. Before a diagnosis is even made, genetic counselling should take place: counselling that sets a clear course for the patient regardless of the diagnosis, and a clear plan of action for the disposition and disclosure of the information once it is available. In this manner, the issue of a duty to warn is likely to be stemmed before it becomes an issue. Once a diagnosis is made, it is advisable that physicians (or more likely and more appropriately, a genetic counsellor) first ask the diagnosed patient to disclose of their own accord (even facilitating a family conference or helping to draft family letters to enhance

II ;A

1111.11.1

3 Management of Medical Information

understanding). Failing this, the physician (or counsellor) shouldnext try to gain the consent of the index case before disclosing to family members.17 Beyond this there are no firm guidelines but the good judgment of the clinician and decisions should be made after consideration of the general circumstances under which a duty to warn is warranted, and discussion will now turn to this. Generally and conceptually, the notion of a duty to warn jibes with basic ethical considerations; there is a point at which rights to individual autonomy must yield to the collective concerns of the broader society. Precisely where this threshold is located is a challenge for bioethicists (and public health ethicists) and for health care providers on the front line. The problem is fairly straightforward (and a detailed discussion of this will be left to the bioethical and public health ethics theorists and to other texts): the relationship between a care provider and patient is often very intimate, and trust and confidentiality are integral to actually making this relationship work. Betraying confidences, even while advancing the needs of the broader public, may utterly destroy the provider—patient relationship and, in turn, erode the foundation of the care giver—patient enterprise. And this enterprise, the Canadian health care and social welfare system, is sacred, dearly cherished, fiercely defended, and scrupulously maintained and paid for by the Canadian public—threatening it is something Canadians do not have much appetite for. Where a dearth of concrete empirical data and a paucity of incontrovertible ethical doctrine create a vacuum, the space is often filled by legislation. Patients' confidentiality is protected by statutory law and policy as was noted above, but there are no specific laws to instruct just when to break these confidences aside from the situations already noted regarding driving, flying, birthing or dying. There is no statutory legislation on obligations placed on health care providers to warn or protect their patients. Rather, this obligation is defined in Canada by case law. These concepts called "duty to warn" (which was introduced above) and "duty to protect" have a fairly rich case law evolution in Canada, but before a discussion of this, it is necessary to make some distinctions between these terms. Duty to warn means an obligation of the health care provider to warn a Third party (or potential third party) of danger from a patient of that health care provider. Duty to protect goes beyond duty to warn and includes the duty to protect a third party—this can include actions performed on the threatening patient, such as holding them against their will under a provincial Mental Health Act.22-23 As discussed above, in the Canadian framework most health professions are self-regulatory; that is, they are governed by professional colleges that have

codes of ethics and conduct, and more importantly are sanctioned by the provincial health acts to license and govern their members. Furthermore, regulatory bodies are to ensure that "best practices" are implemented by their members. It is this arrangement that gives the legal instruments their teeth. Enforcing a duty to warn or penalizing a health care provider for not carrying out this duty is an issue fraught with controversy and difficulty. Patient advocacy groups have in some instances pointed out that the law does not require a health care provider to breach confidentiality but rather pqrmits him or her to do so. Stemming from concerns noted above, this is particularly the case with HIV notifications; HIV patient advocates are quick to point out that the common law does not oblige, but rather grants discretion in disclosure. It is difficult to define just how far to go in carrying out the obligations. Many health care providers are poor assessors of risk or predictors of the actions of their patients, and this imposes a rather openended obligation on them to protect or warn. Legal analysis has suggested that the liability that providers have is in the assessment of the degree of violence or harm that can be done and not in prediction of probability— the magnitude, not the likelihood, takes precedence." There is a rich history of case law in this realm. The American case of Tarasoff v Regents of the University of California is the case that is often first cited (even in Canadian law) in the development of an understanding of the duty to warn. In this case a psychologist was deemed to have an obligation to warn and protect third parties (in this case Tatiana Tarasoff) threatened by one of his patients.25 Canadian case law evolved more slowly and points of law established in increments. It was established that physicians were responsible to act reasonably to warn and protect the general public from harm, both identifiable and non-specific non-identifiable victims.26." Furthermore, a patient's chart and information could be commandeered for the purposes of preventing injustices to others.28 But it was the landmark case of Smith v Jones that set the bar in this topic." In this case a man (Jones) was charged with assaulting a prostitute. He was assessed, at the request o f the defence, by a psychiatrist (Smith) who determined he was very likely to repeat his offence and kill or maim prostitutes in the future. While Jones ultimately pleaded guilty, the issue became one of whether the psychiatrist could disclose the information about his dangerous patient to the appeal judge. Indeed Dr. Smith earnestly wanted to disclose this information as he feared for the greater public. Ultimately, the ruling in the Supreme Court of Canada held that the information was to be revealed to the court, and that the counsellor—client privilege (between a lawyer and his or her client), and by extension the physician—patient privilege, should be sacrificed for the

Unger: Breaching Confidentiality

greater good of_protecting_the public. The opinion of the court outlined several factors that were to be considered in breaking confidential relationships and these continue to provide basic guidance in this realm. So after all this theoretical and legal consideration, there are a few practical things that can be said to guide health care providers in their duty to warn and they were derived in the following forums. In 1998, a working group of medical experts in Ontario established several guidelines for the duty to warn. These included the general recommendations that a physician (and this can be broadened to include other care providers) must warn if a patient threatens a third party with serious harm and if that harm is "more-likely-than-not" to be carried out. This was true of threats on identifiable individuals ("I am going to kill Mary Smith"), or non-specific yet credible and well planned threats ("I am going to plant a bomb in this place . . ." or "I am going to continue to molest 8 year old boys near my home . . ."). Further recommendations were that there should be some standard by which to evaluate these threats and their risks, and in the face of uncertainty that an evaluation by another colleague, as to the extent and seriousness of the threats, should be sought. Finally, there were recommendations that all threats should be seriously considered by health care providers and that it was better to err on the side of caution in notifying authorities.24 These general rules

were adopted by the Royal College of Physicians and Surgeons, the Canadian Medical Protective Association (CMPA), and Ontario professional colleges. A year later the Smith case developed a notion of which factors should be considered in adhering to these recommendations including: what the specific threat is, how serious the threat is, how imminent the threat is and how likely to be carried out (whether the patient actually has the means, etc.), whether the intended victim is specific or general, whether there is a hist ry of this type of violence in the patient's past. Th CMPA which is the institution which oversees the medico- egal affairs of physicians, still adheres to these general principles and uses the Smith case for illumination.30 In sum, in dispatching these duties to warn and protect, the recommendations backed by law have been variously interpreted, but in the broadest of terms include some of the following guidance. If a threat is likely and imminent, and if the potential for serious harm (physical or otherwise) is great, care givers should warn the local law enforcement and even, in some cases, the third party. It is also included in the duty to protect that the care giver could consider detaining the patient under the provincial mental health act against his or her will. Further, almost every professional society recommends that in cases like these the care giver should seek legal counsel of his or her own as soon as possible. F. . .1

Note

The copyright holder for this article would like it noted that the book from which this reading is derived is not a peer-reviewed publication. References 1. 2.

3.

4.

5.

Government of Canada. Criminal Code of Canada. R.S.C. 1985, c C-46, as amended. Section 215 (1). Necessary health care is, of course, a very open-ended term, and can include such things as refusing to give consent for an operation, blood transfusion, cancer treatment, etc. The parents can refuse to consent to these things for themselves but cannot refuse to allow an acceptable standard treatment for their dependent children. Children are not always taken out of or returned to families, these laws pertain to any person that is the legal guardian of children, family or otherwise. The Notifiable Diseases Summary can be viewed at the Public Health Agency of Canada website: http://www.phacaspc.gc.cdpublicat/ccdr-rmtc/04vol30/dr3021b-eng.php. The Canada Communicable Disease Report can be found on the Public Health Agency website: http://www .phac-aspc.gc.ca/index-eng.php.

6. For the record. CMAJ. August 12, 2008;179(4):316. 7. Information on these sites can be obtained through the Public Health Agency of Canada website: http://www .phac-aspc.gc.ca/aids-sida/info/4-eng.php. 8. According to research by the Henry J Kaiser Family Foundation. http://www.kfr.org/hivaids/upload/Updated-Fact-Sheet-HIV-Testing-in-the-United-States.pdf. 9. Grant I. The boundaries of the criminal law. the criminalization of the non-disclosure of HIV. Dalhousie Law Journal. Spring 2008(31). 10. R v. Cuenier; [1998] 2 SCR 371 (Supreme Court of Canada 1998). 11. Symington A. Criminalization confusion and concerns: the decade since the Cuerrier decision. HIV AIDS Policy Law Rev. May 2009;14(1):1, 5-10. 12. R v Aziga, at the time of this writing this case is still open to appeal, and it is certain the final page has not been turned on this case.

II

Lill

IIIIJJ

3 Management of Medical Information

13. One would be remiss to fail to mention a key concept in public health ethics that concerns such as these hinge on; and that is the "Harm Principle." This fundamental concept is attributed to John Stuart Mill: "That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others" (Mill J. On Liberty, 1859). This is, in essence, a justification for intervention by the state that warrants infringements on personal autonomy in the name of harm prevention or reduction. In public health practice this is most commonly considered in the context of a duty to protect the public from harm. 14. A notable exception may be found in Pittman Estate v. Bain (1994) 112 DLR (4th) 258 Ontario Central Division, (Ontario Supreme Court 1994) wherein Dr Bain failed to notify his patient that he could be HIV positive through contaminated blood. In this case the patient was HIV positive, and the doctor was found liable for damages to his patient but also, by extension to the patient's wife whom he also, indirectly, failed to warn. 15. HIV is being used as an exemplary and archetypal case more for the purposes of demonstration of ethical issues and the contour of law and policy in Canada, but the utility of this has its limits as HIV becomes imbued with more peculiarities that conjure new ethical problems. Some of these problems have to do with the changing nature of the disease which, on one hand, represents dangers of transmission if untreated, and on the other hand, relatively minimal risk of transmission if well treated. One new issue is now becoming just where to draw the line in a duty to disclose or warn: does one have an obligation to disclose if one's virus is uncontrolled but not if one's virus is well treated (undetectable in the blood)? There are a host of new problems that attach to HIV specifically that are beyond the scope of this text. 16. Gold J. To warn or not to warn? Genetic information, families, and physician liability. McGill Medical Journal. 2004;8(1):72-78.

17. Lacroix M, Nycum G, Godard B, Knoppers BM. Should physicians warn patients' relatives of genetic risks? CMA J. Feb 26 2008;178(5):593-595. 18. Godard B, Hurlimann T, Letendre M, Egalite N. Guidelines for disclosing genetic information to family members: from development to use. Fam Cancer. 2006;5(1):103-116. 19. Pate v Threlhel, 661 So.2d 278 (Florida 1995). 20. Safer v. Estate of Pack, 677 A.2d (NJ Superior Court App Div 1996). 21. Molloy v, Meier 679 NW.2d 711 (Minnesota 2004). 22. This distinction is important inasmuch as it can mean that confidentiality is not breached at all, and the remedy is to simply detain the patient. 23. Canadian Psychiatric Association. The Duty to Protect. CPA Position Paper 2002. http://publications.cpa-apcorWmedia .php?mid=154&xwm=true. Accessed 28 February 2011. 24. Ferris LE, Barkun H, Carlisle J, Hoffman B, Katz C, Silverman M. Defining the physician's duty to warn: consensus statement of Ontario's Medical Expert Panel on Duty to Inform. CMAJ. Jun 2 1998;158(11):1473-1479. 25. Tarasoff v Regents of the University of California, 17 Cal. 3d 425, 551 P2d 334, 131 Cal. Rptr 14 (California Supreme Court 1976). 26. Carver PJ. Chapter 10: Mental health law in Canada. In: Downie JG, Caulfield TA, Flood CM, eds. Canadian health law and policy. 3rd ed. Toronto: LexisNexis Canada; 2007:399-436. 27. Gray JE, O'Reilly RL. Clinically significant differences among Canadian mental health acts. Can J Psychiatry. May 2001 ;46(4):315-321. 28. R v Ross (1993), 119 NSR (2d) 177 (CA) (Nova Scotia Court of Appeal 1993). 29. Smith v Jones, (1999] 1 S.C.R. 455 (Supreme Court of Canada 1999). 30. Evans KG, Gowling Lafleur Henderson LLP (General Legal Counsel). A medico-legal handbook for Canadian physicians (Sixth edition). Ottawa, Ontario: Canadian Medical Protective Association; 2005.

How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

contribute to new concerns over privacy, confidentiality, and data protection. Two court cases, one in the United Kingdom and one in the United States, provide opportunities for thinking through ethical issues related to these developments. Each case involved selling data for marketing prescription drugs, and in each case the court decided in favor of selling the data. However, the cases were decided on different grounds, raising more general issues of secondary use of health data and the growth of health-related databases, data sharing, data aggregation, and biometric identification.

Bonnie Kaplan Introduction Electronic health records, data sharing, big data, and secondary use of health data enable exciting opportunities for improving health and healthcare. They also

Source: From Cambridge Quarterly of Health Care Ethics 25 (2016) (2), 312-29. Reprinted with permission of Cambridge University Press Journals.

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

Significant health data protection, policy, and ethical considerations are inherent in these cases. The cases call into question just what constitutes "privacy" and "public interest," and considerations for balancing them. They provide an opportunity to weigh privacy against the numerous beneficial uses of data: for individual patient care, public health, research, biosurveillance, and marketing. The cases prompt ethical questions of commodifying medical information and of harmonizing policy across jurisdictional boundaries. They raise concerns of how health data can, and should, be used. Their consequences may affect biomedical informatics, patient and provider privacy, and regulation in ways this article explores, both in the United States and elsewhere. How health data can, and should, be used is at the intersection of public health, research, care, privacy, and ethics. This article provides an ethical analysis of these interwoven ethical issues involving appropriate, judicious, and ethical secondary data use, reflecting a more general discussion of big data benefits and harms, and touching on common dualities of the individual versus the aggregate or the public interest, research (or, more broadly, outside the health field, innovation) versus privacy, individual versus institutional power, identification versus identity, identification versus authentication, and virtual versus real individuals and contextualized information.' I start by discussing what makes health data special, including international consensus on the importance of the clinician's duty of confidentiality and on health data privacy or protection. Next I summarize the court cases. Then I consider who benefits from data disclosure and aggregation, and secondary use for data mining, research, and sale. Throughout, I highlight potential benefits and harms and argue that transparency flexibility, and accountability is needed. Ethical and policy analysis should assess data uses and users, as some are more compatible with societal norms and values than others. Considering how health data should be used in light of these issues suggests policy opportunities concerning patient data and privacy protection. As the use of electronic health records, electronic medical devices, mobile and e-health applications, and biometric, social and behavioral, and genomic data spreads, these considerations are becoming more relevant worldwide. What's Special about Health Data?— International Principles

All countries recognize confidentiality as a patient's right2 that is good for individual patients and

clinicians, and for society as a whole.3 Intimacies are revealed in the interest of good healthcare, so clinicians' professional and fiduciary duties include a duty of confidentiality. Therefore, health information is given special protection internationally, though specific ways of achieving it differ. Lifestyle choices, reproductive abilities, and stigmatizing conditions are considered highly sensitive. But what is included in these categories differs with cultural background, from place to place, and from time to time. What is considered very private, embarrassing, stigmatizing, or grounds for discrimination varies among individuals and groups.4 Countries, likewise, vary in what personal information is treated as needing restricted collection, use, and disclosure.5,6 They also balance privacy and other considerations differently; thus privacy protection is more lax in some places than in others. In India, for example, the judiciary considers privacy on a case-by-case basis, as an exception to the rule that permits government interference in private life. Unlike in Europe and the United States, public interest, welfare, and safety take precedence over individual rights, liberty, and autonomy.' Yet, as discussed subsequently, even if individual clinicians scrupulously meet the professional obligation of confidentiality, confidentiality can be compromised by legal requirements to collect, document, and disseminate personal health information, especially when maintained in computer databases that can be combined easily with other sources of information about the person.8 What patients reveal for the purpose of healthcare may then be used in ways they never intended. Privacy practices have not caught up to these trends. Fair Information Practices and De-identification

The same Fair Information Practices (FIPs) underpin privacy policies in both the European Union and the United States. The European Union and the United States each protect personal data, including data concerning health, albeit differently. The United States approaches privacy by sector; separate laws address confidentiality in distinct domains, such as finance and healthcare. Health data privacy collected in the course of clinical care is governed by the U.S. Health Insurance Portability and Accountability Act (HIPAA) of 1996, extended by the HIPPA Privacy Rule in 2001 and again in 2013 by changes mandated by the 2009 Health Information Technology for Economic and Clinical Health (HITEcH) Act (part of the American Recovery and Reinvestment Act [ARRA] of 2009) and the Genetic Information Non-Discrimination Act (GINA) of 2008.9'1°,11

.11111.1

3 Management of Medical Information

The European Union takes a more comprehensive general approach to privacy; Article 8 of the European Convention on Human Rights includes the right to data protection. This right is embodied in the 1995 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.12 Member states implement directives differently, but the EU Data Protection Regulation establishes a single set of rules for data protection across the EU; the final texts are expected to be adopted by the European Parliament at the beginning 2016 and the new rules to become applicable two years after.13,14 Despite their differences, both the United States and the EU construe privacy as control and protection of data rather than other conceptions of privacy15 Both the United States and the EU also make special note of health information, and both rely on stripping data of content presumed to identify the individual represented by the data. As Paul Ohm points out: "In addition to HIPAA and the EU Data Protection Directive, almost every single privacy statute and regulation ever written in the U.S. and the EU embraces—implicitly or explicitly, pervasively or only incidentally—the assumption that anonymization protects privacy, most often by extending safe harbors from penalty to those who anonymize their data."16 As these safe harbors stipulate, neither HIPAA nor the EU Data Protection Directive applies after data are de-identified. However, relying on de-identification contributes to what is considered an inadequate and problematic legal framework for data protection.17 Addressing concerns over de-identification "would require a significant shift in approach towards data protection across Europe."18 Similar deficiencies plague the United States.'9,2° Privacy protection, then, depends not merely on de-identification but on expectations, transparency, and how data are used. De-identification, or anonymization, presumes that it is possible to identify and enumerate the kinds of data that might contribute to privacy risks and to specify how to prevent harms,21 that such a list is static and sufficient in all contexts,22 and that there are no privacy harms if the individual is not identified, even though individuals may object to uses of their personal data even if they themselves are anonymous.23 Furthermore, HIPAA permits secondary uses of data for research, public health, law enforcement, judicial proceedings, and other "public interest and benefit activities," without individual authorization, thereby assuming that "public interest" is clearly understood.24,25 All are questionable assumptions.

Duty of Confidentiality

Health data privacy relates not only to expectations about privacy in general but also to norms involving professional practice, privilege, autonomy, paternalism, and protected communication and the duty of confidentiality, as well as to requirements for data collection, dissemination, and retention. Physicians and nurses have duties both to their individual patients and to the health of their communities.26 At least since the time of the Hippocratic Oath, it is believed, societal norms and common law have recognized that clinicians' duty to patients includes maintaining confidentiality, except where protecting the public interest or other individuals requires overriding it. The International Code of Ethics of the World Medical Association (wMA) makes respecting the right to confidentiality a duty integral to a physicians' responsibility to patients.27 The WMA Declaration of Helsinki—Ethical Principles for Medical Research Involving Human Subjects (revised 2013) places a duty on physicians "to protect the life, health, dignity, integrity, right to self-determination, privacy, and confidentiality of personal information of research subjects . . . even though they have given consent."28 Recognizing that this personal information, whether collected for research or clinical practice, increasingly is held in databases, in 2002 the WMA adopted the Declaration on Ethical Considerations Regarding Health Databases: "Confidentiality is at the heart of medical practice and is essential for maintaining trust and integrity in the patient—physician relationship. Knowing that their privacy will be respected gives patients the freedom to share sensitive personal information with their physician."29 In this 2002 declaration, the WMA reaffirmed that violating this duty could "inhibit patients from confiding information for their own health care needs, exploit their vulnerability or inappropriately borrow on the trust that patients invest in their physicians" while at the same time recognizing the value of secondary health data use for quality assurance, risk management, and retrospective study3° Thus, a key reason for treating health data as requiring special protection is to maintain trust between clinician and patient in the interest of both social and public order as well as better care for each individual patient. In recognition of this ethical duty, confidentiality is seen worldwide as a health professional's legal duty, one that protects the professional from giving legal testimony, thereby serving the interests of patient and public by maintaining trust during medical encounters. Nowhere can private data about a patient rightly be passed to a third party without that patient's

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

permission, except as required by law. French criminal law makes this universal spirit apparent by criminalizing a physician's breach of confidentiality even in court testimony even if the patient would allow it.31 How people value and respond to concerns about health data privacy is affected by context and common expectations of privacy.32 Many recognize that clinicians need highly personal information in order to care for them, and, because of the long-standing history of trust in professional confidentiality, such patients willingly share sensitive information with those who treat them. As Deryck Beyleveld and Elise Histed eloquently point out: Information that patients provide for their treatment is about very personal and sensitive areas of their lives. Indeed, it relates to their very existence, both physically and symbolically. As such, it is not information that they may be presumed to be prepared to disclose or have used freely. It is their vulnerability, constituted by pain and distress, or fears about their health and lives, that leads them to disclose this information to health professionals. At the same time, people are apt to attach great importance to intimate information about themselves and their bodies, and this can be associated with mystical and religious beliefs, which by their very nature can be idiosyncratic.33

was a plaintiff in one of the court cases) unsurprisingly takes the position that analyzing doctors' prescribing habits contributes to patients becoming informed consumers.36 Yet patients can be harmed when data about them are used to violate privacy: to deny employment, credit, housing, or insurance, and for identity theft and other unsavory purposes. Some fear that patients who are insecure about the confidentiality of prescription or other health record information may withhold information, refuse diagnostic and genetic testing, or decline electronic prescriptions.37,38,39 People do change their behavior and withhold information in order to protect their health information privacy.4° Even before the widespread use of electronic health records, a 2000 Gallup poll indicated that the vast majority of people in the United States opposed thirdparty access to medical data without a patient's permission, and, furthermore, that 67 percent of those polled opposed the release of data to medical researchers.41 Similarly, the Pew Internet and American Life Project reported that, to protect privacy, according to a 1999 survey, nearly one in six patients withheld information, provided inaccurate information, doctorhopped, paid out of pocket instead of using insurance, or even avoided care. Eighty-five percent feared that seeking health information on the Internet would result in changes in insurance coverage or otherwise reveal their information.42

Patient Benefits and Harms

Individuals also may provide health information freely via health-related social networking, web postings, and searches; or because it is required legally, as for prescriptions. Such information could be consolidated and linked with other data for beneficial or nefarious purposes, sometimes without individuals' knowledge. Patients benefit from having their record information available from previous clinical visits, whether or not those visits were with the same clinician or in the same facility, because clinicians can make better care decisions in light of fuller understanding of their patients' past clinical histories. Patients also benefit from public health surveillance and research that depends on combining health information from individual patients to improve public health and develop better treatments. Patients may benefit from making identifiable information concerning adverse drug events available to pharmaceutical companies so that those companies can follow up with patients and improve drug safety, as Source Informatics argued in the U.K. court case discussed subsequently, and as the International Pharmaceutical Privacy Consortium argues more generally.34,35 Data aggregator ims Health Canada (outs Health, Inc.,

Transparency and Consent

As information resources become more ubiquitous and information sharing becomes more profitable, more thought is needed concerning which data uses are acceptable and what control individuals should have over data about themselves. Privacy violations may compromise patient care, the information in patients' records, and patient—clinician relationships. The principles of data protection—transparency, legitimacy, and proportionality—embodied in the EU Data Protection Directive, therefore, specify that the person from whom data are obtained should be informed of what will be done with this information and to whom it will be disclosed. This allows the individual to consent or object and to withdraw or correct the data. Also, according to the directive, the data should be kept only as long as necessary for the specified purpose,43 even though that could compromise later retrospective research. Patients' privacy concerns are exacerbated when patients, and even clinicians, have little idea of what becomes of their data, or just what is protected and what is not.44 Withholding information from one's

3 Management of Medical Information

clinician is neither in the public interest nor beneficial to that patient's individual interest in proper healthcare. Yet, removing identifying information from patient records may not alleviate concerns, especially in light of increasing public awareness of privacy violations surrounding big data and the ease with which data sets that were meant to be kept apart now are combined and used for reidentification.45,46,47,48,49 Further, without transparency, consent is meaningless. Two Court Cases

Two court cases provide occasion for thinking about the ethical implications of data sale and secondary use in light of international principles of health data privacy and protection. Each case involves selling prescription data for pharmaceutical marketing. In both the United States and the United Kingdom, data aggregators successfully challenged restrictions on such data use and sale. The 2011 U.S. Supreme Court case Sorrell v IMS Health Inc. et al.50 was decided on free speech grounds. Although the legalities involve unique features of U.S. constitutional law, a similar case in the U.K. in 2000, R v Department of Health, Ex Parte Source Informatics Ltd.," points to the international nature of the ethical issues. That case was decided on the grounds that selling anonymized (de-identified) data did not violate pharmacists' duty of confidentiality. The decision in each case runs counter to public expectations of health data confidentiality. The public is hardly aware that aggregating and selling prescription and other health data are an international enterprise. Thus, the Sorrell and Source cases raise more general global concerns of privacy and data protection, on the one hand, and appropriate use and secondary use of data for data mining, marketing, research, public health, and healthcare, on the other. Elsewhere I address data de-identification, prescription and other health data aggregation and sale, and issues more specific to these two cases.52 This article explores other ethical issues related to the cases—the benefits and harms of data sale; the trade-offs among privacy, individual health, and public health; and the need for transparency—so ethical dimensions of responsible and ethical health data collection and use can be assessed. Who Benefits?

Clinical data include data that patients are required to provide to receive care. In both the Sorrell and Source cases, prescription data was aggregated and sold. Patients, prescribers, and pharmacies are required by law to collect information related to prescribing. Data

aggregators perform a valuable service in collecting, cleaning, and combining these and other data into useful resources, though the value does not accrue directly to those who are the original source of the data Data aggregators should be compensated for the value added, but the sources deserve some benefit as well. Currently, they primarily bear costs, both financially and in privacy. The combination of required disclosure of personal data and the ease with which data can be collected and disseminated is not unique to pharmacies. It is a cost of healthcare to collect and store patient records, a cost passed on to patients and payers, whether private or governmental. The organizations providing these data obtain it from those legally required to provide it—from individuals who pay directly, or indirectly through their private or public insurers, for its collection and maintenance. These individuals gain little direct benefit from the aggregation and sale of data about them, and they may be harmed by it. It mostly occurs without their knowledge or permission. Even in light of arguments that patients should be required as a condition of treatment to allow data about them to be used for research—a requirement counter to professional norms to provide care—it seems improper to require either patients or clinicians to disclose data they would otherwise choose to keep private so that others may financially profit from them, whether or not the data are de-identified. Secondary use and big data analytics also are affected by the costs of collecting, storing, and organizing data, as well as by the costs of meeting regulatory requirements. To reduce costs, health data processing is outsourced from countries with stronger privacy protections to countries with weaker ones, despite its sensitive nature and consequent privacy risks." Also to reduce costs, U.S. marketing organizations oppose opt-in consenting on the grounds that it would increase the cost of doing business." But costs must be paid somehow. Both the Source and Sorrell cases were fought to protect the commercial value of health information. One way of recovering costs is by selling these data Though some sources provide some data sets at little or no cost to researchers, cost could make it easier for pharmaceutical companies and other commercial enterprises than for researchers to access data.",56 Some fear that the trend toward treating data as private property could make it more difficult to develop comprehensive databases crucial for public health and research.37 Research, trade, and innovation, as well as the globalized healthcare industry, provide considerable public benefit. There are ethical as well as economic

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

costs to privileging privacy, but economic value may not be more important than privacy or other considerations. Law and common ethical practice prevent releasing medical information without a patient's permission, but U.S. law does not prevent selling or transferring rights to records.' Data that can be sold, can be sold and replicated anywhere and, once sold, may be used for good or ill. Tracing the chain of data sales and use is difficult, making transparency and consent nearly impossible the further data are transferred from the original source. Health Data Uses: Big Data, Data Mining, Research, and Biosurveillance

Electronic health records and health information networks provide a wealth of data for public health, health outcome improvements, and research. Data could be used for a range of beneficial purposes, from outcomes and comparative effectiveness research to designing clinical trials and monitoring drug safety. The benefits of these data for public health, marketing, research, drug development, identifying adverse effects, and biosurveillance; for reducing costs and overprescribing; and for regulating devices and software all are intertwined with privacy concerns. For some of these purposes, it is crucial to be able to identify individuals and link together an individual's records, so a requirement for de-identification may further impair research. However, there also could be harms. Patients may withhold sensitive information if they fear it will be used against them, even though it may be useful for other purposes. Studies based on analyzing large data sets could be compromised if individual prescribers or patients withhold information or their consent for data use.59 Privacy advocates, researchers, and public health officials can be at odds over how to achieve benefits while protecting privacy, their disagreements may stem from different values and historical legacies. For example, the U.K.'s National Health Service (NHS), Royal College of Physicians, and the Wellcome Trust led a coalition of leading medical research organizations opposed to the proposed European General Data Protection Regulation, which, unlike the Data Protection Directive it would replace, would bind all 28 member countries. The proposal was acceptable to most EU nations; the European Parliament approved the committee report in full in 2014.60 The regulation affects any organization that gathers, processes, and stores data, whether operating within the EU, doing business with organizations within the EU, or storing data in EU-member countries. As of this writing, most organizations were not

ready for compliance. Research organizations were among those concerned about its impact. It is especially relevant here that the regulation defined personal data as any information about an individual, whether it relates to his or her private, professional, or public life; and thus such data includes medical information. Much of these personal data—a name, a photo, an email address, bank details, posts on social networking websites, or a computer's IP address61—too, are part of medical records. The original proposed regulation, therefore, increased health data protection and would have made illegal the NHS mass database of citizens' health information, which could provide a valuable resource for improving care.62,63,64 Opposition from the NHS and other research organizations contributed to changes put forward by the EU justice ministers in March 2015 to improve data sharing across healthcare services. They also tabled amendments regarding how to manage such special forms of data as health and genetic data, and when patient consent is needed.65 The European Parliament, the Council, and the Commission agreed on the new regulation late in 2015 and it is expected to be adopted by the European Parliament at the beginning of 2016.66 This NHS database also provokes privacy concerns while providing financial benefit, as the NHS sells the data.67 Individuals can opt out of the new care.gov database, which was to contain, for the first time, records from primary care (GP) practices. Privacy concerns delayed including those GP records." Although other rules allow greater third-party access to other NHS databases,69 insurers, pharmaceutical companies, and other private commercial enterprises will receive "pseudoanonymized" records that the NHS claims "will not contain information that identifies you," but that instead include NHS numbers, birth dates, postcodes, and ethnicity and gender information.70 The database was created, according to the NHS England website, to improve NHS services,n and to "drive economic growth by making England the default location for world-class health services research."72 In the United States, too, researchers and bioethicists recognize that privacy protections can impede research and healthcare quality improvement, with calls from such influential agencies as the Institute of Medicine to change the HIPAA Privacy Rule to allow for information-based research—that is, research using medical records or stored biological samples.73 Some innovative approaches to meeting privacy research, and commercial needs for data sharing include the new international Open Humans Network, which "attempts to break down health data silos through an online portal that will connect participants willing to

■ II

I

Al

II

dill

111141

3 Management of Medical Information

share data about themselves publicly with researchers who are interested in using that public data and contributing their analyses and insight to it,"74 and businesses based on similar ideas, such as PatientsLikeMe. Using the data people post, PatientsLikeMe produces publishable material on patient outcomes and comparative effectiveness, which is valuable for effectiveness research. Epidemiologic trends also can be identified through social media postings.75'76'77 Those engaging in this social networking presumably feel it is beneficial to them. Even so, it would be better if they were aware of what is done with their data, instead of being surprised if they have not read subscription agreements carefully enough to know that PatientsLikeMe sells data to pharmaceutical and other companies and that sites such as Face book are not private places.'" Who Sells and Uses Data? One Man's Bread Is Another Man's Poison

As is evident from the multiplicity of uses, health data are valuable. Internationally, the idea of "liberating" data for secondary use is recognized as beneficial for individual and public benefit, research, entrepreneurship, and policy. Though transborder data flow is regulated by international agreements, such as the EU Data Protection Directive, presumably health data could be sold worldwide, to anyone, for any purpose. Balancing this with privacy concerns is fraught.8° Strong privacy protection, such as the rights-centric approach of the European Court of Human Rights, could adversely affect the globalized healthcare industry and innovation and trade.8182'83.84 Entire patient records are among the many possible sources of data for which there is a lucrative market, for laudable as well as unsavory purposes. Incidents of medical identity theft increased by more than 20 percent in 2014 compared to 2013.85 In the active black market in identifiable medical record information, health information is more valuable than U.S. Social Security numbers for identity the ft.86'87 Though prices vary, such information sells for about ten times more than credit card numbers (which typically sell for no more than a few U.S. dollars) because it can be monetized by getting treatment paid for via identity theft or to extort money from hacked corporations.88'89 Electronic records also make it possible for computer or software vendors, intermediaries, or newly created organizations to bundle and sell rights and data," a practice useful for research, policy, marketing, and business. In the United States, there is an exhaustive list of organizations that can use and legally sell health information,91 some for purposes patients and

clinicians would not anticipate. Data sold by both U.S. state and federal agencies can be linked to individuals by using publicly available information, even if some of the data are de-identified:92,93 Some may consider what is done with these data as harmful to some of the individuals who have provided the data and, at the same time, as beneficial to other individuals, depending on what the data reveal. This combination of benefits and harms is evident in a variety of examples in which one's records affect one's services and costs. In the United States, where private medical insurance is the norm, private insurers use prescription and other claims data to deny insurance, charge differential premiums, or exclude some conditions." Businesses often check the MIB (Medical Information Bureau) for job applicants' underwriting data." Aggregators purchase and combine data from the states as well as from pharmacies." Credit agencies are the most frequent buyers of multistate health profiles, though ims Health also purchases data from the states.97 Government fusion centers, designed to promote data sharing among federal agencies and state and local governments, combine data from multiple sources—including health record information—for law enforcement, immigration control, and homeland security."'" Organizations, too, may benefit financially while providing social benefit through data sales. The American Medical Association and the U.S. Centers for Medicare and Medicaid sell provider data, whereas state Health Information Exchanges (Ins) sell secondary data. loo,1o1,1oz The U.K.'s National Health Service, too, sells data.103 Insurance companies or health information technology vendors might aggregate and sell provider-identified data on performance and quality measures, the number of procedures performed, U.S. meaningful use criteria, data security breaches, and other useful compilations. Cash-strapped community health organizations, state Regional Extension Centers (REcs), county hospitals, the U.S. Veterans Administration, the Indian Health Service, the Joint Commission, or hospital associations also could sell data for similarly beneficent purposes. Hospitals routinely sell birth records.'" Genetic data are also double-edged. Such data are needed for research, personalized medicine, and biobanking but also can make individuals and communities vulnerable. For example, in 2000, Iceland's parliament sold exclusive rights to all the genetic and genealogical data from each of its 275,000 citizens to the U.S. company deCODE Genetics. Soon thereafter, deCODE signed a $200 million contract with Hoffman LaRoche to search for several common human genetic

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

diseases. Iceland had an opt-out policy, and the data were encrypted to de-identify individuals. Nevertheless, the Icelandic Supreme Court later ruled that creating the database was unconstitutional because it did not adequately protect personal privacy105 Clearly, provider or patient information is valuable. Hospitals could purchase data about competitors, providers could identify populations for treatment, researchers could conduct studies involving healthcare and public health practices, and government agencies could identify and influence health trends. If such sales were restricted, some fear, the data would not be collected or maintained at all, which could compromise research and new drug development.106,107 The Iceland genetic database sale, for example, led to identification of genes linked to disease,1°8,1°9 though capitalizing on these kinds of discoveries was limited to the company with exclusive rights to this gene discovery. DeCODE's 2009 bankruptcy and the consequent database ownership change from a scientific research company to Saga Investments LLC, and the subsequent sale of the database in 2012 to biotech pioneer Amgen, again raised questions about data privacy and use.110,111 Countries as different as Canada, Estonia, Sweden, Singapore, and the Kingdom of Tonga have developed various models for protecting privacy and differing policies regarding commercial involvement and rights to samples for gene banks, all with the goal of improving the public health of the studied population, and, in some cases, to generate revenue for national healthcare budgets. Though all these policies are intended to maintain confidentiality, all of the data uses require personal identifiers so as to link individuals' records from genetic, medical, genealogical, and lifestyle databases. International controversy over such databases, therefore, centers around confidentiality, consent, to what extent commercial interests should influence policy, and whether commercial ownership facilitates or impedes research,112,113,114 all of which are concerns related to collecting and selling healthcare data in general. As a way of raising additional considerations, I pose possibilities that might occur were there unrestricted selling of health data. Abortion opponents presumably could buy aggregated prescription information for medications that cause abortions, or animal rights activists could buy information about researchers' animal or veterinary medicine purchases. Depending on who buys it and their purpose, such information could threaten or protect researchers', clinicians', and patients' safety and might have adverse effects on research and clinical practice or might open new avenues. Physicians, patients, hospitals, and so on, in one country may be targeted for marketing by commercial

ventures or medical tourism facilities in another. Some may welcome learning of such opportunities, whereas others may feel harassed or violated. Individuals in one country may experience salutary or salacious effects from having (identified or possibly re-identified) data available elsewhere. But without transparency, there is little chance of gaining individual consent or, on both individual and societal levels, assessing harm or benefit. Ownership, Commodification, and De-contextualization

The right to sell data is muddied by lack of clarity over the legalities of data ownership. Law in and outside the United States does not address health data ownership clearly, it is not clear who the owner should be, or whether ownership is better than current or alternative approaches.115,116 It also is not clear where those who sell data analytics services obtain the data, or how they might use them.117 Well-known electronic health record vendors have sold de-identified copies of their patient databases to pharmaceutical companies, medical device makers, and health services researchers.118 Vendor contracts are unusual in that some vendors lay claim to patient record data, whereas businesses and financial institutions typically do not give up their data to their software vendors.119 Regardless of whether the data themselves or the means of access to them are owned by electronic health records vendors, some academic medical centers pay to get data from their own patients' records. Vendors often consider their contracts intellectual property and do not reveal these and other contract provisions, a practice the American Medical Informatics Association considers unethical.120 If health data are property, presumably, whoever owns the data can sell them. Some advocate clearly defined property rights in medical information, giving patients the right to monetize their access and control rights, as a way for individuals to control and benefit from what happens to data about them.121 Others argue against property rights in patient data and advocate instead for public ownership akin to a data commons so that data from multiple sources can be de-identified and combined population-wide for public benefit.'22 Commodifying medical information strikes still others as anathema to professional values and the special relationship between doctor and patient. Privacy is valued because it facilitates ideals of personhood involving autonomy, individuality, respect, dignity, and worth as a human being.123 Therefore, the idea of selling personal health data also disturbs those who think the practice commodifies the self and sullies ideas of personhood.124,125 Compromising of personhood is

uw I

ai

I1..11

IIILIJ

3 Management of Medical Information

compounded because data in databases necessarily are decontextualized. De-identification is an attempt to remove any connection with the person, but even identifiable health record data typically do not include all information a person may consider central to the self. Conclusions Widespread use o f electronic patient record systems enables opportunities to improve healthcare through data sharing, secondary use, and big data analytics. Multiple healthcare professionals, payers, researchers, and commercial enterprises can access data and reduce costs by eliminating duplication of services and conducting research on effective care. However, widespread use of electronic patient records systems also creates more opportunities for privacy violations, data breaches, and inappropriate uses. Ethical and policy analysis related to health data and informatics should consider benefits and harms, taking into account both the uses and users of the information.126,127 Embarrassing an estranged spouse by publishing his or her mental health records is more distasteful than using those records combined with others' to study and improve mental health. As this example suggests, some users (the researcher) are more appealing than others (the spouse). Moreover, an uncontroversial use may be morally offensive if the user is unsavory or controversial.128 How should distinctions be made so that some data uses and users are permissible and some not? On what grounds? And who is best placed to make such decisions: the courts or legislators, clinicians and researchers who are most familiar with their data needs, companies that develop and market new medications, or patients and prescribers, who are most affected by privacy violations and can best weigh the relative importance of various values to themselves.'29 Those most familiar with, closest to, and affected by the potential use should have a strong say. They need to know about those uses, though, to express their preferences in an informed, thoughtful way. Many patients do not know what is, or can be, done with data about them, but keeping them ignorant is not the way to address concerns. Lack of accountability and transparency about health data uses feeds the public's privacy concems,"° undermines the possibility of informed consent, and impairs research, care, and public health. Ethical considerations over data use will, and should, evolve as the public becomes more aware of the value and pitfalls of data sharing, data aggregation, and data mining. Cases like Source and Sorrell encourage debate over propriety and values related to different kinds of data use. They also lead to examining when it

is in the public interest for personal health data to be made available, just what that "public interest" is,131,132 and, for that matter, just what "privacy" comprises and entails as norms evolve.133 The issues include considering, in a healthcare context, the dualities playing out with respect to big data in domains other than healthcare: the individual versus the aggregate, research versus privacy, individual versus institutional power, identification versus identity, identification versus authentication, and virtual people versus real people and contextualized information. They involve big data harms and benefits related to innovation and economic advancement, power shifts, access to knowledge, and freedom of communication. Societies and governments need to grapple with these ethical issues, tensions between privacy and other considerations, and shifting norms. The numerous cross-cutting issues suggest that other areas of law, ethics, and social policy also can inform related ethical and legal considerations. For some time, the legal, bioethics, and informatics communities have been considering issues such as appropriate secondary use of data; patient and clinician relationships in light of the growth of electronic health records and health information technologies;134,135,136 reliance on increasingly untenable de-identification; burgeoning electronic data collection, sharing, transmission, and aggregation; data use for public health, research, and innovation; and the privacy and security of health data. As health information exchanges and health tourism develops; as lifetime electronic health records that follow patients across governmental and institutional boundaries are used more widely; as databases grow and biobanks become digital; as biometric identification becomes more common; as radio frequency identification devices (RFips) are embedded in medical devices, smart-pills, and patients; as home sensors and monitors are increasingly used; as mobile, wearable, and e-health applications expand; and as health information exchanges develop,137,138,139,140 informaticians can add to the conversation among governments, courts, regulatory agencies, professional societies, and other organizations to consider responses to issues involving health-related data. Combining legal and ethics scholarship with informaticians' expertise concerning judicious and ethical data collection and use, together with their technical knowledge of data aggregation and identification, can contribute to more informed policies. The Source and Sorrell court cases can provoke an initial reaction of outrage over privacy violations and data use without consent. Consequently, they call into question just what constitutes "privacy" and "public interest" and stimulate considerations as to how to

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

balance them. They provide an opportunity to weigh privacy against numerous beneficial uses for data. Transparency and accountability are needed so that harms and benefits can be judged through public discussion and so that individual as well as societal decisions can be made on more informed and thoughtful grounds. Using data collected for one purpose (such as prescriptions) for another purpose (such as pharmaceutical marketing) can undermine public confidence,

especially i f the public is unaware of the reuse. Doing so without individuals' permission violates international principles of data privacy.141,142,143,144,145 The court cases prompt ethical questions about commodifying medical information and harmonizing policy across jurisdictional boundaries. Their consequences may affect biomedical informatics, patient and clinician privacy, and regulation in ways this article explores, in the United States, United Kingdom, and elsewhere.

Notes 1. Laura Wexler's comments as a respondent at "The Critical Life of Information," a conference at Yale University, April 11, 2014, outlined dualities related to big data; see http://wgss .yale.edu/sites/default/files/files/Critical%20Life%20of%20 Information%20Program%20spreads.pdf (last accessed 19 Aug 2014) for conference information. 2. Jost TS. Readings in Comparative Health Law and Bioethics. 2nd ed. Durham, NC: Carolina Academic Press; 2007. 3. Institute of Medicine (tom). Beyond the HIPAA Privacy

12.

13.

Rule: Enhancing Privacy, Improving Health through Research. 4. 5. 6. 7.

8.

9.

10.

Washington, DC The National Academies Press; 2009, at 78. See note 3, tom 2009, at 79. Jones P Permission-based marketingunder Canada's new privacy laws. Franchise LawJournal 2004; 24 (2): 267-303. Walden I. Anonymising personal data. InternationalJournal of Law and Information Technology 2002; 10 (2): 224-37. Srinivas N, Biswas A. Protecting patient information in India: Data privacy law and its challenges. NUJS Law Review 2012; 5 (3): 411-24. Kaplan B. Selling health data: De-identification, privacy, and speech. Cambridge Quarterly of Healthcare Ethics 2015; 24 (3): 256-71. United States Government, Department of Health and Human Services, Office for Civil Rights. Summary of the HIPAA Privacy Rule; available at http://www.hhs.gov/ccr/privacy/hipaa/ understanding/summary/ (last accessed 30 June 2013). United States Government, Department of Health and Human Services, Office for Civil Rights. Standards for

Privacy of Individually Identifiable Health Information; available at http://aspe.hhs.gov/admnsimp/ final/ pvcguidel .htm (last accessed 19 Jan 2014). 11. United States Government, Department of Health and Human Services, HSS Press Office, New rule protects patient privacy, secures health information 2013 Jan 17; available a t http://wwwhhs.gov/about/news/ 2013/01/17/new-rule-protects-patientprivacy-secureshealth-information.html (last accessed 1 Jan 2016). See also United States Government, Department of Health and Human Services, Office of the Secretary 45 CFR Parts 160 and 164: 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; final rule. Federal Register 2013 Jan 25: 5565-702;

14.

15. 16.

17.

18. 19.

20. 21. 22. 23. 24.

25. 26.

available at http://www.gpo.gov/fdsys/pkg/FR-2013-0125/pdf/2013-01073.pdf (last accessed 2 July 2014). European Union. EU Directive 95/46/EC- The Data Protection Directive; available at http://www.dataprotection .ie/docs/EUDirective-95-46-EC--Chapter-2/93.htm (last accessed 23 Mar 2014). European Commission, Directorate General for Justice and Consumers. Agreement on Commission's EU Data Protection Reform Will Boost Digital Single Market 2015 Dec 15; available at http://europa.eu/rapid/pressrelease_IP15-6321_en.htm (last accessed 5 Jan 2016). See also European Commission, Directorate General for Justice and Consumers. Reform of EU Data Protection Rules; available at http://ec.europa.eu/justice/ data-protection/reform/ index_en.htm (last accessed 5 Jan 2016). Rossi B. Countdown to the EU General Data Protection Regulation: 5 steps to prepare. Information Age 2015 Mar 24; available at http://www.information-age. com/itmanagement/risk-and-compliance/123459219/ countdown-eu-general-dataprotection-regulation-5-stepsprepare (last accessed 13 May 2015). Solove DJ. A taxonomy of privacy. University of Pennsylvania Law Review 2006; 154 (3): 477-560. Ohm P Broken promises of privacy. Responding to the surprising failure of anonymization. UCLA Law Review 2010; 57: 1701-77, at 270. Taylor MI Health research, data protection, and the public interest in notification. Medical Law Review 2011; 19 (2): 267-303. See note 17, Taylor 2011, at 303. Kaplan B. Patient health data privacy. In: Yanisky-Ravid S, ed. The Challenges of the Digital Era: Privacy, Information and More. New York: Fordham University Press; forthcoming. See note 8, Kaplan 2015. See note 16, Ohm 2010. See note 19, Kaplan forthcoming. Beyleveld D, Histed E Betrayal of confidence in the Court of Appeal. Medical Law International 2000; 4: 277-311. Koontz L. What is privacy? In: Koontz L, ed. Information Privacy in the Evolving Healthcare Environment. Chicago: Healthcare Information and Management Society (timitsS); 2013: 1-20. See note 19, Kaplan forthcoming. See note 8, Kaplan 2015.

mu I

H 11

MILLI

3 Management of Medical Information

27. World Medical Association. International Code of Medical Ethics; available at http://wwwwma.net/ en/30publications/lOpolicies/c8/index.html (last accessed 2 May 2014). 28. World Medical Association. Declaration of Helsinki-

Ethical Principles f or Medical Research Involving Human Subjects; available at http://www.wma.net/ en/30publications/lOpolicies/b3 Oast accessed 2 May 2014). 29. World Medical Association. Declaration on Ethical Considerations Regarding Health Databases; available at http://www.wma.net/en/30publications/lOpolicies/d1 (last accessed 2 May 2014). 30. See note 29, WMA 2014. 31. See note 2, Jost 2007. 32. Malin BA, El Emam K, O'Keefe CM. Biomedical data privacy: Problems, perspectives, and recent advances.

JAMIA (Journal of the American Medical Informatics Association) 2013; 20 (1): 2-6. 33. See note 23, Beyleveld, Histed 2000, at 296. 34. Dunkel YE Medical privacy rights in anonymous data: Discussion of rights in the United Kingdom and the United States in light of the Source Informatics cases. Loyola

47. Benitez K, Malin B. Evaluating re-identification risks with respect to the HIPAA Privacy Rule. JAMIA (Journal of the American Medical Informatics Association) 2010; 17 (2): 169-77. 48. See note 16, Ohm 2010. 49. See note 8, Kaplan 2015. 50. Sorrell v IMS Health, Inc., et al., 131 S. Ct. 2653 (2011). 51. R v Department of Health, Ex Parte Source Informatics Ltd. [C.A. 2000] 1 All ER 786. See also R v Department of

Health, Ex Parte Source Informatics Ltd. European Law Report 52. 53. 54. 55.

56.

57.

of Los Angeles International and Comparative Law Review 2001; 23 (1): 41-80. 35. See note 7, Srinivas, Biswas 2012. 36. See note 5, Jones 2004. 37. Powell J, Fitton R, Fitton C. Sharing electronic health records: The patient view: Informatics in Primary Care 2006; 14 (1): 55-7. 38. Schers H, van den Hoogen H, Grol R, van den Bosch W Continuity of information in general practice: Patient views on confidentiality. Scandinavian Journal of Primary Health Care 2003; 21 (1):21-6. 39. See note 23, Beyleveld, Histed 2000. 40. See note 32, Malin et al. 2013. 41. See note 34, Dunkel 2001, at 70. 42. Choy C, Hudson Z, Pritts J, Goldman J. Exposed Online:

58. 59.

60.

61. 62.

Why the New Federal Health Privacy Regulation Doesn't Offer Much Protection to Internet Users. Health Privacy

43. 44.

45.

46.

Project, Institute for Healthcare Research and Policy, Georgetown University: Pew Internet and American Life Project; 2001, at 4; available at http://www. pewinternet.org/files/oldmedia/Files/Reports/2001/ PIP_HPP_ HealthPriv_report.pdf.pdf (last accessed 11 May 2015). See note 12, EU 2014. McGraw D. Building public trust in uses of Health Insurance Portability and Accountability Act de-identified data. JAMIA (Journal of the American Medical Informatics Association) 2013; 20 (1): 29-34. Curfman GD, Morrissey S, Drazen JM. Prescriptions, privacy, and the First Amendment. New England Journal of Medicine 2011; 364 (21): 2053-5. Tien L. Online behavioral tracking and the identification of Internet users. Paper presented at: From Mad Men to Mad Bots: Advertising in the Digital Age [conference]. The Information Society Project at the Yale Law School. New Haven, CT; 2011.

63.

64.

65.

66. 67.

2000; 4: 397-414. See note 8, Kaplan 2015. See note 7, Srinivas, Biswas 2012. See note 5, Jones 2004. Baxter AD. ims Health v. Ayotte: A new direction on commercial speech cases. Berkeley Technology Law Journal 2010; 25: 649-70. Pasquale F Restoring transparency to automated authority. Journal on Telecommunications and High Technology Law 2011; 9: 235-54. Rodwin MA. Patient data: Property, privacy, and the public interest. American Journal of Law and Medicine 2010; 36: 586-618, at 589. Hall MA, Schulman KA Ownership of medical information. JAMA 2009; 301 (12): 1282-4. Gooch GR, Rohack JJ, Finley M. The moral from Sorrell: Educate, don't legislate. Health Matrix 2013; 23 (1): 237-77. N-is European Office. Data Protection; 2015 Mar 24; available at http://www.nhsconfed.org/regions-and-eu/nhseuropean-office/influencing-eu-policy/data-protection (last accessed 15 May 2015). See note 14, Rossi 2015. O'Donoghue C. EU research group condemns EU regulation for restricting growth in life sciences sector; 2014; available at http://www.globalregulatoryenforcementlawblog.com/ 2014/02/articles/data-security/eu-research-groupcondemns-eu-regulationfor-restricting-growth-in-lifesciences-sector (last accessed 23 Mar 2014). Farrar]. Sharing NHS data saves lives; EU obstruction will not. The Telegraph 2014 Jan 14; available at http://wwwtelegraph .co.uk/ health/nhs/10569467/Sharing-NHS-datasaveslives-EU-obstruction-will-nochtml Oast accessed 23 Mar 2014). European Public Health Alliance. [Update] General Data Protection Regulation; available at http://www.epha.org/ 5926 (last accessed 23 Mar 2014). NHS Confederation. EU ministers table changes to data privacy; 2015 Mar 13; available at http://nhsconfed.org/ news/2015/03/eu-ministers-table-changes-to-dataprivacylaws (last accessed 14 May 2015). See note 13, European Commission 2015. Doctorow C. UK set to sell sensitive NHS records to commercial companies with no meaningful privacy protections-UPDATED; 2014 Feb 4; available at http:// boingboing.net/2014/02/04/uk-set-to-sell-sensitivenhs-r .html (last accessed 5 Feb 2014).

Kaplan: How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales

68. Donnelly L Hospital records of all NHS patients sold to insurers. The Telegraph 2014 Feb 23; available at http:// www.telegraph.co.ukihealth/healthnews/10656893/ Hospital-records-of-all-NHS-patients-soldto-insurers.html (last accessed 24 July 2014). 69. See note 68, Donnelly 2014. 70. NHS Choices. Your records: Better information means better care; available at http://www.nhs.uk/nhsengland/ thenhs/records/healthrecords/pages/care-data.aspx (last accessed 24 July 2014). 71. See note 70, NHS Choices 2014. 72. Ramesh R. NHS patient data to be made available for sale to drug and insurance firms. The Guardian 2014 Jan 19; available at http://www.theguardian.com/society/2014/ jan/19/nhs-patient-data-availablecompanies-buy (last accessed 24 July 2014). 73. Institute of Medicine. Beyond the HIPAA Privacy Rule:

Enhancing Privacy, Improving Health through Research.

74.

75.

76.

77.

78.

79.

80.

81. 82. 83. 84.

85.

Washington, DC: National Academies; 2009; available at http://www.iom.edu/-/media/Files/Report%20 Files/2009/Beyond-the-HIPAAPrivacy-Rule-EnhancingPrivacy-Improving-Health-Through-Research/HIPAA%20 report%20brief%20FINAL.pdf (last accessed 22 Jan 2014). Open Humans Network. Open Humans Network wins Knight News Challenge: Health Award; available at http:// openhumans.org (last accessed 1 July 2014). Christakis NA, Fowler J H. Social network visualization in epidemiology. Norwegian Journal of Epidemiology 2009; 19 (1): 5-16. Christakis NA, Fowler JH. Social network sensors for early detection of contagious outbreaks. PLoS ONE 2010; 5 (9): e12948. Velasco E, Agheneza T, Denecke K, Kirchner G, Eckmanns T. Social media and Internet-based data in global systems for public health surveillance: A systematic review. The Milbank Quarterly 2014; 93 (1):7-33. Andrews L. I Know Who You Are and I Saw What You Did: Social Networks and the Death of Data Privacy. New York Free Press; 2011, at 1-3. Angwin J. Dragnet Nation: A Quest for Privacy, Security, and Freedom in a World of Relentless Surveillance. New York: Times Books, Henry Holt; 2014, at 33-4. Geissbuhler A, Safran C, Buchan I, Bellazzi R, Labkoff S, Eilenberg K, et al. Trustworthy reuse of health data: A transnational perspective. International Journal of Medical Informatics 2013; 83 (1): 1-9. See note 7, Srinivas, Biswas 2012. See note 17, Taylor 2011. Bambauer JR. Is data speech? Stanford Law Review 2014; 66: 57-120. Zarsky TZ. The privacy/innovation conundrum. Lewis & Clark Law Review 2015; 19(1); available at http://ssrn.com/ abstract=2596822 (last accessed 19 May 2015). Dvorak K. Med identity theft continues t o rise; 2015 Feb 23; available at http://www.fiercehealthit.com/story/medidentitytheft-continues-rise/2015-02-23?utm_medium= nl&utm_source=intemal (last accessed 14 May 2015).

86. Avila J, Marshall S. Your medical records may not be private ABC News Investigation. ABC News 2012 Sept 13; available at http://abcnews.go.com/Health/medical-recordsprivateabc-news-investigation/story?id=17228986&page=2 (last accessed 22 Mar 2014). 87. Nguyen V, Nious K, Carroll J. Your medical records could be sold on black market: NBC Investigative Unit surprises strangers with private medical details. NBC Bay Area 2013 June 18; available at http://www.nbcbayarea .com/news/local/Medical-Records-Could-Be-Sold-on.Market-212040241.html (last accessed 22 Mar B210a1c4k) 88. Lawrence D. End of Windows XP support means added opportunity for hackers. Businessweek 2014 Apr 4; available at httpi/ www.businessweek.com/articles/2014-04-04/ end-of-windows-xp-support-meansadded-opportunity-forhackers (last accesced 1 July 2014). 89. Shahani A. The black market for stolen health care data. NPR; 2015 Feb 13; available at http://wwwnpr.org/ sections/alltechconsidered/2015/02/13/385901377/ theblack-market-for-stolen-health-care-data (last accessed 14 May 2015). 90. See note 58, Hall, Schulman 2009. 91. See note 34, Dunkel 2001. 92. See note 47, Benitez, Malin 2010. 93. Roberston J. States' hospital data for sale puts privacy in jeopardy. Health Leaders Media; 2013; available at httpi/ www.healthleadersmedia.com/content/QUA-292963/Stateshospital-data-for-sale-putsprivacy-in-jeopardy (last accessed 14 June 2013). 94. Brief for the New England Journal of Medicine, the Massachusetts Medical Society, the National Physicians Alliance, and the American Medical Students Association as Amici Curiae Supporting Petitioners, William H. Sorrell v IMS Health, Inc. et al., 2010 U.S. Briefs 779 (No. 10-779), 2011 U.S.S. Ct. Briefs LEXIS 299. 95. Holtzman DH. Privacy Lost: How Technology Is Endangering Your Privacy. San Francisco: Jossey-Bass; 2006, at 195. 96. See, for example 12PC Health Data Store CMS MedPAR Hospital Data File; available at httpi/www.healthdatastore.com/ . par-hospital-data-filerspx (last accessed 13 Sept 2c111 01s3m)ed 97. [Winston JS]. States' hospital data for sale puts patient privacy in jeopardy; 2013 June 7; available at haps:// www.annualmedicalreport.com/states-hospital-data-fors2ale o1rtspatient-privacy-in-jeopardy (last accessed 19 Jan 98. Bady A. World without walls-privacy laws should be recrafted for the data fusion age Technology Review 2011; 114 (6):66-71. 99. United States Government, Department of Justice. Fusion

Center Guidelines: Developing and Sharing Information and Intelligence in a New Era; 2006; available at http://www. it.ojp.gov/documents/fusion_center_guidelines.pdf (last accessed Mar 2012). 100. See note 45, Curfman et al. 2011.

3 Management of Medical Information

101. United States Government, Department of Health and Human Services, Centers for Medicare and Medicaid Services. Agreement for Use of Centers for Medicare & Medicaid Services (CMS) Data Containing Unique Identifier s, Form CMSR-0235, OMB No. 0938-0734; available at http://www.cms .gov/Medicare/CMS-Forms/CMS-Forms/downloads//cms-r0235.pdf (last accessed 13 Sept 2013). 102. Hebda T, Czar P Handbook of Informatics for Nurses and Healthcare Professionals. 4th ed. Upper Saddle River, NJ: Pearson/Prentice Hall; 2009, at 321. 103. See note 68, Donnelly 2014. 104. See note 95, Holtzman 2006, at 192. 105. McGraw Hill General and Human Biology Case Studies. Gene Banks versus Privacy Invasion; available at http://www mhhe com/biosci/genbio/casestudies/sellinggenes.mhtml (last accessed 2 May 2014). 106. Brief for the Association of Clinical Research Organizations as Amici Curiae Supporting Respondents, William H. Sorrell v IMS Health, Inc., et al., 2011 WL 2647130 (2011) (No. 10-779), (2011). 107. See note 59, Gooch et al. 2013. 108. See note 105, McGraw Hill 2014. 109. Austin MA, Harding 5, McElroy C. Genebanks: A comparison of eight proposed international genetic databases. Community Genetics 2003; 6 (1):37-45. 110. Gillham WW. Genes, Chromosomes, and Disease: From

Simple Traits, to Complex Traits, to Personalized Medicine.

111.

112. 113.

114.

115. 116. 117.

118.

119.

Upper Saddle River, NJ: Pearson Education, published as FT Press Science; 2011, at 18-19. Amgen. Amgen to acquire deCODE Genetics, a global leader in human genetics; available at wwwamgen.com/ media/media_pr_detail.jsp?releaselD=1765710 (last accessed 2 May 2014). See note 109, Austin et al. 2003. Annas GI Rules for research on human genetic variation-lessons from Iceland. New England Journal of Medicine 2000; 342 (24):1830-3. Gulcher JR, Stefansson K. The Icelandic Healthcare Database and informed consent. New England Journal of Medicine 2000; 342 (24): 1827-9. See note 19, Kaplan forthcoming Evans BJ. Much ado about data ownership. Harvard Journal of Law & Technology 2011; 25 (1): 69-130. For example, GE Data Visualization uses information "based on 7.2 million patient records from GE's proprietary database"; available at http//visualization.geblogs.com/ visualization/network (last accessed 27 Sept 2013). GE Healthcare's healthcare IT solutions-available at http://www3.gehealthcare com/en/Products/Categories/ Healthcare_IT?gclid=CIKQ4Z6P7LkCFcE7OgodTDI APQ and http://www3.gehealthcare.com/en/Products/ Categories/Healthcare_IT/Knowledge_Center (last accessed 27 Sept 2013)-includes patient records and patient portals Sittig DE Singh H. Legal, ethical, and financial dilemmas in electronic health record adoption and use. Pediatrics 2011 Apr; 127 (4): e1042-7. Moore J, Tholemeier R. Whose data is it anyway? The Health Care Blog, 2013 Nov 20; available at

120.

121.

122. 123. 124. 125.

126.

127.

128. 129. 130. 131. 132. 133. 134.

135.

136. 137. 138. 139. 140.

141. 142. 143. 144. 145.

http://thehealthcareblog.com/blog/2013/11/20/ whose-data-is-it-anyway-2 (last accessed 3 Feb 2014). Goodman KW, Berner E, Dente MA, Kaplan B, Koppel R, Rucker D, et al. Challenges in ethics, safety, best practices, and oversight regarding HIT vendors, their customers, and patients: A report of an AMIA special task force. JAMIA (Journal of the American Medical Informatics Association) 2011; 18 (1): 77-81. Hall MA. Property, privacy, and the pursuit of interconnected electronic health records. Iowa Law Review 2010; 95: 631-63. See note 57, Rodwin 2010. See note 3, IOM 2009, at 77. See note 58, Hall, Schulman 2009. Atherley G. The public-private partnership between ims Health and the Canada Pension Plan. Fraser Forum 2011: 5-7. Miller RA, Schaffner KF, Meisel A. Ethical and legal issues related to the use of computer programs in clinical medicine. Annals of Internal Medicine 1985; 102: 529-36. Goodman KW. Health information technology: Challenges in ethics, science and uncertainty. In: Himma K, Tavani H, eds. The Handbook of Information and Computer Ethics. Hoboken, NJ: Wiley, 2008: 293-309. See note 127, Goodman 2008. Data mining case tests boundaries of medical privacy. CMAJ 2011; 183 (9): E509-10. See note 44, McGraw 2013. See note 17, Taylor 2011. See note 57, Rodwin 2010, at 617-18. See note 15, Solove 2006. Goodman KW. Ethics, information technology, and public health New challenges for the clinician-patient relationship. Journal of Law, Medicine and Ethics 2010 Spring: 58-63. Kaplan B, Litewka S. Ethical challenges of telemedicine and telehealth Cambridge Quarterly of Healthcare Ethics 2008; 17(4):401-16. See note 19, Kaplan forthcoming. See note 134, Goodman 2010. See note 135, Kaplan, Litewka 2008. See note 19, Kaplan forthcoming. Roland D. UK to get 200 high-tech factory jobs making "swallowable sensors." The Telegraph 2014 Mar 10; available at http://www.telegraph.co.uk/ finance/10687395/UK-toget-200-high-tech-factory-jobsmakingswallowable-sensors.html (last accessed 17 July 2014). See note 24, Koontz 2013. See note 44, McGraw 2013. See note 23, Beyleveld, Histed 2000. See note 12, EU 2014. Rodrigues RJ, Wilson P Schanz SJ. The Regulation of

Privacy and Data Protection in the Use of Electronic Health Information: An International Perspective and Reference Source on Regulatory and Legal Issues Related to PersonIdentifiable Health Databases. Washington, DC: World Health Organisation (wHo); 2001.

Murray: Genetic Exceptionalism and "Future Diaries" 147

3.4 Genetic Information

Genetic Exceptionalism and "Future Diaries": Is Genetic Information Different from Other Medical Information? Thomas H. Murray A few years ago, at a meeting to discuss the larger implications of the Human Genome Project, the head of the Federal Bureau of Investigation's laboratories leaned over to me and related this story. The FBI lab had conducted analyses of samples connected to the bombing of the World Trade Center in Manhattan. They had an envelope in which an incriminating document had been sent. With a technology known as PCR, polymerase chain reaction, they were able to amplify enough DNA from the back of the stamp to link it with one of the chief suspects through a genetic fingerprint. They also learned that someone else had licked the envelope itself. In other words, if someone cares enough to go to the trouble, it is possible to get substantial information about a person's genetic makeup from the tiny bits of genetic material we scatter around us without much thought. The cells mixed in with our saliva and the bulbs at the base of the hairs we continuously shed are two widely distributed sources of raw material for creating genetic information about each of us. Of course, at least some other medically relevant information about us is just as readily available. We may be pallid and cachectic, morbidly obese, rosycheeked, or missing limbs. We may be depressed or manic, delusional or paranoid, sharp as a tack or forgetful and confused. So the mere fact that people can learn things about us that are medically relevant would not seem to distinguish genetic data from other kinds of health-related information. What, if anything, makes genetic information different from other health-related information? Can it, in concept and in practice, be singled out? Regardless of whether it really is different from medical information, are there characteristics of genetic information or of the society into which it will flow that should lead us to act as if it were different? I was chair of the Task Force on Genetic Information and Insurance of the NIH-DOE Joint Working

Group on the Ethical, Legal, and Social Implications of the Human Genome Project. The task force used the term genetic exceptionalism to mean roughly the claim that genetic information is sufficiently different from other kinds of health-related information that it deserves special protection or other exceptional measures. After many attempts to make the case for genetic exceptionalism, the task force abandoned the effort. At least for the purpose of deciding who should receive health care coverage, we concluded that genetic information did not differ substantially from other kinds of health-related information.' For many reasons, we were reluctant to reach this conclusion. Our expertise lay in genetics, our mandate came from the Human Genome Project, and the problem of genetics and health insurance, as big as it was, was nevertheless tiny compared to the prospect of trying to comprehend—much less change—the entire American health care system. I will explain why we abandoned genetic exceptionalism shortly. But first I want to explore why it is that so many thoughtful and knowledgeable commentators think it is important to treat genetic information as special. In an article on genetic privacy, Lawrence Gostin predicts a move to longitudinal, electronic clinical records for each individual that would "contain all data relevant to the individual's health collected over a lifetime . . . continually expanded from prebirth to death, and accessible to a wide range of individuals and institutions."' George Annas, Leonard Glantz, and Patricia Roche, authors of a model Genetic Privacy Act, conclude a rousing defense of genetic exceptionalism with this claim: "To the extent that we accord special status to our genes and what they reveal, genetic information is uniquely powerful and uniquely personal, and thus merits unique privacy protection."3 The model Genetic Privacy Act, like other legislation intended to protect against genetic discrimination, insists both that genetic information can be distinguished from other medical information and that it ought to be so distinguished. Just why is genetic information, in these authors' words, "uniquely powerful and uniquely personal"? Annas, Glantz, and Roche offer three reasons. First, they liken a person's genetic profile to a "future diary" and claim that genetic information "can predict an individual's likely medical future for a variety of conditions." They go on to argue that we should

Source: From Genetic Secrets: Protecting Privacy and Confidentiality in the Genetic Era, ed. Mark A. Rothstein (Yale University Press, 1997). Reprinted with permission.

lb I

3 Management of Medical Information

think of DNA as a "coded probabilistic future diary because it describes an important part of a person's unique future and, as such, can affect and undermine an individual's view of his/her life's possibilities. Unlike ordinary diaries that are created by the writer, the information contained in one's DNA, which is stable and can be stored for long periods of time, is in code and is largely unknown to the person Most of the code cannot now be broken, but parts are being deciphered almost daily."4 Let us call this, borrowing a phrase from Nancy Wexler, the concern about genetic prophecy. The second reason Annas, Glantz, and Roche cite for regarding genetic information as unique is that "it divulges personal information about one's parents, siblings, and children."5 The biology of the claim is straightforward enough: we get half of our genes from each biological parent, and we pass half of ours on to each of our biological children. Genetic information about each of us, then, is also to some extent information about our ancestors, descendants, and other such biological relations as sisters and brothers. Call this the concern for kin. As their third reason, the authors cite a history of genetics being used to stigmatize and victimize people. Whether they have in mind eugenics programs, ill-conceived genetic screening, or the use of genetic information by employers, insurers, or others, they do not say. Call this the concern about genetic discrimination. Other authors have offered their own lists of reasons supporting genetic exceptionalism. Gostin, for example, offers these "compelling justifications" for protecting the privacy of genetic information: "the sheer breadth of information discoverable; the potential to unlock secrets that are currently unknown about the person; the unique quality of the information enabling certain identification of the individual; the stability of DNA rendering distant future applications possible; and the generalizability of the data to families, genetically related communities, and ethnic and racial populations.' The first two factors—breadth and the potential to unlock secrets—seem to be expansions of the concern about genetic prophecy. The last concern mentioned, the generalizability of the data, is an interesting expansion of the concern for kin. In Gostin's view, the concern does not stop at the family's door but extends to those larger groupings of people who share a certain genetic heritage. I agree. A focus on genetics emphasizes racial and ethnic differences, a very sensitive matter both now and possibly in the future. Stressing the genetics of race has the potential to intensify those divisions, while reinforcing the view that perceived differences are not mere accidents of culture and circumstance but are grounded in biology, which is itself seen as somehow fundamental

and unalterable. Looking toward the future, people might be concerned that genetic information about race and ethnicity could fall into the hands of groups or governments with hostile or totalitarian ambitions. The recent experience of ethnic conflict in Eastern Europe and Africa, as well as enduring racism in the United States, reinforces that concern. Gostin third and fourth factors—the "unique quality" that permits identification of the individual and the possible future uses of stable DNA—require some explanation. The "unique quality" is what enabled the FBI in the World Trade Center bombing case to confirm who licked the stamp and who licked the envelope. Gostin notes an interesting feature of genomic segue nee information. Genetic fingerprinting operates by detecting differences in the DNA sequences of individuals in a number of regions of the genome known to be highly variable. If a database contained sufficient information about the sequence, even if the person's name were not attached to the file, it might be possible to identify the individual whose sequence it is, in a manner similar to the method of genetic fingerprinting. So, although the practice of removing identifying information is usually thought to confer anonymity by making the records impossible to trace to an individual, that may not be the case with records containing significant chunks of DNA sequence data. (Of course, this problem is not limited in principle to genomic information. It is not hard to conceive of databases that, while containing no names, still contain enough information—e.g., community, neighborhood, age, marital status, number and ages of children, occupation, make, model and year of cars—to allow individuals to be identified.) It is also useful to distinguish between two types of genetic collections. One form is databases of genetic information that either exist in electronic form or are readily translatable into a computer-searchable database. Such databases might, in the future, include DNA sequence information about identifiable individuals; they are unlikely to do so now. Researchers and others are less interested in raw sequence than in variants of important genes—mutations and benign polymorphisms—or in short stretches of DNA that are useful for genetic fingerprinting. The other type is collections of tissue samples containing DNA. Each sample could yield, in principle, an entire sequence for each individual. Gostin asserts correctly that such tissue samples contain sufficient information to study any and all genes, as well as the DNA needed to do a genetic fingerprint, which could then be linked to the individual if his or her matching genetic fingerprint were on file. But it is crucial to understand that the

Murray: Genetic Exceptionalism and "Future Diaries"

information in such collections of genetic materials is opaque. Considerable work and expense are needed to get any of the latent information from a sample.' Gostin's worry about the durability of DNA is correct on the bare facts. In the World Trade Center bombing case, the FBI, after all, was able to take microscopic bits of DNA and identify two individuals. But without the motivation and resources required to analyze the DNA, we do not have genetic information per se, merely the raw materials from which genetic information could be derived. (But, then, we would not need a fancy tissue bank either. One of the most massive collections would be in the hands of the Publisher's Clearinghouse—all those stamps and envelopes, with return addresses yet!) Are the arguments in favor of genetic exceptionalism persuasive? The strong form of genetic exceptionalism claims that genetic information is unique—"uniquely powerful and uniquely personal," as Annas, Glantz, and Roche have written. We could admit to hyperbole in that statement and look for a weaker form of genetic exceptionalism, one that claims that genetic information is sufficiently distinctive from other information that it ought to receive greater privacy protection. The policy question, after all, is the one we are most concerned with. Genetic information does not have to be unique in order to warrant special protection, but it does have to be distinctive and especially sensitive. In evaluating the arguments, we should consider the case that could be made if only the weaker form of genetic exceptionalism is supported. The argument from genetic prophecy is not compelling. Genetic information is neither unique nor distinctive in its ability to offer probabilistic peeks into our future health. Many other things afford equally interesting predictions. Some of them would be impossible to conceal and so fall outside the concern of privacy— some people, for example, are avid skydivers or parasailers. Other types of information would be hidden, just like most genetic information; examples include asymptomatic hepatitis B infection, early HIV infection, and even one cholesterol level. These have implications for future health that are every bit as cogent and sensitive as genetic predispositions. Perhaps genetic risks ought to be treated differently because there are potentially so many of them, and such a massive prospective difference in quantity effectively makes a qualitative difference. Or perhaps what is most worrisome is that our genetic risks are occult, hidden; attributes that others could know even as we remain ignorant about them. Here lurks an image of genetic information as a mysterious, powerful, and inexorable force that will dominate and control our futures. Keep this image in mind. We will return to it.

The second argument for genetic exceptionalism is concern about kin—not merely our immediate family but perhaps also the larger ethnic community. Again, it is difficult to claim uniqueness, or even special importance and sensitivity, for genetic information. That one member of a family has tuberculosis is certainly relevant to the rest of the household, all of whom are in danger of infection, along with everyone who works with or goes to school with the infected individual. Likewise, if one partner in a marriage has a sexually transmitted disease, that information is important for the other partner. Or suppose the main wage earner in the household showed early signs of heart disease that could bring disability and death. Wouldn't the other family members have a profoundly important stake in knowing this? They might make very different career training or employment choices if they knew the breadwinner was likely to be struck down soon. Although the concern for kin could amplify the sensitivity of genetic information, it does not render that information unique. It can be very important, to be sure, but whether genetic information is important and sensitive enough to distinguish it from other sorts of information is not yet clear. The fear of discrimination is the third of the candidates for the unique character of genetic information. Here again, genetics is not alone. Institutions and individuals can and have used all sorts of information, both visible and occult, as the basis for discrimination. In underwriting for health insurance, for example, insurers use evidence of current disease or future disease risk—whether it is genetic or nongenetic doesn't matter—to decide who gets a policy, what that policy covers, and how much it costs. Whether this discrimination should be regarded as fair or unfair is debated. But it is difficult to make the argument that it is fair to discriminate on nongenetic factors but unfair to discriminate on genetic ones. Some nongenetic factors, to be sure, can be thought of as a matter of choice, in contrast with one's genes. Such risky behaviors as smoking, thrill-seeking, and the like do seem different from being struck by Huntington disease, over which we have no control. But there are plenty of nongenetic risks that no more reflect genuine individual choice than genetic risks. If the air we breathe and the water we drink are polluted, if our parents or co-workers are heavy smokers, if we are reasonably prudent but injured in an accident nonetheless, it is hard to say that we bear any significant measure of responsibility for the resulting illnesses. Likewise, most links between genes and disease are likely to be very different from the link in Huntington disease. Most will be probabilistic associations

II

Al

111111

3 Management of Medical Information

rather than straightforward causal connections. And many may be modifiable by the individual's actions. No simple equation exists between genetic factors and inexorable fate, or nongenetic factors and being open to individual choice and action. If we are less inclined to worry about discrimination on the basis of health risk factors that are open to modification and individual choice, then let us recognize that as the relevant difference, and not confuse it with the distinction between genetic and nongenetic factors. The concern for genetic information and discrimination may help explain some of the interest in genetic privacy because it broadens and sharpens important perceptions. First, it broadens the pool of possible factors that might be used to discriminate against an individual, and it likewise expands the pool of individuals who might become the subjects of discrimination. Second, it sharpens the widely held moral intuition that we should not be punished for things beyond our control. (This does not contradict the warning just stated about oversimplifying the connection between genes and illness. Health insurers, for example, work on probabilities. It might be enough that a woman has a mutated form of the breast cancer I gene [BRCAI], hence an approximately 85% lifetime risk of breast cancer, for an insurer to deny her coverage. Whose "fault" it might be is irrelevant.) Perhaps what really frightens and galls us about discrimination on the basis of genetic information is its reliance on information about us over which we have no control and may not even know ourselves. Here again is the hidden and mysterious nature of genetic information, joined with its aura of power and ubiquity, lurking close beneath the surface of our discomfort. George Annas's metaphor of our genes as our future diary captures the power and ambiguity of the personal significance of genetic information. Recall the description of an individual's genome as his or her "coded probabilistic future diary because it describes an important part of a person's unique future and, as such, can affect and undermine an individual's view of his/her life's possibilities. Unlike ordinary diaries that are created by the writer, the information contained in one's DNA . . . is in code and is largely unknown to the person."8 The metaphor is ingenious, powerful, and provocative. What features of genetic information does it capture? What is the source of its power? Does it mislead in any way? Diaries are intimate, private places where we confide what is most important about who we are, who and what we love, and what we do. Those same confidences are also just those things we may be most reluctant to broadcast. There may come a time when we

choose to share what we have recorded in our diary, but we want to be able to say when and with whom we share it. The future diary metaphor captures several sources of our uneasiness about genetic privacy. Our "genetic diary" was written not by us but by an agency completely outside our control. It can be read not only by others but without our permission or even knowledge. Last, it is possible for strangers, whose purposes are not ours, to use those genetic "secrets" to harm us. Little wonder that we want to protect whatever is written in that "diary." But the metaphor also misleads. It implies that the contents of that future diary reflect what is most intimate, central, and important about us—that it reveals, in some fundamental way, our social and personal identity, our loves and interests, and our actions. In fact, our genomes have little or nothing to say about any of these crucial matters. The metaphor also promotes genetic determinism. In complex disorders with many contributing factors, such as many cancers and heart disease, genetic information may indicate only a rough range of probabilities, something that falls far short of a "probabilistic future." So much for the case in favor of genetic exceptionalism. What about the case against it? When the Task Force on Genetic Information and Insurance rejected genetic exceptionalism, it did so with reluctance. At least for the purpose of deciding who should have access to health care coverage, the task force could find no sound way to distinguish genetic from nongenetic diseases and risks. Genetic exceptionalism depends on what we have come to call the "two-bucket theory" of disease. According to this model, there are two buckets—one labeled "genetic," the other labeled "nongenetic"—and we should be able to toss every disease and risk factor into one of the two.9 So, Huntington disease goes into the "genetic" bucket and getting run over by a truck goes into the "nongenetic" one. But many diseases and risks don't fit neatly into either bucket. Take breast cancer. Some cases of breast cancer have strong genetic roots, but others have no clear genetic connection. For that matter, not every woman with a mutated BRCAI gene will develop breast cancer. And some apparent risk factors have little or no link to genetics. Similar complexity exists for heart disease: cholesterol is a risk factor, and one's cholesterol level can be modified by diet, exercise, and other factors; but our genes have as much or more to do with the level of cholesterol circulating in our blood as our environment or behavior. Into which bucket, then, should we toss breast cancer? Heart disease? Cholesterol level? On conceptual grounds, the task force agreed that the argument for distinguishing genetic from

Murray: Genetic Exceptional ism and "Future Diaries"

nongenetic information was dubious. Certainly, there were cases of relatively unambiguous and pure genetic or nongenetic diseases and risk factors. But in many more cases, including many of the most interesting and important ones, the two-bucket theory was hopelessly inadequate. Efforts to legislate genetic privacy have confronted the two-bucket problem by confining themselves more or less to whatever fit neatly into the genetic bucket. Karen Rothenberg notes this in her study of state laws, which typically limit the definition of genetic information to the results of DNA testing.w The draft Genetic Privacy Act itself began with a conceptually rich, inclusive definition but moved to a much narrower one. This is how the act's authors describe the decision they faced: "either create a definition for genetic information that is consistent from the viewpoint of theory and principle, but not of much practical value, or design provisions that are capable of practical application that would have the effect of protecting the most private and potentially stigmatizing genetic information." Their definition reads: "The term private genetic information means any information about an identifiable individual that is derived from the presence, absence, alteration, or mutation of a gene or genes, or the presence or absence of a specific DNA marker or markers, and that has been obtained: (1) from an analysis of the individual's DNA; or (2) from an analysis of the DNA of a person to whom the individual is related."u They acknowledge that this definition leaves out biochemical tests for genes—potentially a massive gap, because what we will often be interested in clinically are the products of the genes, not the genes themselves. But they observe ruefully that amending the definition of private genetic information to include testing for gene products "makes the distinction between genetic information and medical information generally more difficult to justify."" This conceptual problem has fed another difficulty with genetic exceptionalism: how to identify and keep separate genetic from nongenetic information in the medical record. If it could not be determined whether some piece of information belonged in one category or the other, it seemed practically infeasible to divide medical records into those portions that were genetic (hence offlimits), from those portions that were nongenetic (therefore available to prospective insurers). More recently, Lawrence Gostin argued that laws offering special protection to genetic information may be problematic because "different standards would apply to data held by the same entity" depending on whether it met the definition of genetic information; because "other health conditions raise similar sensitivity issues" such as HIV and mental illness; and because it could

create enormous practical problems in record keeping and information flow.13 The task force had another reason for abandoning genetic exceptionalism: we concluded that there was no good moral justification for treating genetic information, genetic diseases, or genetic risk factors as categorically different from other medical information, diseases, or risk factors.14 If someone genuinely needed health care, it did not matter whether one could find a genetic root for the disease or whether it was the product of nongenetic bad luck or accident. Some people regard so-called self-inflicted or lifestyle-related maladies as exceptions. The argument here turns on both moral and policy premises—roughly, that to the extent that people are responsible for their own misfortune, there is no moral obligation to share the cost of their care, and that the point of access to health insurance is the proper place to extract that cost or to inflict punishment. Whether everyone finds this position persuasive is, however, not the crucial point. Our need for health care in most cases will be the product of a complex mix of factors, genetic and nongenetic, both within the scope of our responsibility and outside of that scope. The distinction between genetic and nongenetic factors is not the crucial one. We are inclined to regard it as crucial only when we fall for the overselling of genetics—when we fall prey, that is, to genetic reductionism and genetic determinism. Resounding statements about the significance of genetic information are not hard to find. Lawrence Gostin, for example, claims that "the features of a person revealed by genetic information are fixed— unchanging and unchangeable."" George Annas, Leonard Glantz, and Patricia Roche write: 'To the extent that we accord special status to our genes and what they reveal, genetic information is uniquely powerful and uniquely personal, and thus merits unique privacy protection."" Some scholars worry that the significance of genetics has been oversold. In their book The DNA Mystique: The Gene as a Cultural Icon, Dorothy Nelkin and Susan Lindee bemoan what they see as genetic hype: As the science of genetics has moved from the laboratory to mass culture, from professional journals to the television screen, the gene has been transformed. Instead of a piece of hereditary information, it has become the key to human relationships and the basis of family cohesion. Instead of a string of purines and pyramidines, it has become the essence of identity and the source of social difference. Instead of an important molecule, it has become the secular equivalent of the human soul."

II

1111.1.1

3 Management of Medical Information

Other scholars have voiced similar concerns. In a study of factors contributing to the social power of genetics, Eric Juengst identifies genetic determinism and genetic reductionism, along with their familial implications, as the sources of the anxiety caused by genetics. He notes that the link of genetics to disease in the public mind is defined by examples such as Huntington disease, in which having the mutated gene effectively determines the outcome: if one lives long enough, one will get the disease. "This history still affects the way many people think about genetic risk information, by leading them to assume that genetic diagnostics of any kind have more predictive power than other kinds of health risk assessments."18 Juengst notes that images of crystal balls and "future diaries" bolster this deterministic image of genetic information. He also remarks that such images invite fatalism and social stigmatization. Because genetic risks encourage an explanation of disease in biological terms, they are open to reductionistic accounts of both diseases and persons. That is, genetic risks may come to be seen as the explanation for complex multifactorial diseases. They may also be seen as fundamental, defining characteristics of the persons who have such risks, essentially reducing those persons to their genetic propensities. Carl E Cranor considers the concept of causation in genetics and looks critically at both those who underemphasize and those who overemphasize the causal role of genes. He acknowledges the concerns of those worried about possible social misuses of genetic explanations but worries that those critics place too much importance on minimizing the causal significance of genes, when they should be criticizing the misuses they fear. Cranor urges us to "recognize that genes, for at least some single-gene diseases (even if rare), appear to make discernible causal contributions to some diseases, but to deny that this settles all questions concerning scientific understanding, recommendations about research agendas, and the important normative and practical issues as to how such diseases should be treated or `engineered,' if at all."19 That is, we do not have to pretend that genes are unimportant to avoid determinism or reductionism. We should give genes their due, but no more than that. Evelyn Fox Keller points out the simplistic premises of genetic determinism "The idea of 'a gene for X' already presupposes the existence of an organism capable of identifying, translating, interpreting, and making productive use of a particular gene." As she goes on to explain, "All biological functions are composite functions, involving the correct `reading' of many genes. The more complex the function, the more genes are likely to be involved. What genetics can and often does enable us to do is to identify aberrations in some component part that lead to failure of the composite function—always,

however, relative to the other components of that function and almost always relative to particular environmental conditions." Keller warns of the consequences of a too-hasty acceptance of genetic reductionism: "a far more radical depersonalization of medicine than that initiated by the earlier and more general march of medicine from art to science." This step would invite the effacement of not only the patient's environment and history but also the patient him or herself.2° The paths I have explored lead to one destination. Genetic information is special because we are inclined to treat it as mysterious, as having exceptional potency or significance, not because it differs in some fundamental way from all other sorts of information about us. Portions of that mystery and power come from the opaqueness of genetic information, the possibility that others will know things about the individual that he or she does not know, and how genetic information connects individuals to immediate family and more distant kin. The more genetic information is treated as special, the more special treatment will be necessary. Yet none of these factors is unique to genetic information. I propose that genetic exceptionalism—the plea to treat genetic information as different from other healthrelated information—is an overly dramatic view of the significance of genetic information in our lives. It is a reflection of genetic determinism and genetic reductionism at least as much as the product of genuinely distinctive features of genetic information. There is a disturbing corollary to this thesis: there is a vicious circularity in insisting that genetic information is different and must be given special treatment. The more we repeat that genetic information is fundamentally unlike other kinds of medical information, the more support we implicitly provide for genetic determinism, for the notion that genetics exerts special power over our lives. I therefore suggest a revision of the "future diary" metaphor for genetic information. Why not regard our genes as a list of the obstacles we are likely to encounter and perhaps as a somewhat better prediction of how long we will have to do what matters to us, to be with the people we love, and to accomplish the tasks we have set for ourselves? Our genes no more dictate what is significant about our lives than the covers and pages of a blank diary dictate the content of what is written within. Our genes might be regarded metaphorically as the physical, but blank, volume in which we will create our diary. Some volumes have fewer pages in which to write, some more. Certain pages, often toward the back of the volume, may be more difficult to write on. And some leaves may require great skill and effort to open at all. But the physical volume is not the content of the diary. The content we must write ourselves.

Murray: Genetic Exceptionalism and "Future Diaries"

Notes 1. Task Force Report: Genetic Information and Insurance

2. 3.

4.

5. 6. 7.

8. 9.

(Bethesda, Md.: Genetic Information and Health Insurance, National Institutes of Health, National Center for Human Genome Research, 1993). Lawrence 0. Gostin, "Genetic Privacy," Journal of Law, Medicine and Ethics 23 (1995): 320, 321. George J. Annas, Leonard H. Glantz, and Patricia A. Roche, "Drafting the Genetic Privacy Act: Science, Policy, and Practical Considerations,"Journal of Law, Medicine and Ethics 23 (1995): 360, 365. George J. Annas, "Privacy Rules for DNA Databanks: Protecting Coded 'Future Diaries,'" Journal of the American Medical Association 270 (1993): 2346-50; Annas, Glantz, and Roche, "Drafting the Genetic Privacy Act," quotation on p. 360. Annas, Glantz, and Roche, "Drafting the Genetic Privacy Act." Gostin, "Genetic Privacy," 326. Thomas H. Murray and Norman T. Mendel, "Introduction: The Genome Imperative," Journal of Law, Medicine and Ethics 23 (1995): 309-11. Annas, Glantz, and Roche, "Drafting the Genetic Privacy Act," 360. Thomas H. Murray, "Assessing Genetic Technologies: Two Ethical Issues," International Journal of Technology Assessment in Health Care 10 (1994): 573-82.

10. Karen H. Rothenberg, "Genetic Information and Health Insurance: State Legislative Approaches," Journal of Law, Medicine and Ethics 23 (1995): 312-19. 11. Robert J. Pokorski, "The Potential Role of Genetic Testing in Risk Classification," Report of the Genetic Testing Committee to the Medical Section of the American Council of Life Insurance, Hilton Head, S.C., June 10, 1989. 12. Annas, Glantz, and Roche, "Drafting the Genetic Privacy Act," 326. 13. Gostin, "Genetic Privacy," 324. 14. Thomas H. Murray, "Genetics and the Moral Mission of Health Insurance," Hastings Center Report 22, no. 6 (1992): 12-17. 15. Gostin, "Genetic Privacy," 324. 16. Annas, Glantz, and Roche, "Drafting the Genetic Privacy Act," 365. 17. Dorothy Nelkin and M. Susan Lindee, The DNA Mystique: The Gene as a Cultural Icon (New York: W. H. Freeman, 1995), 198. 18. Eric T. Juengst, "The Ethics of Prediction: Genetic Risk and the Physician-Patient Relationship," Genome Science and Technology i, no. I (1995): 21, 30. 19. Carl E Cranor, "Genetic Causation," in Are Genes Us? The Social Consequences of the New Genetics, Carl E Cranor, ed. (New Brunswick, NJ.: Rutgers University Press, 1994). 20. Evelyn Fox Keller, "Masters Molecules," in ibid., quotations on pp. 90, 91, 97.

3.5 Cases

Case 1

George VI's Cancer In September 1951, a cancer tumour was found in the left lung of George VI, the king of Great Britain.'•z George had been a heavy cigarette smoker for much of his life, and the lung cancer was probably a bad side effect of this.3 His lung and the tumour were removed, and George recovered enough to continue with some of his work as king until his death in February 1952.1,4,5 Elizabeth, his daughter and now the queen of Great Britain, took over some of the other parts of his j ob.1'6 Everyone-the king himself, his family, the government, and people in general-knew at the time that George was not in the best of health.2,5.6 They all also knew some details of his health problems. But George and his immediate family did not learn until much later that he had lung cancer.',4 They were all told by the treating physicians that the operation was done to free a bronchial tube or airway in his left lung;

nothing was said about lung cancer.'•2 Surprisingly, the prime minister (Winston Churchill) had access to the king's medical records. With the help of his own doctor, Churchill was able to figure out from this record the true diagnosis. What he learned was soon confirmed by a member of the king's staff (and this information was in turn passed on to some members of the cabinet by the prime minister)! They kept this information a secret from everyone else. Why did the treating physicians not tell the king or even other members of his family the truth?' Back in the early 1950s, the presence of a cancer tumour probably meant that George had not long to live, even if the main cancer tumour were removed successfully. The doctors knew that George would want to carry on with his royal duties for as long as possible.' They also thought that George would be better able to continue if he believed that he was recovering and would live for more than a short while longer. They exercised their therapeutic privilegechoosing what they thought would be best for the patient (and Britain) while keeping the patient in the dark On the

■m I

II

II

IIILL1

3 Management of Medical Information

traditional model of the doctor—patient relationship, deceiving King George in this way was perfectly proper. On some views, however—for example, that of Kant in "On a Supposed Right to Lie from Benevolent Motives" (see page 106)—this was immoral, no matter what the motive. Why did the prime minister have access to the kin0 medical records? Of course, for most ordinary people in Britain or Canada, these medical records would not usually be available to others except with the patient's consent, as Dave Unger points out in "Breaching Confidentiality"

(see page 128). But almost no aspect of the life of members of the British royal family is ordinary. By tradition and law, almost everything they do and everything that happens to them—induding matters of life and death and marriage and birth that would for ordinary people be personal and private—is under the control of others.2.539 The death of a king, for example, is a matter that will affect the public interest.10 If George might die soon, then the government needed to prepare for this and for a new person (Elizabeth, the king's eldest child) to take over these royal duties.

Notes 1. 2. 3. 4. 5.

Philip Ziegler, George VI The Dutiful King (London: Penguin Allen Lane, 2014), ch 6. Sarah Bradford, George VI (New York Viking, 2011), ch 13. I.D. Canacher, "The King's Anaesthetic," Journal of Medical Biography 23 (2015), 139-45. Kenneth Weisbrode, Churchill and the King (New York Penguin Books, 2015), p. 180. Kate Samuelson, "Everything to Know about the Death of King George VI, as Seen on The Crown," Time 9 Nov. 2016.

Case 2

Arndt v Smith In early May 1986 Carole Arndt became infected with chicken pox in the 12th or 13th week of her pregnancy. t•2.3 She talked to her doctor, Dr Margaret Smith, by telephone about her pregnancy and the chicken pox infection in mid-May. Once she recovered from the illness she saw her doctor in person. This happened in late May 1986.' Dr Smith told Ms Arndt that her illness might have affected the fetus she was carrying. As Dr Smith admitted in testimony, she did not tell Ms Arndt all she knew about the effects a chicken pox infection could have on the development of a fetus.' Dr Smith did tell her about the more frequent risk of skin scars and limb defects but not about rarer but still possible and more damaging risks of eye and central nervous system abnormalities.' Ms Arndt continued the pregnancy to term. Unfortunately, the child, a daughter named Miranda, was born with medical problems caused by her mother's chicken pox infection. She required continuous tube-feeding because she was unable to swallow and she needed many operations. Ms Arndt and her partner sued Dr Smith for the extra costs of raising a child with these issues. Her lawyers argued that Dr Smith should have told her about all or at least more of the harms she knew could result from a maternal chicken pox infection. If Ms Arndt had

6. Rosie Waites, 'The Moment a Princess Became a Queen," BBC News Magazine 6 February 2012. 7. The presentation of these events in the Netflix series "The Crown," Episode 1, closely follows the account in Sarah Bradford, George VI. 8. Michael Billig, Talking of the Royal Family (London and New York Routledge, 1998), ch 4. 9. Vernon Bogdanor, The Monarchy and the Constitution (New York Oxford University Press, 1995). 10. Emily Jackson, Medical Law Text Cases and Materials, 3rd edn (Oxford: Oxford University Press, 2013), ch 7.

been given the information, they said, she would have terminated the pregnancy and there would not have been these problems.14 This failure on the part of the doctor was negligent and responsible for the existence of a child with medical issues and the extra costs and other care burdens Ms Arndt and her partner had as a consequence. This case was first heard in the Supreme Court of British Columbia in front of Mr Justice Hutchinson in March 1994 when Miranda was seven years old.' He agreed with Ms Arndt's lawyers that Dr Smith failed to tell Ms Arndt as much about the harms her chicken pox infection might cause to her unborn child as she should have. She failed to fulfill her responsibilities as Ms Arendt's physician.' However, Justice Hutchinson did not agree that the doctoit failure was to blame for the birth of a child with Miranda's health care issues.'•5 He noted that Ms Arndt's pregnancy was carefully planned and that a child was much wanted.'•2 He also noted that Ms Arndt had expressed strong doubts about the validity of the kind of conventional medical and health care knowledge that Dr Smith would have relied on to tell her about the risks to a fetus from maternal chicken pox. Given these two items, the judge doubted that Ms Arndt would have chosen to terminate her pregnancy even if she had been given all the information she should have been given.' So Dr Smith's withholding of information about the risks from Ms Arndt was not responsible for

Cases

the birth of a child with health care problems, the costs of the medical care, or the extra care burdens. The case was next heard in the Court of Appeal for British Columbia in March 1995.2 The appeal was made by the lawyers for Ms Arndt. Mr Justice Lambert wrote the judgment. He also said the issue was not whether Dr Smith had failed to tell Ms Arndt as much as she should have but whether the failure was to blame for the harm that followed.2.5 Justice Lambert's view was that Justice Hutchinson had applied the wrong rule or standard to settle this issue of responsibility." The standard Justice Hutchinson had used was what legal writings call the "subjective test" for causation or blame.° Justice Hutchinson had asked what Ms Arndt would have done if she in particular—not someone else with other beliefs or values or powers of reasoning—had been given all the information her doctor should have provided her about the risks to her fetus from the chicken pox infection. He had concluded she would not have had an abortion. So Dr Smith was not responsible for the harms that followed. According to Justice Lambert, however, the proper test or standard to use to prove that the doctor's failure caused or was to blame for the harm is the

"modified objective" one.6 This was a test devised in an earlier landmark case, Reibl v Hughes.t•2.3.6 On this standard or rule, what should have been found out was what a reasonable person in Ms Arndt's circumstances with a reasonable set of concerns and beliefs would have done if the doctor had provided all the necessary information.' Because in the court's view the wrong standard or test was used, it ordered a new trial. The matter was again heard at court in 1997, this time in the Supreme Court of Canada.3 The lawyers for Dr Smith had appealed the decision of the Court of Appeal for British Columbia; they did not want the matter to be retried.3 The majority opinion was given by Chief Justice Lamer. It confirmed the Court of Appeal's view that the issue was not whether Dr Smith had provided all the information she should have to her patient but whether the doctor's failure to provide information she should have was to blame for the existence of a child with the issues Miranda presented to her parents. Furthermore, the majority agreed that the correct test to apply to answer this question is the modified objective test.3•7 While there were dissenting opinions, the majority opinion was that the case should be retried.

Notes 1. 2. 3. 4.

5.

Arndt v Smith, British Columbia Law Reports vol. 93 2nd series (1994), p 220. Arndt v Smith, British Columbia Law Reports vol. 6 3rd series (1995), p 201. Arndt v Smith, Supreme Court Reports vol. 2 (26 Jun 1997), p 539. At the time of Ms Arendt's pregnancy, Canadian abortion law would have required her to get the approval of a therapeutic abortion committee to terminate her 6. pregnancy. As Mr Justice Hutchinson in the first hearing of the case at the Supreme Court of British Columbia pointed 7. out, this committee might have declined to allow Ms Arndt to have an abortion. Chapter 2 (page 45) considers what a doctor or other medical professional should tell a patient or the patient's substitute decision-maker in order to provide informed consent. A doctor may fail to tell her patient everything she should but not always be responsible for the bad effects

Case 3

The Crash of Germanwings Flight 9525 On 24 March 2015, Germanwings Flight 9525, on its way from Barcelona to Dusseldorf, crashed into a ravine in the French Alps. All 144 passengers and six crew members were killed. t.2.3 In the following days and

that follow. The two issues are theoretically separate. On the other hand, as L.E. and F.A. Rosovsky point out in "The Nuts and Bolts of Obtaining Consent to Treatment" in Chapter 2 (page 45), physicians are sometimes concerned about the first—providing all the information to their patients that they should—because they wish to avoid being blamed and sued for any harms or damage that may follow if they fail to provide all the information they should. Gerald Robertson, "Informed Consent Ten Years Later: The Impact of Reibl v Hughes," Canadian Bar Review 70 (1991), 423. Karen Capen, "Supreme Court Reaffirms Landmark Informed-Consent Ruling in Chickenpox Case," Canadian Medical Association Journal 157 (1 September 1997), 553-4. This comment both separates out and runs together the issue of failing to disclose information that should have been provided by the treating medical professional and the issue of responsibility for harms or bad effects that follow. See note 5 above.

weeks, a picture of the flight's final moments emerged. Soon after the plane reached cruising altitude, the first officer and co-pilot, Andreas Lubitz,4 took control of the plane, locked the captain out of the cockpit, and deliberately flew it into the side of a mountain. Lubitz had a long history of mental health issues 2.5 After two months of pilot school in Germany in late 2008, for example, he withdrew with depression because the

11 111

1

II 1,11

II

11141

3 Management of Medical Information

demands made on him by his training were too great. He had 10 months of psychotherapy and antidepressant medical treatments before returning to pilot school and completing his training program. In December 2014, he started seeing psychologists and other medical professionals again. He complained of stress due to uncertainties about his job at Germanwings. He could not sleep, and his eyesight was deteriorating. An MRI scan showed no physical damage that would explain the problems with his eyes; they may have been psychosomatic—a real disability but one caused by his psychological issues. He was prescribed antidepressants, and at least two of the doctors Lubitz saw wrote letters to him stating that he was unfit to continue work as a pilot. Lubitz ignored these letters and went back to work.5 Could this disaster have been prevented? If the pilot had not been locked out of the cabin by Lubitz, then Lubitz would not have had control of the plane or at least would not have been able to deliberately crash it. But the pilot was locked out by a mechanism designed to stop terrorists or others from entering the cockpit and taking control of the plane.4 Would knowing about Lubitz's depression have allowed his employer or aviation

authorities to predict that he would do what he did and stop him? As Donald Bersoff pointed out in a presidential address to the American Psychological Association, psychologists are expert at diagnosing mental illnesses, but they are not expert at predicting whether someone with a particular illness will do the kind of thing Lubitz did."'8 Of course, doctors treating him did say in letters to Lubitz that he was not fit to return to work as a pilot. However, under Germany's medical confidentiality laws, neither his employer nor aviation authorities were allowed to see these letters without Lubitz's express permission.2'9 Medical confidentiality laws are not so strict in other countries such as Canada. As Dave Unger in "Breaching Confidentiality" (see page 128) explains, Canadian rules allow a medical professional to break patient confidentiality to avoid harm to innocent others, like the passengers on an airplane. But according to philosopher Kenneth Kipnis in "A Defense of Unqualified Medical Confidentiality" (see page 117), the requirement in German law of absolute confidentiality is just as things should be. Even in a case where a disaster like this one results, there should be no breaking of confidentiality except with a patient's clearly stated consent.

Notes 1. 2.

3.

4. 5.

"What Happened on the Germanwings Flight," New York Times 27 March 2015. Jeff Wise, Fatal Descent: Andreas Lubitz and the Crash of Germanwings Flight 9525 (Audible Studios on Brilliance Audio, Feb. 2017). BEA, "Final Report [Germanwings Flight 4U9525]" (March 2016). James Fallows, "Could the Germanwings Crash Have Been Avoided?" The Atlantic 26 March 2015. SPEIGEL Staff, "The Death Wish of a Germanwings Copilot," SPIEGEL Online 27 March 2015.

Case 4

Personal Genome Mapping In 2013, whole genome sequencing—the mapping of a person's entire DNA or genetic blueprint—started to be available in the United States to private individuals."2'3 A company offering this service might require the signature of a physician, but in other ways there wasn't and still isn't anything to stop anyone with enough money from having their complete genetic information analyzed. Websites for companies like GenomeMe in Richmond, BC, claim to offer this service in Canada.4 As well, anyone can have their personal genome completely sequenced as part of a research project like the

6.

7. 8.

9.

David Bersoff, "Protecting Victims of Violent Patients While Protecting Confidentiality," American Psychologist July-August 2014. Suzanne O'Sullivan, "When the Body Speaks," Psychology Today January-February 2017. Denise Grady, "Signs May Be Evident in Hindsight, but Predicting Violent Behavior Is Tough," New York Times 18 September 2013. Editorial Board, "Editorial: Germanwings Crash an Example of When Medical Privacy and Safety Collide" Chicago Tribune 18 March 2016.

Personal Genome Project Canada at the University of Toronto's McLaughlin Centre and the Hospital for Sick Children in Toronto.5 The DNA tests offered by companies like 23andMe and Ancestry.ca in Canada and the US and elsewhere in the world map less than 1 million of the 3 billion base pairs in a person's entire genome.° These are the services considered by Tim Caulfield in "Direct-to-Consumer Genetics and Health Policy" in Chapter 8 (see page 434). In August 2015, Samantha Schilit—a researcher working on human genetics at Harvard University— took advantage of this availability and dropping prices for whole genome mapping to have hers done. 1.2j Who studying in this area would not be curious about their

Cases own genetic makeup and not want to satisfy the curios-

ity if they had the means to do so? But when Samantha contacted her sister, Arielle Schilit Nitenson, to tell her about this project, she found that her sister wasn't as enthusiastic.1,2'7 Arielle was familiar with the issues concerning the confidentiality of genetic medical information from working as a researcher at the US National Human Genome Research Institute.'•z She pointed out that the map of her genome would become part of Samantha's medical record. Anyone with access to this record would have access to her whole genome sequence and whatever it revealed about her risk of inherited health problems. This access might include, for instance, investigators in any health care study she agreed to be a subject in. It would also include insurance companies. The US Genetic Information Nondiscrimination Act (GINA) forbids discrimination based on genetic information by employers and health insurance providers.'•2 There is nothing, however, that prevents life, disability, or longterm care insurers from using this information when it is available to set premiums she would have to pay or even to decide not to offer her insurance. (Canada currently has no law similar to GINA.8) Despite her sisters doubts, Samantha had her genome sequenced.

What neither sister gave very much thought to until later is whether Arielle might have had veto power over Samantha's satisfying her curiosity about her genome.1'2'9 Arielle is not just Samantha's sister but her identical twin. Identical or monozygotic twins come from the same fertilized egg and are nearly genetically identical. In other words, the whole genome sequence for Samantha would also be a whole genome sequence for Arielle. Does that mean Arielle's consent was needed for Samantha to have her genome sequenced?'•z•9 According to Laura Herscher, an instructor and writer in the ethics of genetic counselling, no one has to get the permission of any other family member to have their genetic information analyzed.' But Herscher may be describing what the law says rather than what ethical considerations require. As it turned out, the genome map for Samantha showed something that helped Arielle when Arielle became pregnant with her first child in 2016.' The DNA analysis showed that Samantha—and therefore Arielle—had a mutation that could cause problems with pain medications and delivery methods commonly used during labour. Knowing this, Arielle was able to make arrangements with her obstetrician that avoided these medications and techniques.

Notes 1. 2.

3. 4.

5.

Emily Mullin, "Do Your Family Members Have a Right to Your 6. Genetic Code?" MIT Technology Review 22 November 2016. Samantha Schilit and Arielle Schilit Nitenson, "My 7. Identical Twin Sequenced Our Genome," Journal of Genetic Counselling Online First Articles 2016. 8. Randy Shore, "Personal Genomics: The Test for Everything," Vancouver Sun 27 June 2014. Pamela Fayerman, "BC Gene Sequencing Laboratory Retracts Accreditation Claim after News Story Exposes Bogus Info," Vancouver Sun 25 August 2015. 9. Carolyn Abraham, "Would You Make Your DNA and Health Data Public If It May Help Cure Disease?" The Globe and Mail 7 December 2012.

Elizabeth Murray, "Inside 23andMe Founder Anne Wojcicki's $99 DNA Revolution," FastCompany 14 October 2013. Britt Wray and David Oppenheim, Genomics (National Film Board Canada Ontario Studio, in development, 2017). Kerry Gold, "How Genetic Testing Can Be Used against You—and How Bill S-201 Could Change That," The Globe and Mail 3 April 2016; Yvonne Bombard, Ronald Cohn, and Stephen Scherer, "Why We Need a Law to Prevent Genetic Discrimination," The Globe and Mail 19 September 2016. Anneke Lucassen and Angus Clark, "Should Families Own Genetic Information?" British Medical Journal 335 (7 Jul 2007), 22-3.

3.6 Study Questions 1. Is there any morally important difference between a medical professional lying to a patient and withholding information from the patient about his or her condition or treatment options? 2. Are there ever any circumstances when it would be right for a doctor to offer a patient a placebo as treatment? What if there were no better treatment options? 3. Patients are usually told, or at least believe, that their health care information is private and that hospitals and medical professionals will keep it confidential. Why? Wouldn't it be better for everyone if this information were not kept confidential in any way?

II

al

111111

3 Management of Medical Information

4. Should doctors and other medical professionals always keep information they acquire about their patients completely and absolutely confidential—that is, never, ever break confidentiality? 5. The US Genetic Information Nondiscrimination Act forbids discrimination based on genetic information by employers and health insurance providers. Should Canada have a similar law? 6. If medical records are anonymized—that is, all personally identifying information is removed or disguised—will that always protect the confidentiality of these records?

3.7 Suggested Further Reading Beyene, Yewoubdar. "Medical Disclosure and Refugees: Telling Bad news to Ethiopian Patients." The Western Journal of Medicine 157 (1992): 328-33. Beyleveld, Deryck. "Conceptualizing Privacy in Relation to Medical Research Values," in Sheila McLean, ed., First Do No Harm: Law, Ethics and Healthcare (London and New York: Routledge, 2016), 151-64. Fry, Sara T. "Confidentiality in Health Care: A Decrepit Concept?" Nursing Economics 2 (1984): 413-18. Higgs, Roger, "Truth at the Last: A Case of Obstructed Death?" Journal of Medical Ethics 8 (1982): 48-50. Kermani, Ebrahim J., and Sanford L. Drob. "Tarasoff Decision: A Decade Later Dilemma Still Faces Psychotherapy" American Journal of Psychotherapy 41 (1987): 271-86. Lipkin, Mack. "On Telling Patients the Truth." Newsweek (6 April 1979). Reprinted in Johnna Fisher, ed., Biomedical Ethics: A Canadian Focus, 2nd edn. Pellegrino, Edmund. "Is Truth Telling to the Patient a Cultural Artifact?" Journal of the American Medical Association 268 (1992): 1734-5. Siegler, Mark "Confidentiality in Medicine: A Decrepit Concept." The New England Journal of Medicine 307 (1982): 1518-21. Reprinted in Johnna Fisher, ed., Biomedical Ethics: A Canadian Focus, 2nd edn. Stanard, Rebecca, and Richard Hazler. "Legal and Ethical Implications of HIV and the Duty to Warn for Counselors: When Does Tarasoff Apply?" Journal of Counseling & Development 73 (1995): 397-400. Takala, Tuija. "The Right to Genetic Ignorance Confirmed." Bioethics 13 (1999): 288-93.

-

cro.lessona0 Ethilcs

4.1 Introduction The emphasis on patient autonomy that has supplanted the paternalistic Hippocratic tradition in health care also requires a rethinking of professional ethics. Whereas not too long ago, health care providers, and physicians in particular, exercised greater, often absolute, authority over medical decisions, matters are different today. We saw in Chapters 1-3 that competent patients now have rights to fully informed consent and to accept or refuse any offered treatment. Those rights are reflected in law and are supported by moral reasoning. Today, health care providers' autonomy and responsibilities, and their relationships with patients and each other, must acknowledge and be amended to fit these circumstances. Professional ethics is broadly encompassed by these issues, since it is about the responsibilities and standards of conduct that professionals are expected to live up to. However, there is substantial controversy about the nature and boundaries of professional ethics in health care. Such controversy is undoubtedly due partly to surviving paternalistic Hippocratic influences and biases that remain to be worked out. But debates about professional ethical standards also reflect the intimate connection between professional ethics and a diverse set of genuinely unsettled issues and problems in health care. The articles in this chapter survey much of this ground. They demonstrate the challenges in the way of stating a clear account of professional ethics.

Professional Autonomy and Professional Ethics Because of the acceptance of respect for patient autonomy, it is natural now to ask what the boundaries are to the autonomy of health care providers, usually referred to as professional autonomy. An answer to this question will be shaped by a conception of health care providers' professional responsibilities, or their professional ethics generally. This connection between professional autonomy and professional responsibility can be demonstrated simply enough. As a rough approximation, professional autonomy can be described as "a right to say no" when asked to do things that are in conflict with professional responsibilities. Thus, to have an account of professional autonomy requires that we understand the professional responsibilities of health care providers. Those responsibilities will include their duties to respect patient autonomy, alongside an account of other duties to patients, other health care providers, and their profession. Taken together, these responsibilities and duties reflect the main elements of a general account of professional ethics for health care providers.

nni

II

11

Al

11110

4 Professional Ethics

professional autonomy roughly, health care providers' right to say "no" when asked to act in ways that conflict with their professional obligations.

Making sense of professional autonomy within an account of professional responsibility that recognizes patient rights to autonomy is the aim of the first article in this chapter, Edmund Pellegrino's classic "Patient and Physician Autonomy: Conflicting Rights and Obligations in the Physician—Patient Relationship" (see page 165). Pellegrino argues that despite the frequent appearance of conflict, autonomy and beneficence are actually complementary principles in health care. He argues that physicians need to be respectful of patients' own conception of their health and well-being, noting that "some of the most vital" patient interests are reflected in their choices and their own conception of their "life story" Thus, patient autonomy cannot be separated from patient well-being or beneficence. But Pellegrino goes on to say that patient autonomy cannot override physicians' rights to avoid engaging in activities they honestly believe are not medically required, are harmful, or are "morally repugnant." Recognizing physician autonomy in these circumstances seems plausible because it preserves a sphere of physician discretion and control over their practice of health care. To put it in Kantian terms, respect for professional autonomy prevents health care providers from being used as mere means by patients or society in ways that violate the health care goals of their profession. But there are problems, and it is not clear that the attempt to reconcile autonomy and beneficence is entirely successful or that it provides an adequate direction on other matters related to professional responsibilities and duties. It will be useful to think about these potential limitations as you read the Pellegrino article. Consider, in particular, that if we respect a patient's role in determining what is in their best interests, it will often be unclear when physicians are entitled to have the final say in determining whether a treatment is not medically indicated or is too risky to be undertaken. Whose views matter most in these circumstances? At what point does the threat of harm from a requested treatment entitle a physician to say "no"? Two points are particularly relevant. The first is that health care providers have to acknowledge that, although they are typically in the best position to judge the positive and negative effects of a treatment, they can also be mistaken. They may misjudge the medical efficacy of a treatment, its psychological benefits, and the extent of risk for a particular patient. A second point is that health care providers should acknowledge that patients are typically in the best position to judge whether the anticipated positive benefits of a treatment are worth accepting the risk of harm from treatment, though patients too can make mistakes and regret decisions. This leaves a large area of uncertainty over when physicians and other health care providers can legitimately say "no" when they judge that a patient's requested treatment is not medically indicated or is harmful. This is an unsatisfying position. The uncertainty over when to say "no" can present a "back door" to paternalism and the old Hippocratic approach. Or it may lead to physicians feeling obligated to support or carry out treatments that are inappropriate. Pellegrino's discussion also does not address several key issues with respect to professional responsibility. Can health care providers say "no" to patients' requests for access to scarce resources so that these resources can be used in cases where they have a better chance of producing medical benefits? What obligations do health care providers have to say "no" to patients who may pose threats of harm to others, including other patients or health care providers? These are issues that must be addressed in developing a conception of professional responsibility and professional autonomy. They are addressed in some of the remaining articles in this chapter.

Futility As one practical application of professional autonomy, Pellegrino talks about the importance of recognizing physicians' right to say "no" to "medically useless or futile treatments." Intuitively, this seems plausible, and health care providers often use the notion of futility

Futility

as a ground for refusing requested treatment. We saw it employed in the Rasouli case in Chapter 1 to justify withdrawal of life support. Futility is also used to withhold CPR or organ transplants or chemotherapy for cancer, to name some other instances. But there is great controversy over whether futility is a useful clinical concept that can be legitimately used to refuse treatment. Robert D. Truog, Allan S. Brett, and Joel Frader's "The Problem with Futility" lays out the problems succinctly (see page 176). Truog et al. point out that the notion of futility is typically laden with, and is used to obscure, assumptions about values—for example, that short-term survival is not worth pursuing. In the Rasouli case, this was extended to arguing that a low quality of life was not worth preserving. Rather than reflecting any notion of futility, the Rasouli case and others like it seem to reflect a conflict of values between what physicians and what patients or their families regard as a life that has value. If so, what is at stake is not a clear question about whether a treatment is futile but what sort of an existence or life a patient could find meaningful and what risks are worth taking to pursue such a life. If we accept Pellegrino's argument that patients are typically in a position to know their own best interests, physicians must take very seriously patients' views about the value of survival or low quality of life and engage them or their families in discussion about them if they feel they are mistaken. The Truog et al. article raises important questions about whether futility is used to circumvent such discussions. The authors of the article also point out that it is rarely true that treatments offer zero chance of success. If so, it is reasonable to ask whether treatments with a very low probability of success can be regarded as futile. What often seems to be happening in these cases is that health care providers are reluctant to treat because it wastes scarce resources or is likely to cause more harm to the patient than benefit. Are such considerations really about futility? Aren't they really about whether patients understand the risks of treatment or whether scarce resources should be used where there is little prospect of medical success? If the answer is "yes," futility is used to override discussion about what is most directly relevant and at stake. Truog et al. suggest that the few cases where a treatment offers no chance whatsoever of producing a benefit for the patient (what they call "physiological futility") represents an "ethically coherent ground" for invoking the concept of futility. However, even this limited defense of futility can be challenged, for patients may still derive psychological benefits from being treated or may rightly believe that health care providers can be wrong in their assessment of physiological futility. If there is psychological benefit with no significant or overriding risks or costs, is that not a reason to treat? Why, for example, should a terminally ill patient's hope be denied in such circumstances? This article by Truog et al. has contributed to a substantial and productive literature on futility Their scepticism about the value of the concept of futility is now widely accepted among biomedical ethicists, though it remains in common use clinically. An interesting question not discussed in the article is whether health care could do without the concept of futility completely.

Gatekeeping "Gatekeeping" refers to the role that health care providers play in determining who has access to health care. Gatekeeping can be formulated in negative or positive terms. Negative gatekeeping refers to obligations health care providers have to prevent patients from receiving treatments or tests. In this situation, health care providers stop patients from "getting through the gate" into the health care system. Positive gatekeeping refers to obligations health care providers have to get patients through the gate and into the system. Obligations of negative and positive gatekeeping are inevitable parts of any health care system. Avoiding unnecessary tests or treatments are accepted examples of negative

gatekeeping the role played by health care providers in determining who has access to health care services.

II 131

162

nurI

4 Professional Ethics

tragedy of the commons when each individual seeks to maximize personal benefit from a scarce resource so that demand overwhelms supply and the resource is no longer available to some or all.

gatekeeping. Since treatments and tests have costs and typically involve some risks, this is usually good medical practice. Getting people who need treatments or tests into the system poses no significant moral problems in principle either. But both can come into conflict with other obligations, particularly where resources are scarce. Controversy arises over what obligations health care providers have to act as gatekeepers to control costs or to allocate scarce resources to patients who may be in a better position to benefit from them. Negative gatekeeping is a part of the general problem of fairly allocating scarce health care resources. This issue is discussed in depth in Chapter 7 in which related issues, such as refusing scarce treatment on the grounds of responsibility for illness or considerations of social merit (or lack thereof), are also considered. We introduce discussion of resource allocation here by focusing on conflicting general duties that allocation decisions can place on health care providers. It is often assumed that health care providers owe undivided loyalty to their own patients. If so, there are no conflicting duties, and faced with a choice they will always decide to provide a scarce resource to their own patient, regardless of whether this is an efficient or fair way to allocate scarce resources. In "Should Physicians Be Gatekeepers of Medical Resources?" Milton Weinstein argues, however, that this approach to health care provider responsibilities will lead to a sort of "tragedy of the commons." Physicians who act only in the best interests of their patients without regard to consequences for the overall health care system would soon "collectively reach the limits of health care resources," compromising everyone access to health care, or what he calls the "medical commons." Weinstein argues that some form of collective action is necessary to limit the power of physicians to act solely in their own patients' interests. He rejects proposals for having physicians do cost—benefit analyses at the bedside to determine whether their patients are entitled to scarce resources and argues that the only practical approach that would preserve relations of trust between patients and their health care providers is to impose some sort of external constraints (such as rules of rationing) on the ability of health care providers to act in the best interests of their patients. The physician would then be free to act in the best interests of the patient but within the constraints of the system. Weinstein's proposal puts a substantial burden on health care providers to develop rules of rationing and to manage the expectations of patients, raising questions about preserving relationships of trust between health care providers and patients and about the stability of the system. An alternative not discussed is to have a more publicly accountable system of rationing to establish constraints. This could involve physicians, elected officials, civil servants, hospital administrators, and patients' and citizens' groups working together in a transparent way to solve these problems. This is arguably how a public system of health care like Canada's could operate. It suggests an advantage of public systems of health care over the mainly private American system that Weinstein's proposal is largely directed toward. A largely public system of health care can involve all stakeholders in an open, accountable process for developing a gatekeeping system of rationing scarce resources and managingcompetingclaims on resources. Perhaps a primarily public system of health care would have a greater chance of gaining legitimacy regarding professional obligations and citizens' expectations around rationing and gatekeeping among both health care providers and citizens. But the problem with a public system like Canada's, implicitly noted by Weinstein, is that with relatively low personal insurance costs and little or nothing in the way of user fees, the cost of medical care is not a factor for patients, which may encourage excessive use of health care at the expense of the medical commons. This suggests that carefully written rules of rationing would be necessary or the burden would be back on health care providers to make rationing decisions on a case-by-case basis. But even carefully written rationing rules may be at odds at times with physicians' fiduciary obligations to do what is best for their patients. Such conflicts may be more acute within a system that includes no disincentives to patient access. As well, politicians who make short-term calculations about what policies give

Gatekeeping 163

them the best chance to be elected, or re-elected, may also be tempted to support or ignore loose enforcement of rationing. A possible lesson to be learned from reflecting on Weinstein's discussion is that it is unrealistic and unfair to expect health care providers, either individually or collectively, to bear the main burden of gatekeeping or rationing society's scarce health care resources. Another lesson is that it is one thing to establish publicly agreed upon expectations and standards around gatekeeping but it is crucial that the standards themselves be stable and reasonably self-sustaining. Escalating costs of health care, especially cost overruns, suggest that Canada has not yet found a way to achieve stable and self-sustaining rationing of scarce health care resources, despite having a largely publicly run and accountable health care system. A potential solution that is suggested in Weinstein's discussion would be to have Canadian patients bear more of the direct cost of their care through deductible payments or user fees, though these are widely prohibited now: Another proposal would be to offer positive incentives for prudent use of health care resources—for example, through rebates on insurance fees.

Harm to Others Patients can request treatments or services that can put other patients or health care providers at risk of harm. Such risks can include verbally abusive or violent behaviour or risks of transmission of communicable diseases. Any admission to hospital puts one at a more than everyday risk, and in consenting to the admission the patient accepts that. Even so, it is plausible that patients in hospitals accept only minimal levels of risk. This might be, roughly, the level of risk a reasonable patient would expect other patients to accept in order to permit that patient to receive treatment and care. This would include the stipulation that measures have been taken to minimize risks to patients from other patients. This is one clear professional obligation of health care providers. But it is less clear what the obligations of health providers are to provide care to patients who present more than minimal levels of risk to health care providers. What levels of risk from patients are health care providers required to assume? When are they at the limits of their professional obligations to treat and care for others? It is reasonable to believe that health care providers accept greater levels of risk than do patients. After all, they have consented to become members of professions that involve caring for patients with communicable diseases and for mentally ill patients who may be abusive or violent. But it is not clear what the upper levels of professionally required risktaking are. It is easiest to argue that high levels of risk have been accepted when health care providers have specifically consented to work in high-risk areas, such as emergency rooms and mental health wards. But even when more than minimal risks are accepted by direct consent, it is reasonable to think that health care providers will have conflicting obligations, say, to family members—and of course they have conflicting considerations of self-interest. One attempt to address these issues is given in the Canadian Medical Association's policy statement "Caring in a Crisis: The Ethical Obligations of Physicians and Society during a Pandemic" (see page 188). The CMA policy, written in the aftermath of the 2003 SARS epidemic in Toronto, recognizes a qualified obligation to provide care in a pandemic. It refers to its Code of Ethics as stating that physicians should "provide whatever appropriate assistance you can to any person with an urgent need for medical care." But the policy also recognizes the role of a principle of consent in assessing what risks physicians can be expected to undertake. It says "the concept of personal autonomy allows physicians some discretion in determining where and how they will practice medicine," presumably by choosing specialties and practices that expose them to lower levels of risk. Arguably, this means that health care providers cannot be straightforwardly compared to public safety

II

164

al

1111111

4 Professional Ethics

workers like police and firefighters, who have less discretion about accepting risk by virtue of their choice of profession. So there is a conflict that reflects an uncertainty about what level of risk health care providers are obligated to assume. The CMA policy does not resolve that uncertainty. Instead, it discusses the "reciprocal obligation" society has "to support and compensate" physician and health care workers who "will be expected to put themselves directly in harm's way, and to bear a disproportionate burden of the personal hardships associated with the pandemic." This societal obligation includes making sure that health care workers have adequate resources for responding to the circumstances and that compensation is available for health care workers who are injured, including for their families in the case of death. The CMA policy is instructive for the way that it recognizes the difficulty of assessing the limits of health care providers' obligations to assume risks associated with their profession. It is also useful in clarifying that society must support those who take on such risks if it expects them to assume them. How strong the professional obligation is to accept those risks, even when those supports are clearly in place, remains unclear.

Conscientious Objection

conscientious objection/ refusal when health care providers refuse on moral grounds to provide a treatment that is permitted by the standards of their profession.

Conscience-based objections to providing requested treatment are an accepted, albeit controversial, feature of the provision of health care. Professional governing bodies today recognize that health care providers may "conscientiously object" when patients request treatments that are permitted by the standards of their profession but conflict with a health care provider's moral values—for example, when a health care provider receives a request for physician-assisted dying or an abortion. Currently, the main ethical controversy arises over what should be done to provide care for patients in such circumstances. Does the conscientiously objecting or refusing health care provider have a duty to refer the patient to a health care provider who is willing to provide the care? The ethical argument against this is that if conscientious objection or refusal is accepted, it must also be wrong to require objecting health care providers to make such referrals because this makes them complicit in a practice they find morally objectionable. But, equally, it seems inappropriate to leave patients to their own devices, since this can add stress, create a sense of abandonment, and delay treatment. Also, patients are sometimes not able to act on their own to seek help from other health care providers. Issues around conscientious objection by health care providers and institutions are likely to emerge prominently in Canada in the coming years in light of the Supreme Court of Canada decision in 2015 recognizing a right for some competent adults to physicianassisted dying (PAD). The ensuing federal legislation codifying this right recognizes that health care providers can refuse to participate in PAD, but it does not say anything about a requirement to refer. Health care providers' professional bodies are currently struggling with this issue, although reasonable accommodation of some rights to conscientious objection is universally accepted by all health care professions in Canada. Udo Schuklenk and Ricardo Smalling's "Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies" (see page 191) reviews and rejects classic defences of conscientious objection and the Canadian federal government's, the Supreme Court of Canada's, and health care professional associations' acceptance of some role for conscientious objection in health care. Schuklenk and Smalling believe that it is inappropriate to accommodate physicians' conscientious objections to provide medical services that are uncontroversially legal and that patients have a right to receive. Their argument is multi-faceted, but the core of it rests on a consideration of physicians' voluntary acceptance of the scope of practice of their profession, which is evidenced by their decision to join the medical profession. They note as well the arbitrary nature of which conscientiously held beliefs are or are not thought to be worthy of accommodation

Conscientious Objection

and the potential negative effects on patients from such accommodation. Schuklenk and Smalling's article is provocative and sure to encourage debate. It is worth asking whether they are tolerant enough of pluralism in a liberal democracy and how their views would affect the composition of health care professions.

Professional Ethics Codes It is natural to look to health care professionals' ethics codes to resolve moral problems like the ones discussed above. There is general consensus, however, that professional ethics codes are of little use in these sorts of cases. In "The Functions and Limitations of Professional Codes of Ethics," Dale Beyerstein explains why. In brief, ethics codes do not solve ethical problems or dilemmas because those problems reflect uncertainties about the application of the codes. That is why they are problems! At this point, we must look to deeper moral and often non-moral considerations to solve them. Beyerstein argues that ethics codes nevertheless fulfill other important purposes. They give general guidance to health care professionals; they reflect agreed upon and publicly stated standards of practice and so inform members of the public about expected standards; and they inform members of health care and other professions roughly about what to expect the limitations of cooperation will be when working with each other. This latter point reflects an issue that has not yet been touched upon—namely, that the ethical responsibilities of health care providers include the duty to say "no" when they are asked by other health care providers to do things that are in conflict with the standards of their profession. Beyerstein's article is an apt coda to this chapter. Professional ethics and the duties it imposes on health care providers are filled with problems and uncertainties. It challenges us to think deeply about pressing moral problems. An optimistic view is that over time, many of these challenges will be worked out in policies and practices and incorporated as assumed elements of professional ethics and ethics codes. But even if such optimism is warranted, new problems are bound to arise, or boundaries will be pushed back only to reveal other problems.

4.2 The Nature and Limits of Professional Autonomy and Professional Responsibility

Patient and Physician Autonomy: Conflicting Rights and Obligations in the Physician-Patient Relationship Edmund D. Pellegrino For centuries, physician beneficence went unchallenged as the first principle of medical ethics. To be sure, some physicians had, at times, violated this principle. But no creditable ethical opposition was mounted until a quarter of a century ago when patient autonomy was asserted as a prima facie moral principle

of equal or greater weight than beneficence.' Progressively since then, patient autonomy has become the dominant principle shaping physician—patient relationships. Three serious moral conflicts have emerged as a result: first, beneficence and autonomy have been polarized against each other when they should be complementary; second, the physician's moral claim to autonomy has received little attention; and third, the "autonomy" of medical ethics, itself, has come under a serious threat. This essay will examine each of these three consequences resulting from the rise of patient autonomy. It shall do so from the point of view that the

Source From Journal of Contemporary Health Law & Policy 10, (1994) 1,47-67. Reprinted with permission.

II LE

166

111111

4 Professional Ethics physician—patient relationship is a moral equation with rights and obligations on both sides and that it must be balanced so that physicians and patients act beneficently toward each other while respecting each other's autonomy. Effecting this balance is a morally mandatory and exacting exercise. The compass points that might guide this balancing are to be found in a reflection on the concepts of autonomy and beneficence, the way the content of these abstract notions is provided by the clinical encounter, and the way conflicts may be resolved in particular clinical situations. Out of this reflection, five conclusions will emerge: (1) in concept, autonomy and beneficence are complementary and not contradictory; (2) in theory and in practice, autonomy is a positive as well as a negative principle; (3) the actual content of the concepts of beneficence and autonomy is defined in specific actions and decisions in the light of which conflicts are best understood and resolved; (4) the physician's autonomy as a person and a professional must be factored into the equation; and (5) medical ethics, as an enterprise, must maintain a certain "autonomy" in the face of political and socio-economic pressures. I. The Concept of Autonomy A. Autonomy in General

Autonomy is one of those widely applauded concepts which, on closer inspection, turns out to be difficult to define with precision.2 This is not the place to review the range of construals of the term. Rather, I will limit myself to that construal which centers on the etymology of the word itself, which means "self-rule." What is common to most definitions is the notion that an autonomous person is one who, in his thoughts, words, and actions, is able to follow those norms he chooses as his own without external constraints or coercion by others.3 The history of the concept is complex, and its roots are political as well as moral. Politically, autonomy came into prominence during the Enlightenment as an assertion of the individual's right to be free from tyrannous government—not of law per se, but of unjust law.4 Morally, autonomy encompasses the right of persons to freedom of conscience and to respect as agents capable of making their own judgments in accord with universal moral principles,' or in accord with freely arrived at decisions.6 Autonomy gets its status as a moral right of humans from the fact that human beings have the capacity to make rational judgments about their own lives, choices, and interests. Self-governance deserves respect because it is the way human beings actualize

their powers of choice, and choice is a distinctly human activity. To obstruct the capacity for autonomy is to assault an essential part of a personb humanity because the choices we make are so much an expression of our membership in the human community, of who we are or what we want to be as individual members of that community. Human beings are owed respect for their autonomy because they have an inherent dignity. They do not have dignity because they are autonomous. Human beings who lack or have lost the capacity for autonomous actions are nonetheless humans who retain their inherent dignity. Respect for persons comprises more than respect for autonomy. Autonomy has taken on a distinctive negative connotation. Arising, as it did, as a moral claim against invasion of human rights by tyrannous government, it has come to mean a right of self-determination against those who would usurp that right. In medical ethics, it is conceived largely as a moral and legal defense against physician paternalism and against those who would impose their values—social, moral, or otherwise—on others. But autonomy is also a positive concept. It implies an obligation to foster the human capacity for self-determination, to enhance it, and to remove the obstacles to its full operation. This is especially important in clinical medical ethics where pathophysiological, emotional, and social realities complicate the actualization of patient autonomy. If taken as a strictly negative concept of non-interference, autonomy can be selfdefeating for patients and self-serving for physicians. This positive aspect of autonomy will become clearer as I fill in the content of the concept as it operates in the clinical situation. B. Patient Autonomy and Physician Beneficence

Twenty-five years ago, the political and moral notion of autonomy was appropriated as one of the prima facie principles of medical ethics.' There were good reasons for the emergence of patient autonomy at that time. The rights of patients to refuse unwanted treatment had been neglected for entirely too long. In the mid-sixties, these rights could no longer be denied as participatory democracy, better public education, and the civil rights movements became realities. All authority claims came under suspicion. The abuses of professional and bureaucratic power were widely publicized and no longer tolerable. Moreover, the unprecedented powers of medicine made the choice of medical treatments a far more significant matter than it had ever been in the past. The principle of patient autonomy was seen as the patient's protection against usurpation of his right to participate in decisions that affected his life.

Pellegrino: Patient and Physician Autonomy

This amounted to a denial of the long tradition of medical paternalism (or parentalism), which considered the duty of physicians to decide what was best because the patient lacked medical knowledge and might lose hope if he knew the whole truth about his options or prognosis. Since paternalists acted in the name of beneficence, beneficence was equated with paternalism and thereby came to be interpreted as a counter-principle to autonomy. Morally valid and invalid forms of beneficence and autonomy were not distinguished from each other. The dilemmas of medical decision-making soon were reduced to weighing the principles of autonomy and beneficence against each other.8 Medical paternalism and parentalism, however, are not to be equated with beneficence, conceptually or in practice. Paternalism does not account for the patient's preferences or values that are part and parcel of her good or best interests.9 Paternalism makes the medical good of the patient the only good and subverts other goods to that good. Paternalism violates the patient's autonomy in the name of the patient's best interests while ignoring or overriding some of the most vital of those interests. This cannot be a beneficent act because the patient's own choices are so much an expression of his or her own life story or personhood. To violate or ignore the patient's choices is, by definition, a maleficent act, an injury to the patient's humanity. Only when the patient's human capacity to act autonomously is impaired (i.e., when the patient is incompetent) may we resort to paternalism as a beneficent act to override objections to treatment.10 This is the negative aspect of autonomy. Important as it is, it is a distortion of the idea of autonomy to equate it with total independence from the physician or others in making treatment decisions. The cultural bias against dependence or even the semblance of dependence is strong in American life. However, total independence is unrealistic in any walk of life. Human beings live in community and personal association, especially when they are patients. Patients especially need the input of others if their own choices are to be genuine ones. Physicians are needed to provide information and to discuss this information with patients to enable and empower them to use their autonomy wisely. Patients must compare their values with those of others in the context of some community of belief which they accept in whole or in part. Patients cannot identify with their current choices without reference to some structure of values which they formed in the past and which they reaffirm or reject at the moment of choice. This is part of knowing ourselves, and we know ourselves largely in relation to others.

As Dworkin points out, autonomy implies a "capacity to reflect upon one's motivational structure and make changes in that structure."n Without associating with others and drawing on their preferences and values, we lack the opportunity to alter or reaffirm our values because we do not know what alternatives are available and why they might be preferable. To move from the abstract realm of concept to actual decisions, autonomy needs content, and this comes from reflection not only on our own past values but on the values of others at the moment of choice. It is the physician's obligation to enhance, empower, and enrich the patient's capacity to be autonomous. An autonomous choice requires that we fill in, to the extent possible, the action or choice that maximizes realization of the patient's values. Thus, autonomy has a positive as well as a negative aspect. To become a reality, patient autonomy requires cooperation and assistance from the physician. In short, it requires the physician's beneficent attention to make the patient's autonomy an authentic, as well as an independent, reality. C. Physician Autonomy

In all the current discussions about the moral status of patient autonomy, the autonomy of the physician is often neglected. This philosophy has serious defects. The physician—patient relationship is one of mutual obligation—like any truly ethical relationship. The physician as a human being has the same claim to respect for his or her capacity to make personal choices, to follow his or her conscience about what is good medicine and what is morally acceptable as a person. Personal and professional ethics are not fully separable from each other. Therefore, the patient's moral right of autonomy must be balanced with respect for the physician's autonomy. Autonomy cannot be a unilateral moral right for either patients or physicians. Physician autonomy may be considered under three headings: (1) autonomy as a person, which gives moral status to the physician's personal moral values and conscience; (2) autonomy as a physician, which gives moral status to the physician's knowledge and obligation to use it wisely and well; and (3) autonomy as a member of a profession, of a moral community with collective obligations to patients and society. I have written elsewheren of the moral obligations and the autonomy of medicine as a moral community, and will confine myself here only to the first two construals of the physician's autonomy as an individual. The autonomy of the physician as a person has its roots in the same ground as the autonomy of any other person (i.e., the physician's capacity as a person for rational judgment and expression of preference with

II

11111

Mid

4 Professional Ethics

respect to values and choices). The physician, therefore, cannot be expected to lay aside or ignore his deeper personal beliefs, values, or religious commitments. To be sure, patient autonomy requires that the physician not impose his values in his decisions for the patient. But patient autonomy cannot require the physician to sacrifice his personal moral integrity even for what the patient may believe to be a morally good purpose. Respect for the physician's autonomy also derives from the fact that, under normal circumstances, the physician must write the orders that are carried out by others. The physician cannot avoid the fact that she is the focal point through which harm and benefit of a clinical decision will flow in a majority of cases. The physician therefore is a de facto moral accomplice in what happens to her patient. She cannot place responsibility on others for morally indefensible decisions or for cooperation in decisions that violate her conscience. This inescapable fact of the physician—patient relationship places unavoidable obligation on the physician to avoid action she deems harmful to her patient, even if that action is "required" by state regulation, policy, or law." The physician's autonomy as a physician is also grounded in the possession of expert knowledge needed by sick people and society. The power, itself, conferred de facto by the possession of such knowledge, demands that the physician be free to use it according to her best judgment.'4 If the physician is to fulfill the moral requirement to make her knowledge available to those who need it, she must be allowed sufficient discretionary latitude to apply that knowledge as rationally, efficiently, and safely as possible. This is essential if physicians are to fulfill their part of the covenant with society and with individual patients. Physicians enter this covenant from the first day in medical school, when they accept the privileges and the obligations that go with the acquisition of medical knowledge and skill.15 Clearly, this third sense of physician autonomy can never be absolute. If the physician is incompetent, acts in his own self-interest, or acts paternalistically in the sense I defined earlier, he misuses his expertise and violates his covenant with both the patient and society. That covenant is based on trust in the doctor's Oath which commits him to use his knowledge primarily in the service of the sick. The physician's autonomy as a physician is also limited when she mistakes medical expertise and authority for expertise in questions of values. The physician has no standing as an expert in human values and no authority to set the goals or priorities of public policy relative to the allocation or distribution of health care resources. To be sure, the physician's knowledge

provides essential factual data on which rational social policy should be based. But the actual choices of values are not the prerogative of physicians or any other "experts"—politicians, economists, or even ethicists. Social value questions are a matter of concern for the whole of society In this respect, the expert is like any other member of society with no authority over the values of other individual members of that society or over the society as a whole. The arguments of experts may have more cogency but no more authority than those of others. II. Autonomy: Its Content in the Clinical Context

Autonomy in general, and physician and patient autonomy in particular, might conceivably be defined in the abstract in some general way congenial to a large number of people. However, when we begin to give it content in the context of illness, the problems with absolutism and abstraction become evident, as do the implications of autonomy as a solely negative moral right. First of all, no two persons experience illness the same way. No two persons have the same way of expressing their capacity for autonomous choice. Very few patients demand only "the facts." Some will seek a wide variety of opinions before deciding on their own; some will not. Some patients will prefer to exercise their autonomy by giving it up to a surrogate (i.e., someone they trust to make the decision for them, or perhaps even the physician). The majority will want to express their own way of being autonomous by asking not only for facts but also for the doctor's opinion and the opinions of family and friends. Thus, the content of the idea of autonomy, when it is actualized, will vary with the patient's prior values and cultural, personal, and social relationships. These, in turn, condition a patient's response to illness. Sickness forces a confrontation with the self and with the need to adapt to this illness, here and now. Sickness is a test of our values. For each of us, our response to sickness is unique, and thus the way we express our autonomy is also unique.'6 Patients will vary in the degree of dependence or independence they desire depending on their relationship with the physician, on their relationship to their society or community, and on the degree of trust they impute to others. Second, no matter what degree and kind of autonomy a patient chooses, the very fact of illness physiologically or psychologically compromises the actual expression of autonomy to some degree. The sick person is dependent on the physician's knowledge and

Pellegrino: Patient and Physician Autonomy

help; otherwise she would not need or seek medical help. In addition, in varying degrees, she is in pain, anxious, fearful, and vulnerable. Brain function may be temporarily or permanently compromised by fever, shock, medication, age, or dementia. To restore autonomy, physicians must first attend to reversing these physiological and psychological impediments to the optimal exercise of autonomy. In such cases, medical treatment is essential to restore autonomy. This may require temporarily downplaying or overriding the patient's autonomy until normal sensorial states are attained and then enhancing and empowering it as the capacity for self-determination returns. During this transition, beneficence (i.e., acting in the patient's best interest) modulates the physician's move from "weak" paternalism to enhancement of the patient's full autonomy. Third, no matter what degree of autonomy a patient may want or in what way he wants to express it, the patient is vulnerable to deception in the information he receives. The patient is dependent on the physician's disclosure of diagnosis, prognosis, treatment options, side effects, effectiveness, outcomes, etc. Which facts the physician chooses, which she emphasizes, and which she represses are often subtly or frankly conditioned by her judgment of what she thinks is in the patient's best interests. As any clinician knows, she can get almost any decision she wants from most patients. Therefore, even the most conscientious physician must exert great care to avoid manipulating the patient's choices, even for good reasons. The fact that physicians can so easily influence the patient's choice makes the full operation of patient autonomy problematic. For that very reason, it is morally incumbent on the physician to protect patient autonomy as scrupulously as possible and to try to help the patient realize its positive content. This cannot mean, as some erroneously argue, that autonomy in a sick person is a fiction, that to try to enhance it is a sham, and that we should return to the Hippocratic tradition of benign authoritarianism. Such a reversal would be an intolerable suppression of the patient's human right as a rational being to make uncoerced choices. Physicians and others, therefore, have an obligation not to take advantage of the patient's vulnerability. Informed consent is an empty notion or a charade if the information on which it is based is biased in favor of the physician's preferences. None of this means that physicians cannot advise or persuade patients to do what they think is right. Not to do so is a species of moral abandonment. Patients are entitled to know what physicians think is "best," all things considered. Although the extremes of this

spectrum are not difficult to identify, no one can draw precise lines between advice, persuasion, manipulation, and coercion. But the difficulty of drawing a line does not justify a presumption in favor of paternalism. Rather, it increases the physician's obligation in beneficence to protect autonomy by the most scrupulous selfexamination of his own motives in obtaining consent. Much, therefore, still depends on the physician's character and sensitivity and her possession of the virtue of benevolence. The physician's character may turn out to be the last safeguard of the patient's autonomy and well-being. But, ultimately, the physician and patient must decide together what is to be done. Only in this way can patient autonomy become a cooperative and beneficent enterprise, rather than an adversarial one. All of this applies with special force to surrogate decision-making and to advance directives, which become operative when a competent patient loses the capacity to make his own decisions. Here, the patient's wishes are represented by others or by a written document. The surrogate's wishes have the moral status we usually attribute to a competent patient and should be respected as such. However, family and friends can be in a financial or emotional conflict of interest with the welfare of an incompetent patient. They may even wish, consciously or subconsciously, to relieve themselves of the emotional and physical burdens of caring for a chronically ill person. Their representations of what the patient's autonomous decision would have been were he competent are open to serious question. When "autonomy" is expressed in a living will or other advance directive, an assessment must be made of whether the decision executed in the past, when a person was competent, represents what the patient would want now, when the patient is no longer competent. Is this person, now in a persistent vegetative state, the same person who originally made out that living will? Is autonomy, in its full meaning, so absolute that it binds us to decisions the benefits and the import of which the patient could not possibly have anticipated and which, in the actual context of a particular decision, may not be in his present interests? In these circumstances, the patient is in need of a beneficent agent—one who can be trusted to protect him or her from the autonomous decisions of others, even those who might be legally, but not morally, valid surrogates. This agent may have to be the physician, nurse, or other health professional who acts under the principle of beneficence. Regardless of whether the surrogate is a family member, friend, or the health professional, when the patient's autonomy has been transferred to others, it must be held in trust. If that

Hal

1111111

4 Professional Ethics

trust is violated, the surrogate loses her moral status as well as her "autonomy" to make choices for the patient. Not enough attention has been given to the limitations placed on the "Western" notion of autonomy when applied to the care of patients from different cultural backgrounds. Even in the "West," as Surbone points out in the case of Italian culture, patients may not expect or want to make decisions, preferring to leave them to families or the physician.17 Is it a beneficent or maleficent act to insist on or offer autonomous decision-making in these circumstances? Is it morally wrong, or rather a legitimate compromise, to work within the patient's cultural confines? Are we compelled by the fundamental nature of the principle of "respect for persons" to impose our view of autonomy, or may we compromise it in the name of cultural integrity? These nuances in the full expression of patient autonomy in clinical decisions underscore the fact that autonomy cannot function in actuality without beneficence. Beneficence, properly exercised, is the guarantor of autonomy, rather than its enemy. Enhancing autonomy enabling and empowering the patient to make her own choices, and helping the patient to understand the choices before her in terms of her own past values are all acts of beneficence. These acts enhance the positive content of autonomy and are crucial to any comprehensive notion of the patient's welfare. On this view, the health professional holds the patient's autonomy in trust. While it must be protected, it cannot be divorced from beneficence. It requires the physician's involvement, not her disengagement. In short, if the positive content of the concept of autonomy is to be realized, it will require beneficent action on the part of the health professional. I must emphasize this point because the current pressure to assure patient independence is eliciting two morally dangerous responses on the part of health professionals. One response is to emphasize the negative non-interference dimension of autonomy. This negative conception of autonomy reduces the ethics of the physician—patient encounter to procedure rather than substance. On this view, as long as the procedure allows for autonomy all is well. Autonomy is absolutized in principle and practice. This may lead to the second response, namely, that physicians will accede to whatever the patient or valid surrogate wants. This prompts the physician to transfer all responsibility to patients, family, or friends. This occurs with alarming frequency in the care of infants, the elderly, and demented patients, who may be over- or under-treated because their surrogates demand it. Indeed, one of the most important contributing factors to disagreement between family surrogates and

health professionals is the psychological burden family surrogates carry when they must decide whether to discontinue life support measures. Here, the autonomy owed the patient is transferred to the valid surrogate. In such cases, families often feel they are being asked to sentence a loved one to death or, by their decision, are actually participating in the death of the patient. They need reassurance when the medical situation is one which, in the eyes of the physician, is "hopeless." Physicians cannot simply leave the entire burden to the surrogate or even the patient. They must share that burden. Thus, detachment is not a beneficent act. Often, when families or patients ask that "everything" be done, they are seeking reassurance that everything that could be effective or beneficial be done, not that "everything"—irrespective of probabilities of success—be done. They also want to share their responsibility for cessation of life-support with the doctor. The focus of ethical concern may well shift, however, from the substantive to the procedural when irreconcilable conflict about what constitutes beneficence or the patient's best interests occurs. Ill. Challenges to Physician Autonomy

A seriously neglected facet of the growing dominance of patient autonomy is its impact on the physician's autonomy. The physician—patient relationship, like any ethical relationship, is a reciprocal one. In the justifiable concern for patient autonomy, it is easy to forget that the physician is a moral agent as well as the patient. As such, the physician's autonomy, as well as the patient's, is deserving of respect. When the two are in conflict, the patient's wish does not automatically trump the physician's. The physician's autonomy, like the patient's, has its negative and positive construals. It may seem paradoxical to worry about physician autonomy when it is the patient who is vulnerable and the doctor who holds the knowledge and power the patient needs. This fact rightly imposes the heavier moral burden on the physician in the equilibration of the autonomy relationship. He cannot use his claim to autonomy to violate the patient's capacity to make self-governing choices. But the physician is, like the patient, a human being, entitled to respect for his capacity to reason, judge, and make choices that are authentically "his." He cannot impose his values on the patient, just as the patient cannot impose hers on the physician. The physician—patient relationship is a moral equation with reciprocal rights and obligations. Today, that equation is becoming unbalanced as patient autonomy is elevated to the status of a trumping

Pellegrino: Patient and Physician Autonomy

principle, morally as well as legally. For some, this even implies or includes overriding the physician's values, his discretionary latitude in clinical decisions, and, in some cases, even his rights of conscience. As patient autonomy receives more and more legal sanction, the problem of preserving the physician's moral integrity will grow. This danger is accentuated by the deficiency of "conscience clauses," which could provide statutory protection for physicians who refuse to provide or participate in procedures they find repugnant on moral or religious grounds.18 In the United States, these threats to the physician's autonomy and conscience derive from the evolution of autonomy from a negative to a positive right. The rights of patients to make decisions consistent with their own values was first noted as a right to refuse unwanted treatment in 1914.18 In 1976, it was extended to a right of valid surrogates to refuse lifesaving measures over the physician's objections.2° In 1983, the President's Commission extended autonomy to include the right of participation in "Do Not Resuscitate" orders. This meant that physicians should offer treatments (like resuscitation) that patients might want even if they were not judged medically indicated.21 As a result, the pristine right of refusal of unwanted treatment is now becoming, for some, a right to demand treatment—even over the doctor's best medical judgment.22 Elevating patients' demands for specific kinds of care to moral status under the rubric of autonomy poses several challenges to the physician's right to her own moral integrity. First, there is the challenge to the physician's judgment of what is good medicine (i.e., medicine that is rationally sound in diagnosis, prognosis, and therapeutics). For patients to claim a right to any procedures they wish is to challenge a conscientious physician's integrity as a physician. It depreciates his expertise, reduces his discretionary latitude in decision-making, and makes him a technical instrument of another person's wishes. What is more important is that this can pose a risk to the patient's well-being and subvert the healing purpose for which medicine is intended in the first place. What is demanded may not be indicated, effective, or beneficial. Such demands violate the internal morality of medicine as a practice.23 They can redound to the patient's harm by undermining the physician's moral obligation to provide sound advice and sound practice and to avoid medically useless or futile treatments. This threat is especially pressing today in the debate over medical futility and who defines it. Some would do away with the concept entirely because they consider that defining "futility" is not, and cannot be,

an objective determination. They argue that the idea of futility is so freighted with both the patient's and physician's values that it should be abandoned entirely.24 Others would retain the concept only for obvious situations of total brain death, permanent vegetative state, far advanced malignant disease,25 or when a treatment has failed in the last 100 cases.26 Still others would institutionalize the criteria for futility in hospital policies that would bind the physician to compliance.27 Underlying this debate is the challenge to the physician's expertise to determine when a treatment—or all treatment—is useless, ineffective, or not indicated because the healing, caring, or curing ends of medicine can no longer be attained. One of the Hippocratic Authors made it an ethical obligation for the physician and the patient28 to desist from treatment when the limits of medicine's power had been reached. This limitation on the clinician's discretionary latitude in the use of medical knowledge and skills is especially dangerous when dealing with surrogate decisions for incompetent patients. Do patients or surrogates really know what doing "everything possible" means? Must we respect orders for "no tubes" or "extraordinary measures" when these may well be effective and beneficial and might have been desired if the patient were now competent? Did the previously competent patient really intend to foreswear such measures? Is it ever possible in a living will or medical directive to anticipate what one would wish at the moment of actual decisionmaking? Must vigorous, ineffective, burdensome, and futile treatment be continued because the living will or surrogate requires it? These difficulties do not vitiate living wills or surrogate decisions by those with a durable power of attorney for health. They do warrant caution about the content of autonomy and its actualization in particular cases in which there is doubt about what the patient wanted to be done. They alert us to the fact that the physician's discretion can be so restricted by advance directives that the patient's welfare is compromised. Moreover, a mistaken respect for autonomy or the physician's fear of violating autonomy becomes an excuse for moral detachment, which is actually moral abandonment. The physician is accorded discretionary latitude in clinical decisions because medical knowledge must be applied to individual cases. The care of individual cases is not reduced to a set formula but rather must be modulated by a host of clinical and personal factors peculiar to each patient. Without discretionary latitude, the physician cannot personalize and individualize care; she cannot fulfill her obligation to use her knowledge for the patient's best interests. Without

I I AI

MLA

4 Professional Ethics

constraints on discretionary latitude, the physicianb decisions can violate the patient's values or produce physical harm. The balance between too narrow and too wide a definition of discretionary space is a delicate, but increasingly important, one to strike. Another place where physician autonomy is endangered is in the sensitive realm of the physicianb religious beliefs. In the future, the secular trend in our society and the drive for autonomy may converge to place constraints on the physician's religious convictions and values. Current legal literature already reflects instances of subtle, and sometimes not so subtle, coercion of the consciences of nurses and physicians who oppose or refuse to participate in abortion, sterilization, the use of abortifacient, or to carry out directives to withdraw feeding tubes.29 Medical students and residents are under increasing pressure to learn and to participate in abortion training by practice. Applicants to medical schools are now frequently asked about their views on abortion. No solid data are available on whether their answers influence the admission committee's or interviewers decision to accept or reject them. Nevertheless, the question is asked so often that it seems unlikely to be of only passing interest to interviewers. Fortunately, the right of conscientious refusal on grounds of personal beliefs is currently protected.3° However, past statements of official bodies like the American College of Obstetricians and Gynecologists with respect to training in abortion techniques as a condition of residency approval are worrisome, even though superseded at the present time.3' Another possible challenge to physician autonomy presents itself in the current debate about voluntary euthanasia and assisted suicide. It seems very likely in view of the current drift of public and professional opinion, that one or both of these practices will become legal. When this happens, these procedures will also very likely become "benefits" or entitlements in our future health care system. The Clinton Administration is likely to include abortion among "reproductive services" in its proposed Health Security Act. Pressures on physicians are then sure to mount to provide abortions as part of the "benefit package." We are promised that the rights of conscience of those who find abortion morally reprehensible are to be protected. However, in a climate of moral pluralism, self-determination, and consensus ethics, this could change. I cite these examples not to provoke furious debates about the moral status of the procedures in question but because whatever one's position may be, the moral problem of the integrity of the physician's autonomy and moral rights of expression of conscience

cannot be ignored. If legal or societal sanction for a certain procedure becomes widespread, will this warrant violation of the physician's conscience? Some see these as matters of such societal benefit that the physician's private moral and religious beliefs should be dissociated from his professional life.32 This will pose an impossible situation for the morally conscientious Orthodox Jew, Roman Catholic, or Muslim in certain fields of medicine. Another challenge to the physician's moral integrity and autonomy is one encountered by physicians in countries with "managed" health care systems. The political and economic pressures of health care policy and reform already place the physician in a position of moral conflict. The economic and fiscal drive behind such programs can make the physician a moral accomplice in practices he deems injurious to his patientb well-being. No matter what setting he is in—cost containment, rationing, acting as a gatekeeper, an institutional milieu of managed health care or managed competition, a publicly funded clinic adhering to clinical guidelines, etc.—all place the physician in the position of double or triple agency. The physician's professional commitment to advocacy for her patient may put her at odds with an institutions or society's well-being. Exigency, expediency, and economics, not ethics, drive such systems. Does the "autonomy" of the institution or health policy override the autonomy of either, or both, the patient and physician? How are "good" business, economic, political, and fiscal policy and the moral purpose of medicine to be reconciled? Which takes precedence when conflict is unavoidable? Such questions are sure to become more widespread in the future as the zeal for cost containment and managerial ideologies, rather than the welfare of patients, are enshrined in law and public policy. N. Procedural Ethics and Conflict Resolution

The central moral issues in any attempt to balance patient and physician autonomy are substantive. But when substantive moral issues are unresolvable, procedures for ethically dealing with the conflict are necessary. The autonomy of patients, their surrogates, and physicians all carry moral weight and, on that account, command respect. The ethical goal of any procedure aimed at conflict resolution should be to protect each agent's autonomy to the extent possible. To this end, a variety of procedural moves are morally plausible when a conflict in moral or professional values reaches an impasse. To begin with, the patient or patient's valid surrogate can discharge the physician and engage one who will take care of the patient on his or her terms.

Pellegrino: Patient and Physician Autonomy

Alternatively, the physician can withdraw, respectfully and without recrimination, on grounds of preservation of his moral integrity. However, these alternatives are possible only if another physician is willing to undertake the care of the patient. To withdraw without transferring care to another competent physician is morally and legally unacceptable and constitutes abandonment. Hence, the question about how the transfer should be made arises. In both situations, patients or surrogates might claim a right to assistance in selection of another physician specifically congenial to their moral values. In cases that do not involve a fundamental moral principle (e.g., a patient's choice of a "lump" resection and radiation as against radical mastectomy for breast cancer or the use of non-standard, but not harmful, medical treatment), such cooperation could be ethically appropriate. The issue may be more fundamental, however, when it involves voluntary euthanasia, abortion, physician-assisted suicide, or withdrawal of care from patients in a persistent vegetative state. In such instances, some would argue that the physician who withdraws has an obligation to find another physician to undertake the patient's care under the family's or patienth terms. For others, this would constitute an unacceptable degree of moral complicity by cooperation in an act one considers morally untenable.33 The most difficult situation, for which there is no totally satisfactory solution, is when physicians and patients or their surrogates disagree on a serious and fundamental ethical issue and the possibilities of physician withdrawal, or discharge of the physician by patient or surrogate, are foreclosed by external circumstances. There may be no physician willing to undertake care on the patient's or surrogate's terms. No other physician may be available for reasons of geography or urgency of the clinical situation or lack of the required expertise. Or, the physician may be employed in an institutional setting (i.e., prisons, the military, certain managed care plans, or residency training programs) in which physician choice is limited by virtue of his occupying a specific, socially defined role. In these settings, failure to perform the expected role (i.e., participation in state-ordered executions) could result in significant fiscal penalties, discharge from one's job, and legal or disciplinary action. When there are irreconcilable differences in moral commitments and the physician cannot extricate herself by reasons of exigency or limitations imposed by patient or society, the physician must still be faithful to her conscience. This may mean acceptance of the attendant penalties for refusal to comply with institutional, legal, or socially defined goals (e.g., refusal to

participate in state-ordered executions or in coerced interrogation of war prisoners). Just when and how individual physicians should refuse to comply with social conventions is not a matter of precise formulation. No one can enter the mind and heart of another and untangle the moral psychology of a particular moral choice, but this fact does not vitiate judgment about the ethical probity of the act in question. Some of the most complex and difficult situations occur when surrogates are acting for infants whose future values cannot be known.34 In such cases, the conflict will often be between the surrogate's and the physician's estimates of what is "best" for the infant. The vagaries of "quality of life" estimates complicate the issue because of the impossibility of assessing how the infant would make that evaluation when he or she becomes aware it must live with the infirmities the physicians prognosticate. These infirmities are often complex, may result in a life of prolonged disability and discomfort, and constitute grave emotional, physical, and fiscal burdens for parents and society. In a society propelled by economic constraint, reluctance to sacrifice material goods even for disabled children, and an obsession with physical beauty, it is not uncommon for even conscientious parents to decide to withhold or withdraw life-sustaining treatment in order to spare the infant a "life of suffering" or poor quality. But "quality of life" and "value of life" are not synonymous terms. The modern tendency to use them synonymously is a serious point of discord between secular and religious perspectives on ethical decisions. What is right and good in these circumstances is problematic and may be interpreted in contradictory ways by physicians, nurses, parents, and even governmental regulation.35 In these circumstances, we may search for procedures which will protect the autonomy of all participants, but the substantive ethical disagreement remains. When the physician believes the parents' decision would constitute grave harm to the infant, she has several procedural alternatives. One alternative is to withdraw and ask the parents to engage another physician. This would be permissible if the physician did not think that withdrawal would result in grave moral and physical harm to the infant. Where the physician might judge otherwise, whether treatment is withheld or continued, then that is medically futile.36 At such a time, the physician has an obligation to take whatever measures are available to avert harm, such as appeal to an ethics committee or, if necessary, to the courts. A proponent of absolute parental autonomy might justifiably ask what moral claim a physician can have to judge, or even question, a parent's surrogacy rights.

I

174

II

111111

4 Professional Ethics

This point of view assumes that parents have absolute dominion over their children and that their decisions will invariably be benevolent and altruistic; it also ignores the covenantal trust relationship between the physician and the patient. For a variety of reasons— pride, shame, or unwillingness to confront the expense, financial and emotional, of caring for a disabled child—parents may decide to undertreat. On the other hand, they may opt for futile overtreatment out of lack of information, religious conviction, or fear of being in some way responsible for their infant's death. Furthermore, the autonomy imputed to parents cannot be absolute. Physicians and nurses are obliged by virtue of their commitment to the well-being of their patients to act in the interests of the infant. They must, of course, appreciate that decisions surrounding the care of very sick and potentially disabled infants must involve the family. Indeed, in a real sense, the whole family becomes a "patient," whose collective interests must be safeguarded. The implications of the decision on the future lives of the whole family are, therefore, not to be denied; however, these considerations by themselves do not justify withholding or withdrawing treatment that is effective, beneficial, and not disproportionately burdensome. When there is obvious and overt conflict between the good seen by parents and the medical good of the infant, the obligation is greater to the most vulnerable person (in this case, the infant). Situations involving such irreconcilable conflicts of obligation are sometimes unavoidable. Still, we are obliged to do as much as possible to respect the physician's obligations as physician as well as the autonomy of surrogates or patients. Sometimes both cannot be respected without unacceptable compromises, on one side or the other. Before such an impasse is reached, all other methods of conflict resolution should be exhausted. Ethics committees can serve to clarify the issues and perhaps suggest a way in which compromise could effectively be reached in a manner that preserves the moral integrity of all the participants. Appointment of legal guardians and appeals to the courts are far less satisfactory. In any case, all of these devices address only the procedural resolution of the practical conflict. They certainly do not resolve the ethical dilemma of conflicting claims to autonomous decision-making. V. The Integrity of Medical Ethics

Some would suggest that the problem is with medical ethics itself, with the insistence on universal rules of moral conduct on which physicians base their moral claim to autonomy as physicians. Why not change

medical ethics itself? Why not leave it to be negotiated between physician and patient? Perhaps medical ethics should be a changing, socially constructed contract varying from society to society, era to era, and patient to patient. Some argue that medicine and its ethics must be whatever is negotiated politically between the profession and government. Other socially and politically constructed forms of ethical justification are currently popular as well. They imply that there is no such thing as a universally binding medical ethic, only an ethic of political expediency or societal convention. Some of us, however, think this would be disastrous for medicine, the physician, and the patient. The autonomy—that is really to say the moral integrity— of both physicians and patients must somehow be preserved. So, too, must the integrity of the ethics of medicine itself. Medical knowledge is too powerful a tool to become an instrument of governmental or social pressures, or private negotiation, however benign their motives may appear to be. Medicine is also too powerful to go wholly unregulated. There are too many examples of the subversion of the powers of medicine to evil purpose by unjust political regimes to make the ethics of medicine a subject for political negotiation.37 There are too many examples of the way unregulated medical "entrepreneurs" or morally bankrupt physicians can exploit the vulnerability of the sick. Medical ethics must maintain a degree of independence if it is to protect the sick person. It must remain subject to public criticism but not be controlled by social convention. It must also be protected from subversion by the profession itself. This requires a much firmer philosophical grounding for medical ethics than we now possess. The possibility of achieving universal approbation for a commonly held ethic of the profession seems to be receding today in the face of the multicultural, morally pluralistic, and morally relativistic temper of the times. This climate, however, cannot justify abandoning the effort. Nothing less is called for than a reconstruction of the ethics of the relationship between patient and doctor. This will be difficult, indeed because the "remarkable solidarity" and "singular beneficence," which Osler praised,38 are rapidly disappearing in the worldwide questioning of the moral values that have traditionally undergirded medical ethics. I have purposely said little about the principle of justice, which must also be factored into the equation. On the whole, this facet of the physician—patient relationship has been underdeveloped. It is now necessary to establish the conceptual relationships among justice, autonomy, and beneficence, as well as their

Pellegrino: Patient and Physician Autonomy

actualization in the clinical context. Justice has the 3. Physician autonomy is limited by a competent patient's or valid surrogate's moral right to ref use interesting facet of being both a principle and a virtue. proffered treatment. The physician is obliged, The incorporation of justice into the autonomybeneficence equation will require a prior clarification however, to help the patient arrive at an autonomous decision by enhancing or empowering the of how principles and virtues are conceptually and patient's capacity to make authentic, self-governing practically related. choices. Despite the difficulties, the effort to balance the autonomy equation is not futile. Its importance 4. The patient's autonomy is limited when it becomes a demand for treatment the physician honestly beimpels us to the effort to try to find the points of ballieves is not medically indicated, is injurious to the ance. Autonomy and beneficence are two principles patient, or is morally repugnant. so closely tied to the healing ends of medicine that to violate either is to imperil the moral integrity of both 5. The physician's autonomy is limited on questions patients and physicians. Nevertheless, any compreof value, e.g., on questions of the goals or purposes to which medical knowledge may be put for parhensive moral philosophy for the health professions ticular individuals or societies. must encompass more than these two principles. Justice must be included and account taken of both 6. Societies and institutions must establish mechanisms, with only minimal recourse to law, for univirtue and the moral psychological insights of nonlateral discontinuance of the relationship when principle-based theories. In any case, one step in the larger effort is to try either patient or physician feels personal integrity is being compromised. to achieve a better balance between the two most powerful principles shaping physician-patient rela- 7. The first principle of medical ethics is still benefitions today. Several precepts need to be built into the cence. Beneficence is essential if autonomy is to be current re-examination of the foundations of profesauthentically expressed and actualized. sional ethics: In sum, beneficence and autonomy must be mutually 1. Patient autonomy is a moral right of patients, and re-enforcing if the patient's good is to be served, if it is a duty of physicians to respect it. the physician's ability to serve that good is not to be 2. Integrity of conscience and professional judgment compromised, and if the physician's moral claim to are moral rights of physicians. Society and patients autonomy and the integrity of the whole enterprise of have an obligation to respect them. medical ethics are to be respected. Notes 1. Tom L. Beauchamp & James E Childress, Principles of 11. Biomedical Ethics 67-119 (3rd ed. 1989). 12. 2. Gerald Dworkin, The Theory and Practice of Autonomy 3-6,12-20 (1988). 13. 3. This is a paraphrase of the essentials of Dworkin use of the term. Id. at 7-12. 14. 4. See John Locke, Locke's Second Treatise on Civil Government (Lester DeKoster ed., 1978). 5. See Immanuel Kant, Grounding for the Metaphysics of Morals (James W. Ellington trans., 1981). 15. 6. See John Stuart Mill, On Liberty (Elizabeth Rapaport ed., 1978). 7. Beauchamp & Childress, supra note 1. 16. 8. Tom L. Beauchamp & Lawrence B. Mccullough, Medical Ethics: The Moral Responsibilities of Physicians 22-51 17. (1984). 9. See Edmund D. Pellegrino & David C. Thomasma, For the 18. Patient's Good: The Restoration of Beneficence in Health 19. Care 23-25 (1988). 10. James E Childress, Who Should Decide? Paternalism in Health Care 102 (1982). 20.

Dworkin, supra note 2, at 108. Edmund D. Pellegrino, The Medical Profession as a Moral Community, 66 Bull. N.Y. Acad. Med. 221 (1990). Edmund D. Pellegrino, Societal Duty and Moral Complicity: The Physician's Dilemma of Divided Loyalty, 16 Int'l J.L. & Psychiatry 371 (1993). See David M. Mirvis, Physicians' Autonomy-The Relation between Public and Professional Expectations, 328 New Eng. J. Med. 1346 (1993). Edmund D. Pellegrino, The Ethics of Medical Education, in The Encyclopedia of Bioethics (Warren T. Reich ed., 1994). George]. Agich, Reassessing Autonomy in Long-Term Care, Hastings Center Rep., Nov-Dec. 1990, at 12. See Antonella Surbone, Truth Telling to the Patient, 268 JAMA 1661 (1992). See Lynn D. Wardle, Protecting the Rights of Conscience of Health Care Providers, 14]. Legal Med. 177 (1993). Schloendorff v. Society of N.Y. Hosps., 105 N.E. 92,93 (N.Y. 1914). In re Quinlan, 355 A.2d 647 (NJ. 1976).

uu

I

II Jl

UI

1111113

4 Professional Ethics 21. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Deciding to Forego Life-Sustaining Treatment: A Report of the Ethical, Medical, and Legal Issues in Treatment Decisions 241 (1983). 22. In re Wanglie, No. Px-91-283 (D. Minn. July, 1991). 23. For a further discussion of the morality issues in medicine, see John Ladd, The Internal Morality of Medicine: An Essential Dimension of the Patient-Physician Relationship, in The Clinical Encounter: The Moral Fabric of the PatientPhysician Relationship 209 (Earl E Shelp ed., 1983). 24. R.D. Truog et al., The Problem with Futility, 326 New Eng. J. Med 1560 (1992). 25. Stuart J. Youngner, Who Defines Futility? 260 JAMA 2094 (1988). 26. Nancy S. Jecker & Lawrence J. Schneiderman, Medical Futility: The Duty Not to Treat, 2 Cambridge Q. Healthcare Ethics 151 (1993). 27. Lance K. Stell, Stopping Treatment on Grounds of Futility: A Role for Institutional Policy, 11 St. Louis U. Pub. L. Rev. 481 (1992). 28. Hippocrates, The Art, in 2 Hippocrates 185-217 (W.H.S. Jones trans., 1981). 29. Wardle, supra note 18. 30. Executive Board Minutes from the American College of Obstetrics and Gynecology at 12, Item 6.2 (Jan. 1992) (on file with organization).

The Problem with Futility Robert D. Truog, Allan S. Brett, and Joel Frader

"FUTILITY" is one of the newest additions to the lexicon of bioethics. Physicians, ethicists, and members of the media are increasingly concerned about patients and families who insist on receiving life-sustaining treatment that others judge to be futile. A clear understanding of futility has proved to be elusive, however. Many clinicians view futility the way one judge viewed pornography: they may not be able to define it, but they know it when they see it.' The notion of futile medical treatment may go back to the time of Hippocrates, who allegedly advised physicians "to refuse to treat those who are overmastered by their diseases, realizing that in such cases medicine is powerless."' More recently, the concept has appeared frequently in court decisions and policy statements.3-6 The so-called Baby Doe law exempts physicians from providing treatment that would be "virtually futile."7 The Council on Ethical and Judicial Affairs of the American Medical Association (AMA) recently concluded that physicians have no obligation

31. See Barbara L. Lindheim & Maureen A. Cotterill, Training in Induced Abortion by Obstetrics and Gynecology Residency Programs, 10 Fam. Plan. Persp. 24 (1978). 32. This dissociation is one I have encountered already in private conversation with medical students, colleagues, and influential laypeople. 33. Pellegrino, supra note 12. 34. See Arthur E. Kopelman, Dilemmas in the Neonatal Intensive Care Unit, in Ethics and Mental Retardation 243 (Loretta Kopelman & John C. Moskop eds., 1984). 35. Child Abuse Amendments of 1984, Pub. L. No. 98-457, 98 Stat. 1749 (1984) (amending 42 U.S.C. § 5101 (1974)). 36. See Anne Bannon, The Case of the Bloomington Baby, Hum. Life Rev., Fall 1982, at 63; Michael McCarthy, Anencephalic Baby's Right to life? 342 Lancet 919 (1993); John]. Paris et al., Physicians' Refusal of Requested Treatment: The Case of Baby L, 322 New Eng. J. Med. 1012 (1990). 37. See Edmund D. Pellegrino, Societal Duty and Moral Complicity: The Physician's Dilemma of Divided Loyalty, 1 Int'1].L. & Ethics (June 1994). 38. William Osler, Chauvinism in Medicine, in Aequanimitas: With Other Addresses to Medical Students, Nurses and Practitioners of Medicine 267 (1943).

to obtain consent for a do-not-resuscitate (DNR) order when cardiopulmonary resuscitation (CPR) is deemed futile.8 The fact that this concept has appeared in law and policy may seem to indicate that it is clearly understood and widely accepted. In reality, however, the notion of futility hides many deep and serious ambiguities that threaten its legitimacy as a rationale for limiting treatment. Paradigms of Futility

Contemporary discussions of futility have centered primarily on cases involving patients in a persistent vegetative state and those involving the use of CPR A third type of case, involving organ-replacement technology, has received little attention but is helpful to our understanding of futility. Futility and the Persistent Vegetative State

The first type of scenario involving the question of futility is represented by the recent Minnesota case of Helga Wanglie.9 Mrs Wanglie was an 86-year-old woman who had been dependent on mechanical ventilation and in a persistent vegetative state for more than a year. Her husband insisted that she believed in maintaining

Source: From The New England Journal of Medicine, 326 (1992), 1560-64. Reprinted with permission of the Massachusetts Medical Society

Truog/Brett/Frader: The Problem with Futility

life at all cost and that "when she was ready to go . . . the good Lord would call heel° Her physicians, on the other hand, believed that the continued use of mechanical ventilation and intensive care was futile. When attempts to transfer her elsewhere failed, they sought to have a court appoint an independent conservator with responsibility for making medical decisions on her behalf. The judge denied this petition and reaffirmed the authority of her husband as legal surrogate. Three days later, Mrs Wanglie died. Cases like that of Mrs Wanglie seldom reach the courts, but they are probably not rare. A similar case involving a child with severe brain damage was concluded with a settlement favorable to the family before a judicial decision.11 Futility in Cases Involving CPR

The second prototypical scenario involves the use of DNR orders. Although the techniques of CPR were originally intended only for use after acute, reversible cardiac arrests, the current practice is to use CPR in all situations unless there is a direct order to the contrary. Since cardiac arrest is the final event in all terminal illness, everyone is eventually a candidate for this medical procedure. DNR orders were developed to spare patients from aggressive attempts at revival when imminent death is anticipated and inevitable. Nevertheless, patients or families sometimes request CPR even when caregivers believe such attempts would be futile. Some have argued that in these circumstances a physician should be able to enact a DNR order without the consent of the patient or family.12-14 Futility and Organ-Replacement Technology

Although the bioethical debate over the question of futility has been most concerned with cases involving CPR and the treatment of patients in a persistent vegetative state, a third type of futility-related judgment has gone essentially unchallenged. It involves the increasingly large number of interventions that could possibly prolong the life of virtually any dying patient. For example, extracorporeal membrane oxygenation can replace heart and lung function for up to several weeks. Physicians now use this intervention when they expect organ systems eventually to recover or while they await organs for transplantation. However, it could prolong the life of almost anyone with cardiorespiratory failure, reversible or not. Patients thus kept alive may remain conscious and capable of communicating. Caregivers do not now offer this therapy to terminally ill patients, presumably because it would be futile. This judgment has gone largely unchallenged, yet it is not obvious why a clinician's unilateral decision not to use

"futile" extracorporeal membrane oxygenation is inherently different from a decision not to use "futile" CPR or "futile" intensive care. If all three treatments can be characterized as objectively futile, then unilateral decisions not to offer them should be equally justified. As it is used in these three cases, the concept of futility obscures many ambiguities and assumptions. These can be usefully grouped into two categories: problems of value and problems of probability. Futility and Values

It is meaningless simply to say that an intervention is futile; one must always ask, "Futile in relation to what?" The medical literature provides many examples in which the importance of identifying the goals of treatment has not been fully appreciated. The effectiveness of CPR, for example, is often discussed in terms of whether patients who require the procedure can survive long enough to be discharged from the hospital.' This definition of success usually implies that shortterm survival is a goal not worth pursuing. Patients or family members may value the additional hours of life differently, however. Indeed, physicians and other caregivers have repeatedly been shown to be poor judges of patients' preferences with regard to intensive care."-18 Schneiderman and colleagues have argued that treatments that merely preserve permanent unconsciousness or that cannot end dependence on intensive medical care should be considered futile.19 Although society may eventually endorse decisions to override the previously expressed wishes of patients or the desires of surrogates who demand such treatments, it does not follow that the treatments are futile. Mr Wanglie would have rejected this conclusion, and there is no reason to dismiss his view out of hand. The decision that certain goals are not worth pursuing is best seen as involving a conflict of values rather than a question of futility. Certainly in this context, the plurality of values in our society makes agreement on the concept of futility difficult if not impossible. Several groups have therefore attempted to arrive at a value-free understanding of the concept.2(121 The most promising candidate thus far is the notion of "physiologic futility" As the guidelines on the termination of life-sustaining treatment prepared by the Hastings Center state, if a treatment is "clearly futile in achieving its physiological objective and so offer[s] no physiological benefit to the patient, the professional has no obligation to provide it."2° For example, the physiologic objective of mechanical ventilation is to maintain adequate ventilation and oxygenation in the presence of respiratory failure, and the physiologic objective of CPR is to maintain adequate cardiac output and respiration

I .11

1111.111

4 Professional Ethics

in the presence of cardiorespiratory failure. The New York State Task Force on Life and the Law mistakenly concludes that CPR is physiologically futile when it will "be unsuccessful in restoring cardiac and respiratory function or [when] the patient will experience repeated arrest in a short time period before death occurs."21 CPR is physiologically futile only when it is impossible to perform effective cardiac massage and ventilation (such as in the presence of cardiac rupture or severe outflow obstruction). Saying that CPR is physiologically futile when it will be unsuccessful in restoring cardiac function is like saying that mechanical ventilation is physiologically futile if it cannot restore respiratory function. The immediate physiologic effect of the intervention differs from the broader and more uncertain question of prognosis. Physiologic futility, understood in narrow terms, comes close to providing a value-free understanding of futility. Unfortunately, it applies to a very small number of real cases involving CPR. Similarly, since in the case of Mrs Wanglie mechanical ventilation could maintain adequate oxygenation and ventilation, her treatment could not be considered futile in the physiologic sense. Even the use of extracorporeal membrane oxygenation in terminally ill patients cannot be considered physiologically futile, since it can maintain circulation and ventilation. The concept of physiologic futility therefore falls short of providing guidance in most cases resembling those described above. Futility and Statistical Uncertainty

In most medical situations, there is no such thing as never. Futility is almost always a matter of probability. But what statistical cutoff point should be chosen as the threshold for determining futility? The statement from the Council on Ethical and Judicial Affairs of the AMA concludes that physicians have no obligation to provide futile CPR, but it fails to specify any level of statistical certainty at which the judgment is warranted.8 The AMA statement fails to acknowledge that this is even an issue. Should each physician decide independently what probability of success should be considered to indicate futility? Even if we could agree on a statistical cutoff point for determining futility, physicians are often highly unreliable in estimating the likelihood of success of a therapeutic intervention. Psychological research22'23 has shown that estimates of probability are susceptible to "severe and systematic errors."22 Empirical studies have corroborated the limitations of clinical assessment in estimating both prognosis24 and diagnosis.25 Even in theory, statistical inferences about what might happen to groups of patients do not permit accurate predictions of what will happen to the next such

patient. In addition, the tendency to remember cases that are unusual or bizarre predisposes physicians to make decisions on the basis of their experiences with "miraculous" cures or unexpected tragedies. Schneiderman and colleagues recently argued that a treatment should be considered futile when 100 consecutive patients do not respond to it."' But how similar must the patients be? In assessing the efficacy of mechanical ventilation to treat pneumonia, for example, is it sufficient simply to recall the 100 most recent patients who received artificial ventilation for pneumonia? Or must this group be stratified according to age, etiologic organism, or coexisting illness? Clearly, many of these factors will make an important difference. Futility and Resource Allocation

Although medical practice has increasingly emphasized patients' autonomy there is growing pressure on physicians to slow the increase in health care costs by foreclosing some options. Thus, we have a tension between the value of autonomy exercised in the form of consent to use or omit various interventions, and the desirability of a more Spartan approach to the consumption of medical resources. We promote patients' freedom to request whatever the medical menu has to offer, but we also require that interventions be guided by considerations of cost and the likelihood of benefit.26 Unfortunately, there is no consensus about what constitutes a just method of balancing the preferences of individual patients against the diverse needs of society. To some, the concept of futility provides at least a partial solution to this dilemma it offers a reason to limit therapy without the need to define a fair procedure for allocating resources. This approach allows treatments to be denied on the grounds that they are simply not indicated, apart from the matter of cost. Despite its attractions, there are good reasons why we should not use this concept to solve problems of allocation. First, arguments based on the futility concept conceal many statistical and value-laden assumptions, whereas strategies based on resource allocation force these assumptions to be stated explicitly. Societies may choose to limit the use of therapies that may be of value and have a reasonable likelihood of success in some cases. For example, the much discussed Oregon plan for allocating Medicaid funds27 seeks to reflect community values in ranking various health care goals (placing preventive care ahead of cosmetic surgery, for example). Since rationing policies make explicit the values and probabilities that futility-based arguments leave implicit, it is clearly preferable to develop and adopt them rather

Truog/Brett/Frader: The Problem with Futility

than use futility arguments as a cover for limiting the availability of scarce and expensive resources. Another problem with invoking the idea of futility in the debate over allocation is that we have no reason to believe that it is applicable in enough cases to make a difference in the scarcity of medical resources. Although it may be true that beds in the intensive care unit (especially those used for extracorporeal membrane oxygenation) are relatively scarce, it seems unlikely that patients similar to Helga Wanglie occupy an important fraction of those beds, let alone account for a major proportion of the cost of medical care in the United States. From a macroeconomic perspective at least, we must remain skeptical that an appeal to the idea of futility will get us very far. Moving beyond Futility Our rejection of futility as a useful concept does not imply that we endorse patients' unrestricted demands for interventions such as those described in our prototypical scenarios. On the contrary when providers oppose such demands they are usually acting from a profound sense that further treatment would be fundamentally wrong. Our task is to take account of that sense of wrongness without resorting to unilateral, provider-initiated declarations of futility. In many of the situations in which questions of futility arise, providers believe that the treatment in question would not be in the patient's interests, even from the patient's perspective, and that any insistence by the patient (or surrogate) on further interventions is based on faulty reasoning, unrealistic expectations, or psychological factors, such as denial or guilt. In these circumstances, providers are obligated to make every effort to clarify precisely what the patient intends to achieve with continued treatment. If the patient's goals appear to reflect unrealistic expectations about the probable course of the underlying illness or the probable effect of medical interventions, providers should attempt to correct those impressions. Because inadequate or insensitive communication by providers probably accounts for a substantial proportion of unrealistic requests, such discussions will successfully resolve many conflicts.14'28 Empirical studies of ethics consultations have demonstrated precisely this point.29.30 Although this appeal to the patient's interests may seem to contain some of the same ambiguities as arguments using the concept of futility, there is a subtle but important distinction between the two. Judgments about what is in the patient's interest are properly grounded in the patient's perspective, whereas judgments cast in the language of futility falsely assume that there is an

objective and dispassionate standard for determining benefits and burdens. Nevertheless, even after providers make sustained attempts to clarify patients' preferences, some patients or surrogates will continue to demand life-sustaining interventions when the caregivers feel deeply troubled about providing them. In many such cases unrestrained deference to the wishes of the patient or surrogate conflicts with two other values that do not require a unilateral judgment of the futility of treatment: professional ideals and social consensus. The ideals of medical professionals include respect for patients' wishes, to be sure, but they also include other values, such as compassionate action and the minimization of suffering. Consider, for example, a bedridden victim of multiple strokes who has contractures and bedsores and who "communicates" only by moaning or grimacing when she is touched. Physicians asked to perform chest compressions, institute mechanical ventilation, or use other life-sustaining interventions in such a patient may regard these actions as cruel and inhumane.31 Moreover, physicians and other caregivers have a legitimate interest in seeing that their knowledge and skills are used wisely and effectively. For example, if surgeons were repeatedly pressured to perform operations that they believed to be inappropriate, they would certainly suffer a loss of dignity and sense of purpose. Although appealing to professional ideals can serve as a convenient means of protecting the interests of physicians at the expense of patients' values, these ideals are legitimate factors to weigh against other values. To dismiss this perspective as irrelevant in decision-making is to deny an essential part of what it means to practice medicine. Although we believe that health care professionals should not be required to take part in care that violates their own morals, the law in this area remains uncertain. On the one hand, courts have upheld a state interest in protecting the ethical integrity of the medical profession. This may provide some basis for protecting doctors who wish to refrain from cruel or inhumane treatment, despite the wishes of the patient or surrogate.32 On the other hand, in the two cases that have led to court decisions (those of Helga Wanglie3 and of Jane Doe in Atlanta33) the judges upheld the surrogates' decision-making authority. Clearly, this area of the law remains to be defined. Finally, social consensus is yet another expression of the values at stake in some medical decisions. In a pluralistic society, differences in personal values and interests occasionally run so deep that they cannot be resolved by the introduction of additional facts or by further private debate. At certain critical junctures, the resolution of these conflicts may require an explicit public process of social decision-making.34 Social consensus has been sought, for

179

I I 31

111111

4 Professional Ethics

example, to address the issue of fair allocation of resources.27 The involvement of society is also essential when the most highly charged questions of morality are at stake, as in the increasingly heated debate over euthanasia.35 In the prototypical scenarios described at the outset of this article, an ongoing attempt to achieve social consensus is perhaps most conspicuous with regard to the prolongation of life for patients in a persistent vegetative state. From a legal perspective, the relevant decisions began with the case of Karen Quinlan36 and have extended through that of Nancy Cruzan.37 These cases have increased awareness of the ethical issues raised by the situation of patients in a persistent vegetative state and have helped to consolidate the view that it is acceptable to withdraw life-sustaining treatment from patients in such a state. Controversy does remain about who has the ultimate authority to make these decisions. Some hold that the choice must remain with the patient or surrogate, whereas others believe that under some circumstances this prerogative may be overridden. For example, the Hastings Center38 and the Society of Critical Care Medicine39 have concluded that providing intensive care to patients in a persistent vegetative state is generally a misuse of resources, and the President's Commission stated that such patients should be removed from life support if such action is necessary to benefit another patient who is not in a persistent vegetative state.4° It is unclear how this debate will conclude, but the confluence of medical, legal, and ethical thinking about the persistent vegetative state is an example of how social consensus may evolve. In summary, the Wanglie case demonstrates how the resolution of these conflicts must proceed on many levels. Most such cases will benefit from sustained attempts to clarify the patient's values and the likelihood of the various relevant outcomes and to improve communication with patients or their surrogates. When this

approach fails, physicians and other caregivers should ask themselves whether the care requested is consistent with their professional ethics and ideals. When these ideals appear to be violated, either alternative venues for such care should be found or the conflict should be addressed in a public forum. This broader review could be provided through institutional mechanisms, such as the hospital's ethics committee, or by the courts. The public scrutiny that attends such cases will further the debate over the appropriate use of medical resources and foster the development of consensus through legislation and public policy. Conclusion In outlining the perspectives of the principal stakeholders—patients and their surrogates, physicians, and society—we have avoided the construction of a rigid formula for resolving conflicts over interventions frequently regarded as futile. Because of clinical heterogeneity, pluralistic values, and the evolutionary nature of social consensus, most clinical decisionmaking on behalf of critically ill patients defies reduction to universally applicable principles. The notion of futility generally fails to provide an ethically coherent ground for limiting life-sustaining treatment, except in circumstances in which narrowly defined physiologic futility can be plausibly invoked. Futility has been conceptualized as an objective entity independent of the patient's or surrogate's perspective, but differences in values and the variable probabilities of clinical outcomes undermine its basis. Furthermore, assertions of futility may camouflage judgments of comparative worth that are implicit in debates about the allocation of resources. In short, the problem with futility is that its promise of objectivity can rarely be fulfilled. The rapid advance of the language of futility into the jargon of bioethics should be followed by an equally rapid retreat.

References 1. Jacobellis v. State o f Ohio, 84 S Ct 1676 (1964). 6. 2. Hippocrates. The art. In: Reiser SJ, Dyck AJ, Curran WJ. eds. Ethics in medicine: historical perspectives and contemporary concerns. Cambridge, Mass.: MIT Press, 1977:6-7. 7. 3. Capron AM. In re Helga Wanglie. Hastings Cent Rep 1991 ;21 (5): 26-8. 8. 4. Lantos JD, Singer PA, Walker RM, et al. The illusion of futility in clinical practice. Am J Med 1989;87:81-4. 5. Standards for cardiopulmonary resuscitation (CPR) 9. and emergency cardiac care (ECC). V. Medicolegal considerations and recommendations. JAMA 10. I 974;227:Supp1:864-6.

Appendix A the proposed legislation In: Do not resuscitate orders: the proposed legislation and report of the New York State Task Force on Life and the Law. 2nd ed. New York: The Task Force, 1986:83. 1984 Amendments to the Child Abuse Prevention and Treatment Act. Pub Law 98-457.1984. Council on Ethical and Judicial Affairs, American Medical Association. Guidelines for the appropriate use of do-not-resuscitate orders. JAMA 1991;265:1868-71. Miles SH. Informed demand for "non-beneficial" medical treatment. N Engl J Med 1991;325:512-5. Brain-damaged woman at center of lawsuit over lifesupport dies. New York Times July 5,1991:A8.

Truog/Brett/Frader: The Problem with Futility

11. Paris J J, Crone RK, Reanlon F. Physicians' refusal of re- 25. quested treatment: the case of Baby L N Engl J Med 1990;322:1012-15. 12. Blackball U. Must we always use CPR? N Engl J Med 26. 1987;317:1281-5. 13. Hackler JC, Hiller F C. Family consent t o orders not t o resuscitate: reconsidering hospital policy. JAMA 1990;264:1281-3. 27. 14. Murphy DJ. Do-not-resuscitate orders: time for reappraisal in long-term care institutions. JAMA 28. 1988;260:2098-101. 15. Bedell SE, Delbanco TL, Cook E F, Epstein FH. Survival 29. after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983;309:569-76. 16. Danis M, Gerrity MS, Southerland LI, Patrick DL. 30. A comparison of patient, family, and physician assessments of the value of medical intensive care. Crit Care 31. Med 1988;16:594-600. 17. Danis M, Jarr SL, Southerland LI, Nocella RS, Patrick DL. A comparison of patient, family, and nurse evalua- 32. tions of the usefulness of intensive care. Crit Care Med 1987;15:138-43. 33. 18. Danis M, Patrick DL, Southerland LI, Green ML. Patients' and families' preferences for medical intensive 34. care. JAMA 1988;260:797-802. 19. Schneiderman LJ, Jecker NS, Jonsen AR. Medical futility its meaning and ethical implications. Ann Intern 35. Med 1990;112:949-54. 20. The Hastings Center. Guidelines on the termination 36. of life-sustaining treatment and the care of the dying. Bloomington: Indiana University Press, 1987:32. 37. 21. Appendix C: New York Public Health Law Article 29-B-orders not to resuscitate. In: Do not resuscitate 38. orders: the proposed legislation and report of the New York State Task Force on Life and the Law. 2nd ed. New York: The Task Force, 1986:96. 39. 22. Tversky A, Kahneman D. Judgment under uncertainty: heuristics and biases. Science 1974;185:1124-31. 23. Elstein AS. Clinical judgment: psychological research 40. and medical practice. Science 1976;194:696-700. 24. Poses RM, Bekes C, Copare FJ, Scott WE. The answer to "What are my chances, doctor?" depends on whom is asked: prognostic disagreement and inaccuracy for critically ill patients. Crit Care Med 1989;17:827-33.

Should Physicians Be Gatekeepers of Medical Resources? Milton C. Weinstein The Medical Commons Resources available for health care are finite. This means that it is impossible for physicians collectively to

Poses RM, Cebul RD, Collins M, Fager SS. The accuracy of experienced physicians' probability estimates for patients with sore throats: implications for decision making. JAMA 1985;254:925-9. Aaron H, Schwartz WB. Rationing health care: the choice before us. Science 1990;247:418-22. Eddy DM. What's going on in Oregon? JAMA 1991;266:417-20. Youngner SJ. Who defines futility? JAMA 1988;260:2094-5. Brennan TA. Ethics committees and decisions to limit care: the experience at the Massachusetts General Hospital. JAMA 1988;260:803-7. La Puma J. Consultations in clinical ethics-issues and questions in 27 cases. West J Med 1987;146:633-7. Braithwaite 5, Thomasma DC. New guidelines on foregoing life-sustaining treatment in incompetent patients: an anti-cruelty policy. Ann Intern Med 1986;104:711-5. Meisel A. The right to die. New York John Wiley & Sons, 1989:104. In re: Doe, Civil Action No. 093064 (Fulton County, GA, October 17, 1991). Callahan D. Medical futility, medical necessity: the-problem-without-a-name. Hastings Cent Rep 1991;21(4):30-5. Misbin RI. Physicians' aid in dying. N Engl J Med 1991;325:1307-11. In the Matter of Karen Ann Quinlan, an alleged incompetent. 355 A.2d 647; or 70 NJ 10. March 31,1976. Annas GJ. Nancy Cruzan and the right to die. N Engl J Med I 990;323:670-3. The Hastings Center. Guidelines on the termination of life-sustaining treatment and the care of the dying. Bloomington: Indiana University Press, 1987:112. Task Force on Ethics of the Society of Critical Care Medicine. Consensus report on the ethics of foregoing lifesustaining treatments in the critically ill Crit Care Med 1990;18:1435-9. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Deciding to forego life-sustaining treatment: ethical, medical, and legal issues in treatment decisions. Washington, DC: Government Printing Office, 1983:188-9.

offer all technologically feasible and clinically beneficial medical services to all patients. Individually, though, physicians practise medicine under a basic ethical tenet which compels them to do whatever is in their power to help their patients. Herein lies one of the fundamental ethical issues in modern medicine: how can physicians fulfil their moral obligations as fiduciary agents for individual patients while being responsible stewards of the finite pool of resources?

Source Fromfourrial of Medical Ethics 27 (2001), 4,268-74. Reprinted with permission of the BM J Publishing Group.

I I DII

111111

182 4 Professional Ethics

In a 1975 article in the New England Journal of Medicine,' Howard Hiatt likened the situation in medical care to a parable described by Garrett Hardin in a classic article, entitled "The Tragedy of the Commons."' According to Hiatt's adaptation of Hardin's parable, physicians are like herdsmen who used to feed their cattle (patients) on a common pasture. Acting in their own interests, the herdsmen could allow their cattle to feed on the land without limit, as long as their numbers and appetites were small compared to the resources on the land. As the number and appetites of the cattle grew to the point where, collectively, their wants exceeded the capacity of the common resource, the desires of the herdsmen to extract the maximal nutrition for their cattle led to overgrazing. At first, the less aggressive cattle failed to get adequate nutrition and died. Later, herdsmen were forced out of business. And, in the end, the rich pasture turned into an overgrazed wasteland. Although Hardin's essay was written in the context of population growth, Hiatt saw its relevance to health care. Physicians, each acting in the best interests of their own patients, collectively reach the limits of health care resources, with the result that access to care and quality of care are compromised. There is no obvious ethical solution to the problem of rationing the medical commons, because any solution involves comparisons between the value of health services provided to different patients with different conditions. Nonetheless, the remedy must lie in some form of collective action: physicians, like the herdsmen in Hardin's parable, can save the commons only by adhering to a set of mutually acceptable covenants which govern and limit their use of the shared resource. If health care were "free," there would be no need to limit its use. Health care is not "free," because the use of resources (physician time, hospital beds, health care budgets) by some precludes the use of those resources by others. The overall result of failing to adopt covenants that lead to restrained use of resources by well-meaning physicians is unacceptable. If a society mandates universal access to health care, and if all physicians provide their patients with the most beneficial treatments available, then the cost of health care will be unacceptably high to their patients, either as taxpayers, payers of insurance premiums, at the point of care, or in combination. The alternatives are compromises, either with the principle of universal access or with the principle of unlimited care. Failure to acknowledge these trade-offs can lead to inefficiencies and inequities that compromise both principles: hidden barriers to access such as queues and administrative hassles, exclusion of entire segments of the population, or erecting rigid and

arbitrary barriers to introducing new treatments while older treatments of questionable efficacy continue to be used. This essay begins with a view of this problem from the perspective of society as a whole. Adopting a utilitarian stance, different allocations of resources may be judged by how well they achieve some generally agreed upon measure of social benefit. This representation leads to cost-effectiveness analysis as a guide to resource allocation from the societal perspective. The next section describes the fundamental principles of cost-effectiveness analysis as a guide to resource allocation. The following three sections of the essay contrast the perspectives of patients, consumers (i.e., patients ex ante), and physicians on resource allocation. Evidently there are differences among these perspectives regarding the degree to which resource cost should enter into decision-making. Concepts that overlap economics and moral philosophy, such as moral hazard and fiduciary relationships, are introduced along the way. I conclude that neither patients nor physicians can be expected to attend to the rationing of the medical "commons" without some form of collective action but that collective action alone will not succeed without genuine acceptance of the need to ration health care resources by both physicians and patients. The remainder of the essay concerns the structure of collective solutions, the purpose of which is to permit physicians to fulfil their responsibilities as fiduciary agents to patients, while being responsible gatekeepers of resources. Forms of collective action may range from government regulation to decentralised budgeting of providers, to non-binding clinical guidelines. The promises and limitations of such devices as capitated payments, limited access to technology, and voluntary guidelines are also considered. An analogy is drawn with public regulation of pollution. The essay concludes with observations regarding the role of costeffectiveness analysis at the bedside. The Cost-Effectiveness Paradigm

Suppose that a society wants to provide the maximum aggregate health improvement in its population, but it has limited resources to do so. Each potential health intervention delivered to a defined group of persons with a particular condition yields a health improvement and entails a cost. If health improvement is measured in units which reflect the values of the society, and if costs are measured in units which reflect the extent to which the resource budget is depleted by the intervention, then the maximum societal health improvement can

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

be achieved by applying the following simple decision rule: rank order medical interventions in decreasing order of their expected health improvement per unit expected cost, and adopt them from the top of the rank list to the point on the list that resources are depleted.3 The ratio of benefit to cost from each intervention represents its "value for money." To make this decision rule operational even at the societal level, one needs measures of predicted health improvement or benefit and predicted resource cost. (For present purposes, complexities arising from uncertainty regarding anticipated benefits and costs are set aside, and are summarised in terms of the average, or expected, values of each.) The health benefit can be measured in units that reflect the preferences of the community, considering their desire for increased longevity but also the value they place on limitation of function, pain, and other dimensions of health-related quality of life.4 One such measure is the quality adjusted life year (QALY).3-5 Quality adjusted life years measure the number of years of remaining life, each adjusted by a preference weight (generally between zero and one, or possibly even less than zero for health states judged worse than dead) that reflects the relative value of the health state on a scale between perfect health (one) and dead (zero).3-6 Since the amount of health benefit is uncertain before the intervention, the measure of benefit for a health care intervention can be expressed as an average across similar persons in the target population, in terms of quality adjusted life expectancy. The QALY gain for an intervention should include all health consequences, both positive and negative, and is therefore a measure of net health benefit.5 Monetary Terms

Costs are usually measured in monetary terms—dollars, pounds, euros—but it should always be remembered that money is only a proxy for the real resources— physicians, nurses, hospital facilities, pharmaceutical development and production—that are consumed in providing the service.5 From society's perspective, costs are also net of any savings in future health care resources that might have been consumed but for the intervention—such as the costs of treating strokes that are averted because of a blood pressure control programme. From a societal perspective, costs also include resources contributed by patients and family members and other caregivers, including their time, and not just resources financed through health insurance (public or private) or billed to the patient. In common practice, the ratio reported from costeffectiveness analyses is the reciprocal of "value for

money," namely, the cost per quality adjusted year of life gained.3-5 Thus, interventions having low values of this ratio go to the top of the rank list, and those with high values may fall below the line for a particular budget. Some programmes may actually save more resources than they consume while still providing a health improvement. Such "cost-saving" programmes automatically go to the top of the rank list because they do not deplete the resource pool; indeed, they add to it. Contrary to popular beliefs, however, very few programmes, even in the area of prevention, are truly cost-saving.' 8 What is the ethical basis for seeking to allocate society's health care resources with the aim of maximising quality adjusted life expectancy? If each individual measured the value of his or her own health in terms of quality adjusted life expectancy, and if each of them paid for his or her own health care through individual savings accounts, then economic theory would lead them each to allocate their own assets according to the cost-per-QALY rule.9 Because of the uncertainties about the need for health care during one's life, however, and for reasons of equity, health care services are typically covered by pooled risk-sharing arrangements in the form of insurance or national health care. Under these arrangements, the measure of aggregate QALYs at the societal level entails an interpersonal comparison: the implicit assumption is that a QALY is a QALY, no matter who gets it. Much has been written regarding the ethical justification for such a social value. Citizens behind a primordial veil of ignorancel" might elect to pre-specify that health resource allocations be made so as to maximise their expected utility as viewed from behind the veil. If their preferences after the veil is removed are to maximise their own quality adjusted life expectancies, then a society that maximises aggregate QALYs might appeal to them. Rawls, however," comes to a rather different conclusion from the same initial premise. He concludes that individuals behind the veil would attach more weight to gains in well-being should they emerge with disadvantages, such as poverty or poor health. Thus, more weight would be applied to health improvements for those persons in the most disadvantaged positions in society. Alternatives to QALYs that reflect societal preferences for differential weighting of health gains to the least healthy have been proposed by Nord." Whether a society wishes to count all QALYs equally or to weight them inversely to the health status of the beneficiary, the result is a utilitarian measure of societal health benefit that could be used to guide

.1 .11

1

1.11

II ill

muu

4 Professional Ethics

resource allocations in health care through an appropriately constructed ratio. The fundamental question in this essay is this: if there is general consensus that resources should be allocated in such a way as to maximise aggregate health benefit, who is responsible for allocating the resources? Should the patient voluntarily deny himself health services out of a sense of communal obligation to conserve the commons? Should physicians be expected to balance their responsibilities to do the best they can for their patients with a responsibility to be the gatekeeper of the commons at each and every encounter with patients? Or is some form of collective action required whereby citizens empower their physicians to practise medicine within a system that imposes limits on the resources available to them? My conclusion is that a combination of all three is necessary in order to allocate medical resources efficiently. The Role of Patients

Patients expect physicians, as fiduciary agents, to do everything in their power to provide them with the best possible health care. Americans in particular, perhaps in contrast with citizens of other industrialised countries, demand the maximal use of available technologies. They are none the less accustomed to the fact that their physicians are already forced to ration care because of constraints that are neither financial nor under their control. For example, physicians have limited time during the day, which results in barriers to scheduling office appointments or conducting the most thorough possible examinations. Access to facilities, such as hospital beds and especially diagnostic technologies such as magnetic resonance imaging and computed tomography, may be limited, forcing physicians to delay, or even forgo, some diagnostic information. In sum, physicians ration care to some extent, or at least set priorities, and patients know that other patients may take precedence for their physician's attention, depending on the urgency and severity of their problems. Unfortunately, the services that physicians are most likely to forgo under these pressures are those which patients do not actively seek, much less demand. Some of these services, such as periodic screening for colorectal and breast cancer, or blood pressure monitoring and treatment, may be far more cost-effective (by the cost-per-QALY criterion) than the services that physicians are compelled, by their sense of obligation to patients, to provide. Patients themselves ration their own health care to some degree. In the United States, insured patients often face co-payments or deductibles. Looming even

larger as resource constraints upon patients are their own time, including travel to the doctor's office, and other out-of-pocket costs such as transportation and child care. In general, these constraints on physicians and patients are relatively weak deterrents to the use of maximally beneficial medical care. They are not strong enough by themselves to allocate the commons, and to the extent that they limit care, they may not do so efficiently. The Role of Consumers (Patients ex ante)

As patients, people have a different view of health care costs than they have as consumers. While people may expect that physicians do everything possible for them when they are sick, they complain bitterly when their insurance premiums rise, when the prices of goods and services go up because labour costs to employers reflect rising health care benefits, or when their taxes go up. Consumers have to pay for their collective use of health care resources, but they don't want to bear responsibility for the collective costs at the point of their own care. Part of the motivation for individuals is explained by the economic theory of insurance—groups of people can become better off by pooling risks and avoiding major losses in the event they become sick. As a result, citizens demand that health insurance be provided by their employers (or made available at nominal cost) or by government. Moreover, subsidised health insurance enables citizens who would not otherwise be able to afford basic health care to obtain it; there is a redistributional aspect to the provision of health insurance. When patients have medical insurance, they face different incentives in going to the doctor to seek care. Price becomes less of a factor, because the patient does not bear the full cost of care at the time care is sought. In effect, the price facing the patient is lower than the full social cost of health care. The gap between perceived price and resource cost creates an incentive to utilise more health care than the patient would otherwise be willing to buy. This phenomenon is known as "moral hazard" in the economics literature.14 It tends to promote excessive use of the "medical commons," leading to increases in the cost of health insurance beyond what consumers believe is reasonable. Ironically, the word "moral" in "moral hazard" suggests that patients are at risk of being "immoral" if they overutilise health care services relative to the value of the services they receive. This nomenclature stands in sharp contrast to the predominant ethical problem faced by physicians in the presence of limited

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

resources: their ethical obligation to the patient at hand makes them immoral if they underutilise services relative to the maximum they could do. In reality, neither the physician placed in the fiduciary role as agent for the patient, nor the patient facing artificially low prices at the point of care, can be faulted for providing and expecting, respectively the best health care technology can offer. The Role of Physicians

Physicians and patients engage in what economists refer to as principal—agent relationships. This refers to the fiduciary trust that patients (the principals) place in physicians (the agents) to make decisions that maximise the well-being of their patients and to act as advocates for their patients in the health care system. This fiduciary responsibility has been described by Fried as entailing doing whatever is possible to "[preserve] life capacities for the realization of a reasonable, realistic life plan."15 As ideal agents for patients, physicians would consider not only the health consequences of their decisions but also the economic and psychological consequences for their patients. Out-of-pocket costs, time and inconvenience, and reassurance from diagnostic tests are all part of what patients value. A perfect agent would consider all of these and weigh them against one another if necessary, as would the patient. A physician who recommends an expensive, new treatment that is not covered by the patient's insurance but that is equally effective as an insured procedure would not be acting as a good agent. But suppose the patient preferred a far more expensive, insured procedure whose only advantage was that it avoided a small scar or a few hours less waiting time. Where does the physician's obligation as an agent stop? Clearly, physicians often depart from perfect agency even with the most honourable and selfless of intentions. Principal—agent relationships do not always achieve the goal of perfect proxy decision-making and selfless advocacy if the incentives facing the agent lead to deviations from the decision that is best for the principal. For example, physicians may obtain different levels of remuneration, professional stature, or satisfaction from their actions, and these incentives compete with the incentives that are aligned with the interests of their patients, such as a sense of obligation and cognisance of external monitoring of their quality of care. As noted before, physicians are often unable to be perfect agents for their patients because of constraints placed upon them. They have limited time during the day (and even if they work extra hours, their

performance may suffer). They are limited by resource constraints in hospitals and laboratories, such as intensive care beds and magnetic resonance equipment. Sometimes constraints force the physician to make explicit choices between the interests of different patients, as in the setting of emergency triage or in the selection of organ transplant recipients. In these situations, the choices concern patients with names and faces, all of whom are under the care of a single physician or provider organisation. The question at the societal level is whether physicians should be expected to allocate resources between their patients and other nameless, faceless patients who could, perhaps, obtain more benefit if the resources were conserved to benefit them. The question, in other words, is whether physicians should consider the cost-effectiveness of the decisions they make for their patients, recognising that resources are limited. Is it the physician's responsibility to protect the medical commons? An entirely different view of the physician's ethical responsibility would be as an agent for society at large rather than for individual patients. Under this view of agency, the physician would be compelled to allocate resources in the most cost-effective manner in order to achieve maximal value for money on behalf of society. In such a world, physicians would make decisions that are in less than the best interests of their individual patients, because not all medically beneficial procedures would be provided. But, on aggregate, consumers would be better off because more health benefit would be achieved. Or would they? Such an ethic would compromise the fiduciary relationship between patient and physician. It would place the physician in a position of making trade-offs with faceless patients, and then explaining to patients why it was not "cost-effective" to do extra computed tomography (CT) or to prescribe the more expensive drug with a slightly better side effect profile. Physicians trained to do the best for individual patients would balk at this social agency role. Major Sea Change

The dilemma, then, is that consumers collectively, concerned about health care costs, have an interest in costconsciousness in the clinic and at the bedside, while as patients they expect a perfect agency relationship. One possible class of solutions may entail collective action to constrain (not necessarily regulate!) physicians to act in the collective interest while allowing them to strive to maximise the welfare of their individual patients. This is the type of solution that economists advocate, for example, to induce manufacturers to restrict pollution

186

4 Professional Ethics

or to induce consumers to recycle. But in health care, it is not clear that consumers will accept such collective solutions until they first buy into the premise that resources are limited. I believe that a combination of incentives, constraints, and a major sea change in citizen attitudes toward the finitude of medical care will be required to save the medical commons. The Role of Collective Action

The goal of a collective solution to protect the medical commons would be served by placing constraints on physicians' choices such that, even as they strive to do the best they can on behalf of their patients, the result of their doing so leads to cost-effective resource allocation. The constraints would, in effect, force physicians to consider the opportunity cost of their decisions, just as shopping consumers are bound by the cash in their wallet and their credit balance. The consumer seeks to maximise her well-being, subject to a budget constraint. The idea is to get the physician to take cognisance of the collective budget constraint. Many examples of incentives to consider opportunity costs pervade the existing health care systems of the industrialised world. Capitated (per patient) payments to physicians encourage them to use their time efficiently. Capitated budgets for hospitals have a similar effect at the institutional level. One problem with capitation as an incentive at the local or individual physician level, however, is that it mitigates the advantage of insurance and risk pooling. If all providers allocate their separate budgets to maximise health improvement, the overall consequence may be suboptimal if the most cost-effective opportunities to improve health are spread unevenly among providers. Over time, monitoring of caseloads and case-mix could be used to adjust the budget levels and thereby equalise the incremental value of resources in different settings. If physicians consider both QALYs (or another socially desirable measure of health outcome) and cost when deciding how to allocate their budgets, the result would produce the societal allocation implied by costeffectiveness analysis. Physicians in that setting would be led to consider the cost per QALY of alternative decisions, and the result would be the maximum possible production of QALYs by that physician. If physicians consider other values, for example, if they give additional priority to the sickest patients as suggested by Nord,' then the result would reproduce the costeffective result based on those values. The current situation in the United States, dominated by managed care, has largely eschewed this sort of decentralised, provider-centred system in favour of direct

regulation of medical services. Managed care organisations, bound by bottom-line concerns, keep a close watch on the resource allocations of their affiliated providers. The result has been less this sort of decentralised decisionmaking than a system dominated by prior approval and sanctions against providers for overutilisation. Whether insurers and managed care organisations actually consider cost-effectiveness in their regulation of care is uncertain16 If they do, their perspective on both cost and effectiveness departs from that of society at large by virtue of their short time horizons, responsiveness to popular demands over clinical effectiveness, and other factors.17 The result has been a large number of dissatisfied consumers and an even larger proportion of disgruntled physicians. While some economists might applaud the decentralised, provider-centred solution in theory, it does have a number of practical flaws. For one, physicians would have to fend off patients whose demands for health care services were not being met. There would be a tendency to give more attention to the loudest, most assertive patients, relative to the cost-effectiveness of their claims on the resource pool, and relatively less attention to more passive patients. Perhaps a major barrier to successful implementation of the providercentred model is the need for buy-in on the part of both physicians and consumers. I will return to this essential ingredient later. The Role of Physicians under Resource Constraints

A responsible physician who cares for a panel of patients, but who is either faced with a resource budget or accountable for the resources he consumes, would be placed in a situation similar to the emergency room physician performing triage. The physician would be responsible for setting or implementing priorities for care and could invoke cost-effectiveness data to help guide these priorities. This would enable physicians to consider the incremental health benefit they could offer to each patient, with perhaps some added moral consideration given to applying resources to the most desperate cases.n Whatever metric the physicians use, whether implicitly or explicitly, they would be striving to maximise the welfare of their patient population. Like parents taking care of their several children—decisions about what clothing to buy for each, which ones to send to sports camp, which to send to college—physicians could approach these interpersonal comparisons in a caring, compassionate way. Given the complexity of medical decisions, and in light of the growing recognition of the role of evidence on effectiveness and cost in medical decisions, physicians must work together to develop and interpret the

Weinstein: Should Physicians Be Gatekeepers of Medical Resources?

evidence to support a mutually acceptable framework for cost-effective decision-making. This entails physician participation at two levels: helping to set the constraints within which they practise and formulating flexible guidelines to help them allocate the resources under their stewardship. These, then, are the additional roles physicians must play in order to ensure resource allocations that are responsive to patients' needs and preferences. If physicians accept this role of stewardship for their portion of the medical commons, they can continue to exercise their roles as agents and advocates for patients. An Apt Analogy: Pollution Control It would be unrealistic, and a violation of the trust patients place in their physicians, to expect physicians to practise cost-effectively without some form of external constraint. The ethical forces acting upon physicians will pull in the opposite direction if physicians are left to make trade-offs between their own patients and other, unseen, patients. The force compelling them to attend to their own patients is, and should be, stronger, with the result that society's commons problem is left unattended. The situation is analogous to the control of pollution. As in health care, self-interest (it does not matter whether the self-interest reflects economic or health values) competes with the interests of the greater society. The managers of companies, especially publicly owned ones, have a primary fiduciary obligation to their shareholders. If profit maximisation entails the discharge of pollutants, then it would be a violation of their agency relationship to shareholders to volunteer to regulate pollution more than their shareholders would want. But other members of society, who are not shareholders, have an interest in pollution control. Their interest in the company, while perhaps not zero because the costs of pollution control would be passed on to them in part through higher product prices, would be greatly attenuated by comparison with the adverse effects of the pollution. Should these managers be expected to restrict pollution voluntarily below the levels their shareholders would want? Would it be ethical for them to do so? Recognising that a valid argument could be made that it would be unethical for the managers to fail to abide by their shareholders' wishes, society adopts pollution control regulations (or pollution discharge quotas or taxes). Responsible companies, while they may lobby for less stringent standards, higher quotas, or lower taxes, generally accept the rationale for collective action to protect the greater interest of the community. They then set about their business of maximising profits within the constraints. Sometimes they attempt to bypass the constraints or advocate loosening of the

standards on behalf of their shareholders, but the laws are enforced, and they learn to do the best they can for the shareholders within the constraints. Concluding Remarks:The Need for"Buy-In". Should physicians do cost-effectiveness analysis at the bedside, even if they are not compelled to ration care by externally imposed constraints? To do so would place them in an untenable position as agents for two, sometimes adversarial principals—the individual patient in the clinic and the larger community. Patients would lose the undivided advocacy to which they are accustomed, and the pressures on physicians to be responsive would be overwhelming. Collective action is required to constrain individual providers of health care to practise within their collective means. These constraints could be implemented in and by organisations as diverse as managed care organisations and national health services. Or they could be agreed upon by consensus of provider organisations and citizens. Decision-making within these constraints would invite cost-effectiveness analysis by individual providers, but now the choices would be among members of the physician's own "family" of patients, just as the parent decides how to tend to the needs and wants of her several children. The physician would be free to advocate for each patient, to do the best he can to "[preserve] life capacities for the realization of a reasonable, realistic life plan."15 Such a means of protecting the "medical commons" could result in resentment on the part of physicians and patients alike, unless both accept the underlying rationale for the constraints. The system of utilisation review by managed care organisations in the United States has spawned a movement in support of a return to physician control of decision-making. But a return to unconstrained use of technologies and resources will fail because of the limits at the societal level. The only way out of this dilemma is for citizens and physicians to accept the concept and consequences of resource limits, just as they accept speed limits, zoning laws, and other self-imposed constraints in the interest of the greater good. Attitude of Consumers How can this acceptance of resource constraints come about? There is some indication that physicians are already willing to accept the idea of limits, as long as they have some degree of control of or participation in the process of setting them. Medical specialty societies have begun to incorporate cost-effectiveness into their formulation of clinical guidelines. A recent study showed good concordance between the rankings by a

II

MILLI

4 Professional Ethics

consensus panel of physicians of appropriateness of a medical technology in different indications-coronary angiography after myocardial infarction-and costeffectiveness ratios for these indications.'8 The biggest obstacle to cost-effective resource allocation along these lines is the attitude of consumers. People who expect everyone to have access to all possible medical care regardless of cost are bound to be disappointed. In the United States, the sustenance of this myth has come at a high price. Even in times of economic prosperity, an increasing proportion of Americans is uninsured, and millions of them have limited or no access to medical care. Some of the most cost-effective but least

glamorous medical interventions, such as screening for colorectal cancer and adult vaccinations against influenza and pneumonia, are underutilised in favour of the procedures consumers want and demand. An informed populace, aware of the finite benefits of health care services and of the rationale for constrained choice in the clinic and at the bedside, is an essential ingredient if physicians are to be burdened with the responsibility for cost-effective decision-making. The evidence suggests that the medical profession will accept that responsibility if it is given the authority to set the rules by which its members are to play and if their patients are at peace with the principle of living within their means.

References

1. Hiatt HH. Protecting the medical commons: who is respon- 11. sible? New England Journal of Medicine 1975;293:235-41. 2. Hardin G. The tragedy o f the commons. Science 12. 1968;162:1243-8. 3. Weinstein MC, Stason WB. Foundations of cost- 13. effectiveness analysis for health and medical practices. New England Journal of Medicine 1977;296:716-21. 4. Patrick DL, Erickson P. Health status and health policy: 14. allocating resources to health care. New York: Oxford 15. University Press, 1993. 5. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Costeffectiveness in health and medicine. New York: Oxford 16. University Press, 1996. 6. Torrance GW. Measurement of health state utilities for economic appraisal. Journal of Health Economics 1986;5:1-30. 7. Russell LB. Is prevention better than cure? Washington: 17. The Brookings Institution, 1986. 8. Weinstein MC. The costs of prevention. Journal of General Internal Medicine 1990;5(suppl):89-92s. 9. Garber AM, Phelps CE. Economic foundations of cost-effectiveness analysis. Journal of Health Economics 18. 1997;16:1-31. 10. Harsanyi JC. Cardinal welfare, individualistic ethics, and interpersonal comparisons of utility. Journal of Political Economy 1955;63:309-21.

Caring in a Crisis: The Ethical Obligations of Physicians and Society during a Pandemic Canadian Medical Association

Inherent in all health care professional Codes of Ethics is the duty to provide care to patients and

Rawls J. A theory of justice. Cambridge, MA: Harvard University Press, 1971. Nord E. An alternative to QALYs: the saved young life equivalent. British Medical Journal 1992;305:875-7. Aaron H J, Schwartz WB. The painful prescription: rationing hospital care. Washington: The Brookings Institution, 1984. Pauly MV. The economics of moral hazard. American Economic Review 1968;58:231-7. Fried C. Medical experimentation: personal integrity and social policy. New York: American Elsevier Publishing Co, 1974. Prosser LA, Koplan JP Neumann PJ, Weinstein MC. Barriers to using cost-effectiveness analysis in managed care decision making. American Journal of Managed Care 1999;6:173-9. Weinstein MC. Principles of cost-effective resource allocation in health care organizations. International Journal of Technological Assessment of Health Care 1990;6:93-103. Kuntz KM, Tsevat J, Weinstein MC, Goldman L. Expert panel vs decision-analysis recommendations for postdischarge coronary angiography after myocardial infarction. Journal of the American Medical Association 1999;282:2246-51.

to relieve suffering whenever possible. However, this duty does not exist in a vacuum and depends on the provision of goods and services referred to as reciprocal obligations, which must be provided by governments, health care institutions, and other relevant bodies and agencies. The obligations of government and society to physicians can be seen as comparable to the obligations of physicians to their patients.

Source: From CMA POLICY (2008). Reprinted with permission of the Canadian Medical Association.

Canadian Medical Association: Caring in a Crisis

The recent experience of Canadian physicians during the SARS epidemic in Toronto has heightened the sensitivities of the medical profession to several issues that arose during the course of dealing with that illness. Many of the lessons learned (and the unanswered questions that arose) also apply to the looming threat of an avian flu (or other) pandemic. Canadian physicians may be in a relatively unique position to consider these issues given their experience and insight. The intent of this policy is to highlight the ethical issues of greatest concern to practising Canadian physicians which must be considered during a pandemic. In order to address these issues before they arise, the CMA presents this paper for consideration by individual physicians, physician organizations, governments, policy-makers, and interested bodies and stakeholders. Although many of the principles and concepts could readily be applied to other health care workers, the focus of this paper will be on physicians. Policies regarding physicians in training, including medical students and residents, should be clarified in advance by the relevant bodies involved in their oversight and training. Issues of concern would include the responsibilities of trainees to provide care during a pandemic and the potential effect of such an outbreak on their education and training. A. Physician Obligations during a Pandemic The professional obligations of physicians are well spelled out in the CMA Code of Ethics and other documents and publications and are not the main focus of this paper. However, they will be reviewed and discussed as follows. Several important principles of medical ethics will be of particular relevance in considering this issue. Physicians have an obligation to be beneficent to their patients and to consider what is in the patient's best interest. According to the first paragraph of the CMA Code of Ethics (2004), "Consider first the well-being of the patient." Traditionally physicians have also respected the principle of altruism, whereby they set aside concern for their own health and well-being in order to serve their patients. While this has often manifested itself primarily as long hours away from home and family and a benign neglect of personal health issues, at times more drastic sacrifices have been required. During previous pandemics, many physicians have served selflessly in the public interest, often at great risk to their own well-being. The principle of justice requires physicians to consider what is owed to whom and why, including what

resources are needed and how these resources would best be employed during a pandemic. These resources might include physician services but could also include access to vaccines and medications, as well as access to equipment such as ventilators or to a bed in the intensive care unit. According to paragraph 43 of the CMA Code of Ethics, physicians have an obligation to "Recognize the responsibility of physicians to promote equitable access to health care resources." In addition, physicians can reasonably b e expected to participate in the process of planning for a pandemic or other medical disaster. According to paragraph 42 of the CMA Code of Ethics, physicians should "Recognize the profession's responsibility to society in matters relating to public health, health education, environmental protection, legislation affecting the health and well-being of the community and the need for testimony at judicial proceedings." This responsibility could reasonably be seen to apply both to individual physicians as well as the bodies and organizations that represent them. Physicians also have an ethical obligation to recognize their limitations and the extent of the services they are able to provide. During a pandemic, physicians may be asked to assume roles or responsibilities with which they are not comfortable, nor prepared. Paragraph 15 of the CMA Code of Ethics reminds physicians to "Recognize your limitations and, when indicated, recommend or seek additional opinions or services." However, physicians have moral rights as well as obligations. The concept of personal autonomy allows physicians some discretion in determining where, how, and when they will practise medicine. They also have an obligation to safeguard their own health. As stated in paragraph 10 of the CMA Code of Ethics, physicians should "Promote and maintain your own health and well-being." The SARS epidemic has served t o reopen the ethical debate. Health care practitioners have been forced to reconsider their obligations during a pandemic, including whether they must provide care to all those in need regardless of the level of personal risk. As well, they have been re-examining the obligation of governments and others to provide reciprocal services to physicians and the relationship between these obligations. B. Reciprocal Obligations towards Physicians While there has been much debate historically (and especially more recently) about the ethical obligations of physicians towards their patients and society in general, the consideration of reciprocal obligations towards physicians is a relatively recent phenomenon.

uW

II

I

;ill

III W

4 Professional Ethics

During the SARS epidemic, a large number of Canadian physicians unselfishly volunteered to assist their colleagues in trying to bring the epidemic under control. They did so, in many cases, in spite of significant personal risk and with very little information about the nature of the illness, particularly early in the course of the outbreak. Retrospective analysis has cast significant doubt and concern on the amount of support and assistance provided to physicians during the crisis. Communication and infrastructure support was poor at best. Equipment was often lacking and not always up to standard when it was available. Psychological support and counselling was not readily available at the point of care, nor was financial compensation for those who missed work due to illness or quarantine. Although the Ontario government did provide retrospective compensation for many physicians whose practices were affected by the outbreak, the issue was addressed late and not at all in some cases. It is clear that Canadian physicians have learned greatly from this experience. The likelihood of individuals again volunteering "blindly" has been reduced to the point where it may never happen again. There are expectations that certain conditions and obligations will be met in order to optimize patient care and outcomes and to protect health care workers and their families. Because physicians and other health care providers will be expected to put themselves directly in harm's way and to bear a disproportionate burden of the personal hardships associated with a pandemic, the argument has been made that society has a reciprocal obligation to support and compensate these individuals. According to the University of Toronto Joint Centre for Bioethics report, We Stand on Guard for Thee, "[The substantive value of] reciprocity requires that society support those who face a disproportionate burden in protecting the public good, and take steps to minimize burdens as much as possible. Measures to protect the public good are likely to impose a disproportionate burden on health care workers, patients and their families." Therefore, in order to provide adequate care for patients, the reciprocal obligation to physicians requires providing some or all of the following:

•

•

During a Pandemic

• • •

• • •

•

• Prior to a Pandemic

•

Physicians and the organizations that represent them should be more involved in planning and decision-making at the local, national, and international levels. In turn, physicians and the organizations that represent them have an obligation to participate as well.

Physicians should be made aware of a clear plan for resource utilization, including: - how physicians will be relieved of duties after a certain time; - clearly defined roles and expectations, especially for those practising outside of their area of expertise; - vaccination/treatment plans—will physicians (and their families) have preferential access based on the need to keep caregivers healthy and on the job; - triage plans, including how the triage model might be altered and plans to inform the public of such. Physicians should have access to the best equipment needed and should be able to undergo extra training in its use if required. Politicians and leaders should provide reassurances that satisfy physicians that they will not be "conscripted" by legislation.

•

Physicians should have access to up-to-date, realtime information Physicians should be kept informed about developments in Canada and globally. Communication channels should be opened with other countries (e.g., Canada should participate in wHo initiatives to identify the threats before they arrive on our doorstep). Resources should be provided for backup and relief of physicians and health care workers. Arrangements should be made for timely provision of necessary equipment in an ongoing fashion. Physicians should be compensated for lost clinical earnings and to cover expenses such as lost wages, lost group earnings, overhead, medical care, medications, rehabilitative therapy, and other relevant expenses in case of quarantine, clinic cancellations, or illness (recognizing that determining exactly when or where an infection was acquired may be difficult). Families should receive financial compensation in the case of a physician family member who dies as a result of providing care during a pandemic. In the event that physicians may be called upon in a pandemic to practise outside of their area of expertise or outside their jurisdiction, they should contact their professional liability protection provider for information on their eligibility for protection in these circumstances. Interprovincial or national licensing programs should be developed to provide physicians with

Canadian Medical Association: Caring in a Crisis 191

•

•

•

backup and relief and ensure experts can move from place to place in a timely fashion without undue burden. Psychological and emotional counselling and support should be provided in a timely fashion for physicians, their staff, and family members. Accommodation (i.e., a place to stay) should be provided for physicians who have to travel to another locale to provide care or who don't want to go home and put their family at risk, when this is applicable, i.e., the epidemiology of the infectious disease causing the pandemic indicates substantially greater risk of acquiring infection in the health care setting than in the community. Billing and compensation arrangements should ensure physicians are properly compensated for the services they are providing, including those who may not have an active billing number in the province where the services are being provided.

since there is now some time to plan for the next pandemic and to ensure that reciprocal obligations can be met prior to its onset. Physicians have always provided care in emergency situations without questioning what they are owed. According to paragraph 18 of the CMA Code of Ethics, physicians should "Provide whatever appropriate assistance you can to any person with an urgent need for medical care." However, in situations where obligations can be anticipated and met in advance, it is reasonable to expect that they will be addressed. Whereas a physician who encounters an emergency situation at the site of a car crash will act without concern for personal gain or motivation, a physician caring for the same patient in an emergency department will rightly expect the availability of proper equipment and personnel. In order to ensure proper patient care and physician safety, and to ensure physicians are able to meet their professional obligations and standards, the reciprocal obligations outlined above should be addressed by the appropriate body or organization.

After a Pandemic

•

•

Physicians should receive assistance in restarting their practice (replacing staff, restocking overhead, communicating with patients, and any other costs related to restarting the practice). Physicians should receive ongoing psychological support and counselling as required.

C. How Are Physician Obligations and Reciprocal Obligations Related?

Beyond a simple statement of the various obligations, it is clear that there must be some link between these different obligations. This is particularly important

Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies Udo Schuklenk and Ricardo Smalling Introduction

It is not unusual for students in any given bioethics class to offer something like the following defence of conscientious objection rights: "Remember the Nazi

Conclusion

If patient and physician well-being is not optimized by clarifying the obligations of physicians and society prior to the next pandemic in spite of available time and resources necessary to do so, there are many who would call into question the ethical duty of physicians to provide care. However, the CMA believes that, in the very best and most honourable traditions of the medical profession, its members will provide care and compassion to those in need. We call on governments and society to assist us in optimizing this care for all Canadians.

experiments and the abuse of prisoners there and then? It is good that conscientious objection rights exist to protect good doctors refusing to participate in such crimes." There are more sensible versions of this argument, such as one published by an admittedly not "very courageous" doctor who wrote in the British Medical Journal that he would have hoped to object on conscience grounds to the abuse of prisoners in Stalinist Russia and claims that knowing that he was "part of an independent medical profession with allegiance to something higher and more enduring than the regime of the day" would have increased the odds of him doing the right thing.' We know, of course, in bioethics, that

Source: From Journal of Medical Ethics 43 (2016), 4. Reprinted with permission.

MI

192

I

I I Al

II I La

4 Professional Ethics

whenever a Nazi analogy is brought up to defend a particular normative stance, the odds are that that stance is weakly supported, if not outright indefensible. Unlike contemporary liberal democracies, the Nazis and Stalin's Soviet Union did not respect conscientious objectors and neither did Pol Pot's henchmen and henchwomen. Well-known pacifists such as Hermann Stohr, the leader of the German branch of an international pacifist organisation, perished in German concentration camps.' Conscientious objection typically does not flourish in dictatorships. Toleration of conscientious objectors in liberal democracies does little to support conscientious objectors in North Korea. The concept has evidently little currency there: its utility and legitimacy cannot be defended by pointing to Nazi Germany or Stalinist Russia. The same cannot be said for liberal democracies where respect for both individual as well as professional autonomy ranks highly. The medical historian John J. Michalczyk got it probably right when he noted that "those who invoke the Nazi analogy in a broad or general fashion are pressing the limits of valid analogy simply because the broader the scope of their reference, the harder it becomes to understand exactly what they think the Holocaust was, and thus why it is of moral relevance to the current issue."3 We aim to take the ethical debate about conscientious objection in medicine back to where it currently properly belongs, namely, liberal Western democracies where some medical doctors wish to see their private moral or religious objections to the provision of certain professional medical procedures accommodated by regulatory regimes. This is invariably at a cost to patients hoping to access medical services that they are legally entitled to access. Liberal democracies rightly do not take a stance on the substance of their citizens' moral or religious or other convictions; what is protected in liberal democracies' constitutions is the citizens' right to hold such beliefs and live by them—within reason. Typically, no distinction is and should be drawn between religious and other moral convictions, but that is not always the case.4'5 In the kinds of societies that we are concerned about in this paper, the vast majority of litigated cases are triggered by religious conscientious objectors as opposed to secularists or atheists.6 That in its own right is not a reason to disregard such complaints, because the protection of an individual's rights to adopt significant beliefs and live by them is at the heart of what living one's own life in a liberal polity is all about. For all practical intent and purposes, we are discussing predominantly religiously motivated conscientious objectors in the medical profession who ask that their objections to the delivery of particular professional services are protected by the secular state. There might

be instances of conscientious objection in other kinds of societies. There might also be conscientious objection in other contexts, for instance, in case of military conscripts. This article will not address those contexts. We are concerned only with conscientious objectors who decided to join a particular profession (in this case medicine) voluntarily and who then wish to be exempt from providing services that are typically expected of that profession. Conscientious Objection in Liberal Democracies: Why Should We Respect It?

Let us begin by trying to understand what conscience actually is. Daniel Sulmasy describes it as something that seems to operate both retrospectively as well as prospectively, it impacts on particular past actions, and it impacts on how we evaluate normatively possible future actions.' He conceptualises conscience both as our conviction that we should act in accordance with our individual understanding of what morality demands of us and on wilfully and voluntarily acting in accordance with what we consider to be morally good and right' It is uncontroversial that asking someone to violate their conscience-based convictions in matters that are of great importance to them is also asking such individuals to accept a potentially fairly high psychological cost. Typically, when we act contrary to our conscience, we will be plagued by guilt and possibly worse. Hermann Stohr died rather than recant his pacifist convictions. There are others like him. Why then should we, prima facie at least, tolerate conscientious objectors in liberal societies? There are various reasons that have been suggested for this. One is obvious from the above. Prima facie no society should force conscientious objectors to suffer the psychological cost that they would incur if they were forced to act against their conscience. Peter West-Oram and Alena Buyx have offered a number of other reasons. They write, "the right to exempt oneself from the fulfilment of a generally held duty is typically justified on the grounds that such a right is vital for the preservation of freedom of conscience. The latter is itself argued to be a core value of pluralist, liberaldemocratic states, and 'a moral right' Further, the rights to freedom of conscience and conscientious objection are argued to be constitutive of liberty and autonomy and to be necessary for the preservation of individual moral integrity. In promoting these goods, the rights are argued to be vital for the adequate toleration of different moral and philosophical perspectives in a pluralistic society"9 These are powerful reasons. On this reading, respect for an individual's conscience

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

is something of a corollary of the principle of respect for persons. Mark Wicclair has suggested in an influential paper that a blanket refusal to tolerate conscientious objection would constitute a significant threat to some doctors' moral integrity? Similar views are expressed by Pellegrino? It is not surprising then that there appears to be a broad consensus between religious voices in bioethics° and liberals° according to whom the failure to tolerate, protect, and respect some conscientious objectors is incompatible with what it means to live together in a liberal multicultural society. It is also fair to say that some bioethicists have voiced more or less strong opposition to the view that we should accommodate medical doctors' conscientious objections.14-18 As mentioned in the "Introduction" section, based on the review of hundreds of legal cases reported in Brian Leiter's book Why Tolerate Religion?, respect for conscience in the 21st century translates into a fairly one-sided affair: it is fought for and demanded by religious healthcare professionals without much regard for actual patient care and health outcomes or indeed respect for these patients' moral choices. Considering the significant power differential between patients and doctors, this is remarkable in its own right Whenever there is a conflict, on this account, a doctor's private ideological convictions generally take precedence in the professional practice of medicine. Various bioethicists have tried to suggest limiting criteria to avoid the possibility that arbitrary stances taken by doctors could pass successfully the conscience muster." Typically, conscientious objections occur in the context of reproductive health (e.g., abortion, in vitro fertilisation, contraceptives), end-of-life care (e.g., assisted dying), and also affect negatively gays, lesbians, and patients with gender dysphoria. Why Conscientious Objection Has No Place in the Practice of Medicine

It is worth noting that scepticism about the importance of an individual's conscience claims within a community has been expressed by influential thinkers for many years. Thomas Hobbes had this to say in his Leviathan: another doctrine repugnant to civil society is that whatsoever a man does against his conscience is sin; and it dependeth on the presumption of making himself judge of good and evil. For a man's conscience and his judgement is the same thing; and as the judgement, so also the conscience may be erroneous. Therefore, though he that is subject to no civil law sinneth in all he does against his conscience, because he has no other rule to

follow but his own reason, yet it is not so with him that lives in a Commonwealth, because the law is the public conscience by which he hath already undertaken to be guided. Otherwise in such diversity as there is of private consciences, which are but private opinions, the Commonwealth must needs be distracted, and no man dare to obey the sovereign power farther than it shall seem good in his own eyes.18 Poignantly, Hobbes makes this point in a chapter on "those things that weaken or tend to the dissolution of a commonwealth." Of course, Hobbes is not quite our archetypical defender of liberal democracies, but the point he is making here is valid; it applies to the case of conscientious objectors in liberal democracies, too. Let us begin our argument by making the general case for why conscientious objection in medicine should not be accommodated. We recognise that conscientiously objecting healthcare professionals can have various rationales to support their opposition to the participation of doctors in particular medical procedures. They include typically a recourse to tradition, the Hippocratic Oath, the Bible, the Quran, and any number of other documents that have no legitimate bearing on the practice of 21st century medicine. It is easy to show that even the objecting doctors' standard recourse to the Hippocratic Oath when it comes to abortion and assisted dying is not always credible.19,2° It is evidently the case that other ideological convictions held by particular doctors motivate their deployment of the argument from tradition, in the guise of the Hippocratic Oath, because only the lines from the Oath that suit a particular objecting doctor's interests are typically cited, while those they disagree with are quietly ignored. But in any case, the Oath is not a defensible ethical guide to modern medical practice." Mercier might be right in this context when she writes that religion is all about believing that one's beliefs are right, but not about having right beliefs. If firstorder religious beliefs had content, their content could be checked against the truth. It is precisely because such beliefs lack content that one can go on about believing that one believes them despite any and every evidence. But the price of second-order belief in vacant first-order beliefs is self-deception." On this background, it is fortunate from the conscientious objector's perspective that secular liberal democracies do not typically test whether the views

II I

I I II

111111

4 Professional Ethics

conscientious objectors profess to subscribe to are defensible. What matters is that they are deeply held or, more to the point, that the conscientious objectors claim that they hold those convictions deeply.23 Even on this count we are incoherent. A female Muslim doctor refusing to see a male patient would not be granted a conscientious objection exemption, whereas a pharmacist refusing to sell contraceptives in some countries might. Let us assume that both of these conscience decisions were loudly proclaimed to be genuinely held by the professionals in question. Why should one be respected and the other one not? It could be argued that in liberal democracies, constitutional documents and human rights legislation variously protect against gender-based discrimination but may not guarantee access to contraception. While this may be true, it misses the point of the argument—that the substance of conscience-based objections that are protected is arbitrary. Take as an example the unfolding debates around marriage equality. Until quite recently this was a concept alien to liberal constitutions and human rights legislations. Those with conscience-based objections to same sex marriage were free to assert those beliefs against such couples. However, that is increasingly no longer the case. Nothing has changed about the same sex couples or the beliefs of the objectors; society has simply decided that such objections will no longer be tolerated. Furthermore, just as we cannot test the plausibility of the ideological dicta that lead to conscientious objections (there is no test for the existence of "God," for instance, or truth of the Bible), it is also impossible to ascertain whether conscientious objectors actually hold the views they profess to hold. The US Supreme Court writes on this issue, What principle of law or logic can be brought to bear to contradict a believer's assertion that a particular act is "central" to his personal faith? Judging the centrality of different religious practices is akin to the unacceptable "business of evaluating the relative merits of differing religious claims" . . . it is not within the judicial ken to question the centrality of particular beliefs or practices to a faith, or the validity of particular litigants' interpretation of those creeds . . . courts must not presume to determine the place of a particular belief in a religion or the plausibility of a religious claim.24 If that is the case, it remains unclear why untestable conscience claims from privileged professionals who voluntarily chose to join a particular profession, and who have been endowed by society with a monopoly

on the provision of particular procedures, should be accommodated, given that this toleration subverts the very objectives the profession is designed to achieve. This does not deny anyone the right to hold any number of private religious and moral views, as they see fit and as they choose to hold. What we are denying is that professionals are entitled to subvert the objectives of the professions they voluntarily joined by prioritising their private beliefs over the professional delivery of services to the public, especially when they are monopoly purveyors of these services. Legal scholar Alta Charo called it right, when she wrote, "claiming an unfettered right to personal autonomy while holding monopolistic control over a public good constitutes an abuse of the public trust—all the worse if it is not in fact a personal act of conscience but, rather, an attempt at cultural conquest."25 Various authors have warned that an unqualified right to conscientious objection in medicine would result in harmful consequences, for instance, for women wanting to have an abortion, who could be denied timely access to this healthcare procedure not because of legal constraints but because of healthcare professionals refusing to provide such a healthcare service on conscience grounds. Frequently, a middleground is sought, as legal systems are indeed rare where conscientious objection rights cannot be overridden by a competing stronger right (e.g., a patient's right to access healthcare in a timely manner). Typically, then, the argument is not over whether conscience rights are defensible but about what kind of compromise is reasonable for everyone affected.26,27 Unfettered versus Fettered Conscientious Objection Accommodation

It could be objected that the argument advanced in this paper is too simplistic as it appears to target only unfettered conscientious accommodation stances, when the focus in today's policy debates is about fettered conscientious objection accommodation. It may be argued that the question really is: where and how could society draw reasonable lines that take into account societal interests as well as those of individual objectors? The preceding section explains to some extent why we have taken this stance. The courts in various jurisdictions have already, rightly so, conceded that it is a moot point trying to establish the truth or even plausibility of the views purportedly held by conscientious objectors. That is not an insignificant point, because this concession opens the door to any number of more or less arbitrary and random conscientious objection claims. For policy-makers aiming to establish a functioning healthcare system with predictable service delivery and

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

guaranteed service levels to the people who finance the system, this constitutes an insurmountable problem. It is nigh impossible to predict which healthcare professional, in which part of the system, will demand accommodation for which kinds of purported or real convictions. It is also evidently impossible to verify whether objecting healthcare professionals even hold the views they profess to hold. Such claims may merely be a convenient way out of the provision of inconvenient healthcare services. In light of this, it seems to us that whatever attempt at a principled fettered accommodation is made, it is by necessity resulting in arbitrary outcomes. Regardless of the accommodations that we make today, these can and likely will be challenged time and again by new generations of conscientious objectors. Today it might be abortion and assisted dying, tomorrow it might be the use of the tools of personalised medicine or something else altogether. Limiting conscientious objection accommodation to defensible claims seems impossible to us, unless we overcome the two problems mentioned: demonstrate the truth of the foundations of the conscientious objection and demonstrate evidence that objectors actually genuinely hold the views they claim to hold. Failing that, as we will show, the inevitably ensuing arbitrary accommodation demands will have harmful real-world consequences as far as healthcare outcomes and patient access to care is concerned. A Few Canadian Examples

Lobbyists for doctors usually defend the view that doctors must never be required to provide services that they object to on conscience grounds. Their focus is on ensuring that conscientious objectors do not have to oblige patients asking for such healthcare services or even have to assist them in finding a healthcare provider who will oblige them. A good example of this is Dr Jeff Blackmer, the Director of the Canadian Medical Association's Ethics Office. Dr Blackmer writes on behalf of the Canadian Medical Association that medical doctors are neither obliged to provide abortion services, nor are they obliged to transfer patients on to doctors they know will provide abortions to women seeking one.28 Women are legally entitled to access abortions in Canada if they so wish, and typically abortions are fully funded by provincial public healthcare systems. Canadian doctors are the only professionals legally entitled to provide such procedures. They are monopoly providers. Dr Blackmer's considered view on this subject matter is that pregnant women who wish to have access to a medical procedure that they are legally entitled to, and that is fully funded by the state, should

have to depend on the goodwill of volunteering doctors who also happen to be the only licensed providers of this procedure in the country. A second example: At the time of writing, Canadian legislators grapple with the question of how to implement a Supreme Court of Canada judgement that effectively decriminalises providing assistance in dying to competent patients suffering from an irreversible chronic illness that renders their lives not worth living in their own considered view.2° Dr Cindy Ford, the current president of the Canadian Medical Association, mirrors Dr Blackmer's stance on the obligations of conscientious objectors. In a statement made to the Canadian parliamentary committee tasked with investigating possible legislative options on assisted dying, Dr Ford reportedly stated that it is unreasonable to require conscientious objectors to even refer such patients to a clinician who they know will be prepared to render assistance in dying.3° Dr Blackmer agrees with her, while at the same time trying to assure the Canadian parliamentary committee members of reliable professional service delivery; as he says, "I can sit here . . . and guarantee that from simply a number perspective, access will not be an issue."31 Dr Blackmer's argues that given that 30% of Canadian doctors are willing to provide assistance in dying, there should not be a problem for patients wanting to access such services. If it was just a numbers game, Dr Blackmer would certainly be correct. Given the relatively small number of eligible patients likely to ask for assistance in dying, there should be a sufficient number of doctors willing to provide assistance to such patients. However, as so often, the devil is in the detail. The problem is— we suspect this is the reason why Dr Blackmer does not actually offer unconditionally guaranteed access on behalf of his association—that this likely will not help seriously sick and vulnerable patients who happen to live in remote areas of Canada where only few doctors happen to practise and where all of them might be opposed to assisted dying on conscience grounds. It is irrelevant then that elsewhere in the country plenty of doctors would be willing to provide assistance in dying on request if a particular patient is unable to travel to them. To these patients, Dr Blackmer's "guarantee" is worth very little. We should note here that the Canadian Medical Associations stance is not unusually radical among its peers. The British Medical Association, too, produced a guideline stipulating that "whenever possible physicians who are conscientiously opposed to forgoing treatment should be permitted to hand over care of the patient to a colleague," but they are not obliged to do so.32 The entitlement mentality that drives these

195

II

196

,d11

1111.1.6/

4 Professional Ethics

associations' stances is unsurprisingly supported by the doctors they represent.33 A further example of the potential impact of conscientious objection on patients' access to care in Canada can be seen in the recent application filed by various religious physicians and associations representing them in the Ontario Superior Court. The application challenges the College of Physicians and Surgeons of Ontario (cPSo) Policy Statement #2-15: Professional Obligations and Human Rights, which among other things requires that physicians "provide a referral to another appropriate health-care provider for the elements of care the physician is unable to manage directly."34 The applicants, including the Christian Medical and Dental Society of Canada and the Canadian Federation of Catholic Physicians' Societies, described the referral provisions of the CPSO policy statement as "unconscionable" and a violation of their freedom of conscience and/or freedom of religion.35 This example is hardly unique.36 These kinds of hardline positions disregard the needs of patients. Statements and actions of this kind strongly suggest that Professor Charo was not attacking strawmen in her article. Respect for Private Conscience Choices Is Already Limited

Regardless of the views espoused by representatives of the Canadian Medical Association and the religious doctors' groups mentioned, there are already limits placed on the conscience choices of Canadians in various settings. The Supreme Court of Canada stated that "freedom of conscience and religion protects the right to entertain beliefs, to declare them openly and to manifest them, while at the same time guaranteeing that no person can be compelled to adhere directly or indirectly to a particular religion or to act in a manner contrary to his or her beliefs."3' In that same case, the Court went on to say that the "state's duty to protect every person's freedom of conscience and religion means that it may not use its powers in such a way as to promote the participation of certain believers or nonbelievers in public life to the detriment of others."38 These principles are seen as cornerstones of maintaining a free and democratic society.39 This is not to say that physicians and other medical professionals should be denied the right to hold religious or conscience beliefs. In liberal democracies, a diversity of beliefs and opinions is welcomed, and individuals are free to advocate for societal acceptance of their particular world view. It is also trite to say that the law in liberal democratic societies may protect a particular social or moral position espoused by a particular

religion; however, it should not do so on the basis that it is a religious position but on the basis that "in reason its merits commend themselves."4° Therefore, in order to maintain a defensible balance between competing beliefs in a liberal and democratic society, such societies have generally considered "the practice of religion and the choices it implies to relate more to individuals' private lives or to voluntary associations."' To this end, it has been recognised that within the public realm an individual's freedom of religion and conscience may be legitimately burdened. In another case, the Supreme Court of Canada explicitly acknowledged that legislative or administrative actions may increase the cost of practising or otherwise manifesting one's religious beliefs.42 Further, such legislative or administrative burdens may be justified where they prevent conduct that would potentially cause harm to or cause interference with the rights of others.43 The reasoning of the Court supports the view that the conscientious belief of a physician may be legitimately halted at the point that it interferes with the rights of patients to access services they are entitled to receive. That interference, however, is precisely what happens each time a conscientious objector is accommodated and the patients are unable to access the required service elsewhere. Respect for Private Conscience Choices Will Result in Avoidable Suboptimal Access to Healthcare

Any society that grants medical professionals a conscientious objection-based opt-out will have to accept suboptimal health outcomes with regard to procedures that are considered at that point in time part and parcel of good professional healthcare practice in those societies. Why would toleration of conscientious objection lead invariably to suboptimal health outcomes? Take the Canadian assisted dying example. It is apparent that the representatives of Canadian doctors are unconcerned about the distress they would inflict on—often—dying patients in rural areas who cannot access easily a doctor willing to provide assistance in dying. In many cases, doctors' objection could be more than an inconvenience., it could be an insurmountable barrier to access. A case in point is the small Canadian province of Prince Edward Island (also known as PEI). Women in that province, at the time of writing, are required to leave the province if they wish to have an abortion because healthcare providers are not offering this service in the province. Empirical evidence elsewhere suggests strongly that "travel is a barrier to accessing legal abortion."44 The situation in PEI is not unique in the world, far from it. Minerva reported recently that some 70% of Italian gynaecologists conscientiously object to performing abortions,

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

which is arguably one of the reasons for staggeringly high backstreet abortion rates in that country.45 Patient interests come clearly last here. As we have seen, the representatives of the monopoly providers of these healthcare services in Canada are quite content with that outcome, as long as the association's members' consciences—or, more precisely, their untestable claims about the content of their consciences—are not burdened or otherwise inconvenienced. This attitude is quite the opposite of what it means to be a professional, where a promise is made to serve the public good and to serve patient interests first and foremost. Apparently such central features of what it means to be a professional can readily be sacrificed when arbitrary, untestable individual conscience claims are mounted by professionals. Respect for Private Conscience Choices Will Result in Avoidable Inequitable Workloads for Doctors

It seems also uncontroversially true to us that toleration of conscientious objectors will have a detrimental impact on healthcare practice in other respects. Because patients can expect to be shepherded around among different healthcare professionals choosing "conscientiously" to provide different levels of healthcare, there will be an increasing level of upheaval visited upon healthcare systems. Worse, if the Canadian Medical Association had its way, patients would not even be shepherded through the system, it would be up to them to find doctors obliging them for particular ailments that might be of concern to doctors with any number of idiosyncratic private views of the universe and on what may or may not be ethically acceptable. An important consequence of such a situation would be that doctors who are willing to provide services conscientiously refused by some or many of their colleagues would have to carry an inequitable load of such work. We cannot think of a good reason for why they ought to accept such an inequitable burden and, indeed, why a just society should design a health delivery system that burdens such doctors unfairly. Respect for Private Conscience Choices Will Result in Unpredictable and Unfair Service Delivery

Because of the unpredictable nature of what it is that conscientious objectors will object to, patients can never be quite certain about the kinds of services that they will be able to receive from a particular doctor, even their long-time family doctor. That is the opposite of what can reasonably be expected of a profession and professionals. Uniform service standards are arguably one of the hallmarks of what constitutes a

profession. Another aspect of this has to do with the unfairness of this situation vis-à-vis the patient. Patients, as Ruth Purtilo rightly notes, "can expect to be treated fairly. Persons seeking treatment should not be given advantage on the basis of arbitrary favouritism or be left out on the basis of arbitrary dislike."46 Given the intractability of conscience claims, it is not unwarranted to characterise them as essentially arbitrary dislikes. They might not be arbitrary in the eyes of the objector, but we cannot even be certain of that, given our inability to test the objector's conscience claims. They are arbitrary with regard to what kinds of services particular patients can or cannot expect from particular professionals belonging to the same profession in the same jurisdiction. The nature of professions is that they provide privileged services to the public. They are quite unlike political parties, churches, animal welfare, or environmental organisations. While the latter cater to particular sectarian audiences, professionals must cater uniformly to everyone within the scope of professional practice. The services professionals provide to the public must not be subject to their private normative judgements about individual needs of members of the public, as long as what these members of the public require falls within the scope of professional practice. After all, patients do not seek out doctors for their individual convictions but to receive professional services that they are trained to provide. There are other sounds reasons for refusing to tolerate conscientious objectors in medicine. They have to do with the voluntary choices made by those who decided to join the profession. Choosing to Join a Profession Is a Voluntary Activity Undertaken by an Autonomous Adult

It seems to us that the argument from voluntariness is important in this context. Nobody is forced to join a particular profession, medicine included. Pace Christopher Cowley's argument that the practice of medicine for some is a "calling"—one can still choose not to answer the "call."47 The decision to join the medical profession is quite distinct to, say, joining the military as a conscript. Medical schools the world all over reject most applicants, not because the applicants are unsuitable to become doctors but because the number of places they have available is always much smaller than the number of people who apply. Anyone joining a profession knows, or should know, that it is ultimately up to society to determine the scope of professional practice. Society can do this through professional bodies, but it can also choose to do so by other regulatory means. Having a monopoly on the provision

II

.41

111111

4 Professional Ethics

of services within that scope of practice is not something that comes cost-neutral to those who choose voluntarily to join a particular profession. Indeed, given the high-stakes competition to join the medical profession in the first place, it is reasonable to suggest that doctors refusing to provide professional services that are within the scope of practice should be replaced by someone who is willing to undertake the work. If at any given time a doctor is unable to continue practising due to their—ultimately arbitrary—conscience views, nothing would stop them from leaving the profession and taking up a different vocation. This happens across industries and professions very frequently. Professionals can be expected to take responsibility for the voluntary choices they make. Does the Nature of a Particular Procedure Matter?

It has been suggested that not all conscientious objections fall into the same category. The reason why it would not be acceptable for the female Muslim doctor in our example to refuse to see a male patient had to do with that being in violation of the core values in medicine. Assisted dying or abortion, on the other hand, would then be seen if not in violation of traditional medical values but at least being sufficiently controversial to justify the accommodation of conscientious objectors. Wicclair argues for one possible threshold that a conscientious objection must meet in order to be prima facie worthy of accommodation. He proposes that a conscientious objection "has significant moral weight only if the core ethical values on which it is based correspond to one or more core values in medicine."48 Wicclair is, therefore, supportive of conscientious objection to abortion or assisted dying because to his mind neither of these medical procedures corresponds uncontroversially to these core values. We think that he is mistaken about what he considers to be the core values of medicine, which would be applicable in these two instances, but we do not wish to debate the merits of this claim here. Rather, his argument also fails because despite his protestations to the contrary it is essentially an argument from tradition. What he calls core values could be more aptly described as traditional values. However, it is in the nature of such values that they are changing. Medicine is currently arguably undergoing a paradigm shift from maintenance of human life at nearly all cost to a quality-oflife focused approach.49 The traditional core values of medical practice are in flux, and that is one reason for why abortion is available in most liberal Western democracies, and that is the reason why the number of jurisdictions that are decriminalising assisted dying is steadily increasing.

To highlight just one example of this, again from Canada, the past president of the Canadian Medical Association, Dr Chris Simpson, on behalf of the association, referred to assisted dying as a "therapeutic service," a mere 15 years after Wicclair suggested that core professional values would make such a description impossible.8° It is unlikely that the association would refer to assisted dying as a "therapeutic service" if it thought that the provision of this service would violate Wicclair's claimed core values of the profession. If it is a therapeutic service, as we agree it is, Canadian doctors wanting to go the conscientious objector route to avoid providing this service to eligible patients could no longer avail themselves of Wicclair's analysis, because the "moral integrity of the medical profession" would not be served by accommodating individual doctors who refuse to provide therapeutic services. It is important to recognise that medical practice is also a cultural practice that changes over time. No democratic society and no medical profession should leave the scope of the provision of professional medical services to the vagaries of its members' personal ideological convictions. What about Our Societal Ability to Recruit People to Join the Medical Profession?

An argument could be mounted that if we prevented doctors from excusing themselves from providing professional services that they happen to feel strongly about, we might find it difficult to find a sufficient number of people willing to become doctors. This seems far-fetched to us, but it is conceivable that some people might reconsider joining the medical profession if the accommodation of their feelings with regard to particular professional services is not guaranteed. We indicated already that the profession and society would likely be better off if such people chose not to join the profession, seeing that they think that their idiosyncratic views of the universe should be of greater significance than the patients whose interests they promised to further first and foremost as professionals. The concern about recruitment problems does not appear to be based on real-world evidence to begin with. A recent survey reports that countries such as Sweden, Finland, and Iceland, which do not permit their doctors to opt out of the provision of abortion service on grounds of conscience, have not experienced harmful consequences as far as these countries' healthcare systems' ability to train a sufficient number of doctors is concerned." A somewhat related argument suggests that by preventing doctors from refusing service based on religious or conscience grounds, we would run the

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

risk of losing the participation of those kinds of individuals in the profession. However, that argument does not hold. Those who object to particular procedures could choose specialities that would not require that they violate their conscience; for example, they could opt for dermatology instead of gynaecology if they are opposed to abortion. A self-imposed limit on one's choice of speciality is not inherently unfair. Individuals with certain physical limitations can also be excluded from certain specialities in the medical profession. That does not preclude them from joining the medical profession. Does It Matter When Professionals Joined a Profession?

It could be argued that it matters when conscientious objectors joined the profession. The idea here is that they agreed to join a profession that at that time had a particular scope of practice that these professionals agreed to work in. In a way, this scope of practice constitutes a kind of contract between society and its professionals. If a society wishes to change a profession's scope of practice, it should accommodate conscientious objectors among those who joined the profession prior to a given controversial change, for example, by grandmothering them in. Some of the reasons we mentioned above mitigate against granting objectors a blanket right to such an accommodation. There is another reason that suggests that this argument is not particularly strong. It is surprising that doctors seem to think in large numbers that society provided them with a monopoly on the provision of medical services but that society at no time would add or subtract from the range of services it considers scope of practice of that profession. This is puzzling given the nature of medicine as an everevolving scientific and cultural pursuit. Most of us who do not work as medical professionals have encountered employers who wish to change our scope of practice. Many academics the world all over are suffering under their employers' desire and pressure to introduce online learning programmes, despite our objections on pedagogical and other grounds. However, nobody would think that our employers are not well within their rights to ask us to contribute to such programmes. Unlike doctors, we are not even entrusted with typically well-remunerated service delivery monopolies. Having said this, we do think that we should try at least to accommodate conscientious objectors who have practised for a long term under a different kind of contract than doctors who join the profession after particular significant changes to the scope

of practice have been made. However, that is a far cry from suggesting that doctors have a right to such accommodation. Could Not the State Be the Guarantor of Access?

Proponents of a fettered accommodation system might argue that we should try to accommodate conscientiously objecting healthcare professionals regardless of the arguments presented by us so far. They could argue that, for instance, institutions other than the medical profession ought to guarantee reliable access in a healthcare system that is proactively being subverted by accommodation-demanding conscientious objectors. Why should not the health ministries in Canada's provinces organise reliable access systems based on doctors who are not conscientiously objecting? We know from the mentioned experiences of pregnant women seeking an abortion in the province of PEI that this might not be feasible in parts of the country. The odds are fairly high that the more rural an assistance-seeking patient is located, the more difficult it could become to guarantee access. We have yet to see a persuasive reason, given the arguments presented so far, why any healthcare system should burden itself, and ultimately patients, with these sorts of logistical problems when there is an obvious, more efficient alternative: saying no to the conscience-based accommodation requests of healthcare professionals. A profession that is unable to guarantee reliable access to its services for no other reasons than its desire to accommodate the private ideological convictions of some or many of its monopoly provider members is failing in its mission. Conclusion

Medical professionals have no moral claim in liberal democratic societies to the accommodation of their individual conscientious objections. To accommodate such objections would subvert some of the very reasons for why the medical profession was created in the first place. To accommodate them would also permit such medical professionals to abuse the monopoly privileges that society endowed their profession with. Medical professionals practising medicine in the 21st century would be well advised to accept professional service delivery, as defined by the scope of professional practice, as one inevitable corollary of their voluntary decision to join the profession. Forcing patients to live by the conscientious objectors' values constitutes an unacceptable infringement on the rights of patients.

MU I

4 Professional Ethics

References 1. Smith VP. Conscientious objection in medicine: doctors' Amselem: Syndicat Northcrest v. Amselem, [2004] 2 SCR freedom of conscience. BMJ 2006;332:425. 551, 2004 SCC 47. 2. Braun H. Hermann Stohr (1898-1940). In: Hummel 24. Smith: Employment Division, Department of Human K-J, Strohm C, eds. Zeugen einer besseren Welt. ChristResources of Oregon v. Smith, [1990] 494 US 872. liche Martyrer des 20. Jahrhunderts. Evangelische Ver- 25. Charo RA. The celestial fire of conscience-refusing to deliver medical care. N Engl J Med 2005;352:2471-3. lagsanstalt. Leipzig: Butzon & Bercker, 2000:87-105. 3. Michalczykl]. Medicine and ethics in the Third Reich: historical 26. Cantor], Baum K. The limits of conscientious objection: and contemporary issues. NewYork: Sheed and Ward, 1994:8. may pharmacists refuse to fill prescriptions for emer4. Weinstock D. Conscientious refusal and healthcare gency contraception? N Engl J Med 2004;351:2008-12. professionals: does religion make a difference? Bioethics 27. Lynch HE Conflicts of conscience in health care: an institutional compromise. Cambridge MA: MIT Press, 2008. 2014;28:8-15. 5. Leiter B. Why tolerate religion? Princeton: Princeton 28. Blackmer J. Clarification of the CMA's position concernUniversity Press, 2013. ing induced abortion. CMAJ 2007;176:1310. 6. Laycock D. Regulatory exemptions of religious behavior 29. Carter: Carter v. Canada (Attorney General), [2015] 1 SCR 331, 2015 SCC 5. and the original understanding of the establishment clause. Notre Dame Law Review 2006;81:1739-842, at 1839 for 30. Zimonjik P. Doctors with moral objections to assisted dying should be able to opt out: committee hears. CBC the USA. Strasbourg Consortium for European Court of Human Rights cases at http://www.strasbourgconsortium. January 27, 2016. http://www.cbc.ca/news/politics/assisted-dying-committee-evidence-hearings-1.3422352 org/. We owe these references to Leiter B 2013, n4. (accessed 28 Jan 2016). 7. Sulmasy DP. What is conscience, and why is respect for 31. Zimonjik P 2016. it so important? Theo Med Bioeth 2008;29:135-49. 32. As quoted in Wicclair 2000, 218. 8. Sulmasy DP 2008, at 139. 9. West-Oram P, Buyx A. Conscientious objection in 33. Lawrence RE, Curlin FA. Physicians' beliefs about conscience in medicine: a national survey. Acad Med health care provision: a new dimension. Bioethics 2015. 2009;84:1276-82. dot 10.1111/bioe.12236. 10. Wicclair MR. Conscientious objection in medicine. Bio- 34. College of Physicians and Surgeons of Ontario (CPSO). Policy Statement #2-15: Professional Obligations and Human ethics 2000;14:205-27. Rights. March 2015. http://www.cpso.on.ca/CPSO/media/ 11. Pellegrino E. The physician's conscience, conscience documents/Policies/Policy-Items/Human-Rights clauses, and religious belief: a Catholic perspective. Fordham Urban Law J 2002;30:221-44. .pdf?ext=.pdf (accessed 31 Jan 2016). 12. E.g., Pellegrino E 2002. 35. End-of-Life Law and Policy in Canada (EOL). De13. E.g., Weinstock D 2014. velopments to watch-Christian Medical and Dental 14. LaFollette E, LaFollette H. Private conscience, public Society of Canada v. College of Physicians and Surgeons of Ontario. 2015. http://eol.law.dal.ca/wp-content/ acts. J Med Ethics 2007;33:249-54. uploads/2015/06/2015_03_20_Notice_of Application-3 15. Savulescu J. Conscientious objection in medicine. BMJ .pdf (accessed 31 Jan 2016). 2006;332:294-7. 16. Schuklenk U. Conscientious objection in medicine: pri- 36. Laycock D. How the Little Sisters of the Poor case puts religious liberty at risk. Washington Post 2015, March 20. vate ideological convictions must not supersede public https://www.washingtonpost.com/opinions/how-theservice obligations. Bioethics 2015;29:ii-iii. little-sisters-of-the-poor-put-religious-liberty-at-risk/2016/ 17. E.g., Wicclair MR 2000, Sulmasy DP 2008. 03/20/eaaa6a34-e4b4-11e5-a6f3-21ccdbc5f74e_story. 18. Hobbes T. Leviathan (chapter 29). London: Andrew html (accessed 21 Mar 2016). Crooke, 1651. https://ebooks. adelaide.edu.au/h/hobbes/ thomas/h681/chapter29.html (accessed 26 Feb 2016). 37. Saguenay: Mouvement laique quebecois v. Saguenay 19. Donahue BS. Infant euthanasia is morally unacceptable. (City), 2015 SCC 16, paragraph 69. 38. Saguenay, paragraph 76. J Thorac Cardiovasc Surg 2015;149:1684-5. 20. Schuklenk U. The case against assisted euthanasia has not 39. McFarlane: McFarlane v. Relate Avon Ltd [2010] EWCA Civ 880, paragraph 21. Saguenay, paragraph 75. been made. J Thorac Cardiovascr Sur 2015;149:1685-6. 21. Veatch RM. Hippocratic, religious, and secular medical 40. McFarlane, paragraph 21. ethics: the points of conflict. Washington DC: Georgetown 41. Lafontaine: Congregation des temoins de Jehovah de StUniversity Press, 2012. Jet-time-Lafontaine v. Lafontaine (Village), 2004 SCC 48, 22. Mercier A. Religious belief and self-deception. In: Blackparagraph 67. ford R, Schuklenk U, eds. 50 Voices of disbelief: why we 42. Edwards Books: R. v. Edwards Books and Art Ltd., [1986] 2 SCR 713, 1986 CanLII 12 (SCC), paragraph 97. are atheists. Oxford: Wiley, 2009:41-7, at 47. 23. Smith: Employment Division, Department of Human 43. Amselem: Syndicat Northcrest v. Amselem, [2004] 2 SCR 551, 2004 SCC 47, paragraph 62. Resources of Oregon v. Smith, [1990] 494 US 872.

Schuklenk/Smalling: Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation

ST, Ridell J, Brown A, et aL Characteristics of women who present for abortion towards the end of the mid-trimester in Scotland: national audit 20132014. Eur J Contracept Reprod Health Care 2015;1-6. doi:10.3109/ 13625187.2015.1111326. Minerva P. Conscientious objection in Italy. J Med Ethics 2015;41:170-3. Purtilo RB. Professional-patient relationship: ethical issues. In: Jennings B, ed. Bioethics. 4th edn. Farmington Hill, MI: Macmillan, 2014:2547-55, at 2551. Cowley C . A defense o f conscientious objection in medicine: a reply to Schuklenk and Savulescu. Bioethics 2015. doi:10.1111/bioe.12233. Wicclair 2000, at 205.

44. Cameron

45. 46.

47.

48.

The Functions and Limitations of Professional Codes of Ethics Dale Beyerstein

Many people become cynical about codes of ethics because they suspect that such codes are of no help in resolving moral dilemmas. In fact, some cynics take delight in pointing out holes in their own professional codes that are large enough to drive a truck through, thus making them next to worthless for providing moral guidance. Other cynics speculate on the reason for this that codes of ethics exist primarily to make professionals look moral, not to promote morality in the profession—in other words, to allow professionals to protect themselves, either individually or collectively. It is not my purpose here to either affirm or deny this claim. As far as I can see, it is perfectly possible for professions to design codes of ethics for this purpose, just as it is possible for individual professionals to use the codes of their professions for this purpose. But it is equally possible for whole professions and individuals to use codes for other, more respectable ends, and it is these ends which I wish to explore here. The point of this paper, then, is not to examine how codes of ethics are in fact used but rather to say a few things about how they might be used. My hope is that this will shed some light on the question whether there is any point in a profession's having a code of ethics and in a professional's ever bothering to consult one. The argument of this paper is twofold: the first point is that the intuition behind the cynicism I have alluded to is perfectly correct—that is, codes of ethics cannot resolve moral dilemmas. The second is that it does not matter very much that codes cannot fulfill this function. Codes of ethics have other purposes

49. Singer P. Rethinking life and death. Text: Melbourne, 1994. 50. CBC. Doctor-assisted suicide a "therapeutic service," says Canadian Medical Association. February 6, 2015. hap:// www. cb c.c a/new s/health/doc t or-a s si s ted- suicide-a therapeutic-service-says-canadian-medical-association1. 2947779 (accessed 30 Jan 2016). 51. Fiala C, Gemzell Danielsson K, Heikinheimo 0, et al. Yes we can! Successful examples of disallowing "conscientious objection" in reproductive health care. EurJ Contracept Reprod Health Care 2016;1-6. do i:10.3109/13625187.2016.1138458. Schuklenk U, Smalling R, 4. Law, ethics and medicine. J Med Ethics 2016;0: 1-7. doi: 10 .1136/medethics-2016-103560.

than to resolve moral dilemmas, and we have other (and better) means besides codes of ethics to resolve moral dilemmas. Resolving moral dilemmas requires the application of a moral theory which is more general than the moral theory which informs the code of ethics of a particular profession. So the cynicism implied in the above criticism of codes of ethics is misplaced: we simply should not expect the code of ethics to resolve moral dilemmas. A second sort of cynicism I wish to address holds that moral dilemmas are irresolvable. The usual argument for this view rests upon the correct observation that codes of ethics do not resolve moral dilemmas but leaps from this premise to the unwarranted conclusion that moral dilemmas are not capable of rational resolution. The reason that this inference is unwarranted is that it ignores the fact that moral theory can be invoked in those cases where codes provide no guidance. Fortunately it is not my job here to actually resolve any moral dilemmas. My aim is much less ambitious: to simply direct you to the right place and let you take it from there. I shall begin by distinguishing professional codes from moral theories. The Oxford English Dictionary defines "code" in the sense we are interested in as "a systematic collection or digest of the laws of a country, or of those relating to a particular subject." Two points in this definition are important for our present purposes. First is the emphasis on "systematic": a good code of ethics will present its injunctions in a systematic manner. However, note the second point: the connection with the laws of a country. The laws of the land merely assert what is to be done or, more usually, what must not be done. They do not explain why this is so or provide the reasons behind these pronouncements. Thus, if the legal

Source: FromApplied Ethics: A Reader, ed. E.R Winkler and J.R Coombs (Blackwell, 1993). Reprinted with permission.

II

all

111111

4 Professional Ethics

system is systematic, a mere inspection of the code is not sufficient to reveal this. It is only by examining the code in the light of legal and moral theory that we can judge the code to be adequate or inadequate. A code of ethics is essentially a societal code "writ small"—that is, one that governs the conduct of a subset of the population who engage in a particular set of tasks that are not commonly engaged in by the wider public. Since only a small number of people are doing these things, and usually only those people doing them have the knowledge to tell whether they are done well or badly, it would be too cumbersome and without point to have these rules listed in the whole society's code. Instead, they can be listed in a code made available to the profession itself and to those who deal directly with these professionals. But the main point of a code is that it is a statement of the rules which govern, or which ought to govern, a profession. Such a statement of rules will necessarily be incomplete. The point of listing a set of rules is so that people can follow them, and if the list is so long that we cannot find the rules when we want to refer to them, then the list is of no use to us. This is not to say that the rules of the system are or must be incomplete; my point is that the statement of them must be incomplete if the list is to be of any use. For example, the law setting the speed limit at 50 kph on city streets is clear and unambiguous: driving at 51 kph counts as a violation. But if a man is pulled over by a policeman and explains that he was speeding in order to get his wife to the hospital before she delivers the baby, this constitutes an exception to the law but not one that is recognized in the law itself as an exception. The reason for this is threefold. First, the number of exceptions is so large that the statute could not be learned by the public if all the known exceptions were listed. Secondly, the traffic police officer still would have to use his or her judgment to determine whether this case constituted the type of situation the exception was framed to capture. Third, it is arguable that it would be impossible to state in advance all possible types of situations where an exception ought to be made. So, the procedure we follow is to rely on the individual judgment of the police officer involved at the time. In probably 99.9 per cent of cases, no exceptions apply, and in only about 0.1 per cent of cases is an individual judgment called for. Similarly, the professionals' code of ethics will work very well for the bulk of cases, and it would be foolish to attempt to write into the code provisions for moral dilemmas which will be encountered very rarely and which require a great deal of moral deliberation to resolve. A moral theory, on the other hand, is not itself a set of principles but is an attempt to explain why we place

the weight we do on the principles we use to guide our lives; to explain why we rank them in the order we do; and to provide reasons for preferring one principle over another when they conflict. Those of us who act in the world, as professionals do, are primarily interested in this function of a moral theory as a method for providing justifications for what we think we ought to do. In moral dilemmas where two morally good actions are mutually exclusive, a moral theory resolves the dilemma by explaining why one action has a higher value than the other. How does it do this? Actions are done for a purpose, an end, and so what we need when we resolve a conflict between actions is a theory of value: what kinds of things are valuable? Actions? Subjective states such as pleasures, contemplations, intentions, hopes . . . ? How is it that we come to have duties? Is it when the opportunity to do some sort of good presents itself to us or only when we put ourselves under an obligation, as when we make a promise? Do the duties that we contract when we take on a client give our clients special claims on us, even when they are suffering less, or are generally less morally worthy, than a non-client? Who counts from the moral point of view anyway? Who has interests that we must take into account when we decide what to do? Do fetuses, gerbils, trees, or the environment as a whole have interests? If so, how can we weigh them against our own? These are the kinds of questions that we need to answer in order to solve difficult moral dilemmas. Fortunately we need not answer them now in order for me to make the central point of this paper that, given the functions that we expect a code of ethics to serve, we find that it will in most cases be helpless in answering these kinds of questions and therefore helpless in resolving moral dilemmas. To show this I need to outline the functions of a code of ethics. A professional code serves four functions: (1) it gives guidance to the individual professional; (2) its principles are statements of the agreed-upon standards of ethical practice and thus provide guidance for ethics committees struck to rule on professional misconduct of members; (3) it is a public pronouncement of the ethical principles agreed upon by the majority of the profession and thus serves to inform clients about what they have a right to expect from their professionals; and (4) it informs other professionals with whom the professional must work about the kind of cooperation they have a right to expect from the professional and the limits to the cooperation that a professional ought to give. All four of these functions always involve a consensus of the profession. Their point is to announce the moral principles that the profession has already agreed to endorse. Now, this agreement

Beyerstein: The Functions and Limitations of Professional Codes of Ethics

must have been arrived at either by imposition by a group of people whose authority is unquestioned or by some sort of democratic decision-making process. In the former case, there is little to recommend that decision. Part of what makes someone a professional is that she is thought to have the capacity for independent judgment; therefore, it is contrary to the professional ethos for her to do something simply because she is told to. This point holds even when the person doing the ordering is another member of the same profession.' Therefore, this leaves the second option, that the provision of the code will have been arrived at by the usual methods of democratically arriving at a consensus: negotiation and bargaining. And we know what is the usual result of this process: agreement about that which is obvious and prevarication over that which is difficult or contentious. This necessity for compromise severely weakens the first function of a code, to give guidance to individual professionals. While the provisions of the code may be adequate to handle the clear and obvious moral problems that arise, we shall see in a moment, when we examine the nature of moral dilemmas, why the methods of compromise and Trasle-066 are unable to produce a code that is adequate to the task of resolving moral dilemmas. Now let us tum to the second function of the code we have specified: to state the standards of ethical practice that are invoked by an ethics committee to judge a professional's conduct. These standards must be promulgated in advance, just as a law must, if someone is to be accused of violating it. We consider it unjust to charge a citizen with violating a law that was promulgated after the alleged offense was committed, and it would be similarly unjust for a professional society's disciplinary body to discipline a member for something that is not clearly forbidden by the society's code of ethics in advance of the action which is the subject of the investigation. If a professional stands in front of an ethics committee charged with misconduct, then that person is accused of behavior or judgment which falls very much below the standards of his or her colleagues. Therefore, there must exist an agreed-upon standard which this person has failed to meet. The third function of the code, that is, its providing a public pronouncement of the ethical principles of the profession, again requires the consensus of the majority of the profession. As already mentioned, this pronouncement serves to inform clients about what they have a right to expect from the professionals with whom they deal, and therefore such a pronouncement would be hypocritical if it were not agreed upon by the vast majority. It is for this reason that many provisions

in a code tend to be general statements that everyone could interpret in a way he or she finds acceptable. But this generality is precisely what makes these statements useless as a guide to solving a moral dilemma The fourth function of a code is to inform other professionals with whom the professional must work about the kind of cooperation they have a right to expect from the professional and the limits of the cooperation a professional ought to give. Often, in such a case, the individual is pressured by these other professionals to do something he or she morally cannot do, and the code serves to buttress the professional's resolve to resist this pressure and to provide an authority for the professional's arguments against the other professionals: "I cannot do what you request because our code forbids it." It also indicates to the others that there is no point in simply asking some other member of the profession to do this job, since the other members are bound by the same code. But, again, if the code is to serve this purpose, it must express the resolve of the majority of the profession. If it did not, it would be disingenuous for a professional to state that provision of the code as evidence that this is the case. Dilemmas sometimes arise in these cases where one professional, who has one set of duties, asks for cooperation from a second professional, who has a different, conflicting set of duties. In this sort of case, there will be a conflict between the two professionals' codes of ethics. An example of this sort of conflict between codes would be that of a psychiatrist who is approached by his client's lawyer asking that the client be kept in hospital two weeks longer, although she was ready for discharge, so that the lawyer could have the trial deferred and would therefore have time to prepare her case for court. In this sort of case, each professional has a code of ethics which enjoins him or her to promote certain ends. So, here, we have a conflict between codes, as well as a conflict about which end—an honest diagnosis or a fair trial—is most important from the moral point of view: It is understandable in these cases that each professional, given his or her training and the values that led that person into the profession, will come to value things differently. And it is perfectly natural that, having once decided that one of these is the most important, she will struggle to bring it about. But it will not do to say about this sort of case that one professional ought to follow his code and the other professional hers and both simply ought to try to bring about what their respective codes enjoin. In the sort of dilemma we are considering, both professionals have as their ultimate responsibility to look after the client's overall interests, and in so far as both professionals are

II

4111

'ma

4 Professional Ethics

trying to achieve this end, they cannot be indifferent to what ultimately happens. Thus, both professionals must re-examine their codes and look for guidance from moral theory to determine what ought to happen in this particular case and which code has to be revised in order to be consistent with that result. To summarize, then, it appears that three of the four functions of a code of ethics necessarily involve a consensus already having been established among the members of the profession. But with respect to the first function of a moral code, to provide guidance in ethical deliberation, an appeal to what the majority would do cannot, by itself, serve as a moral reason for acting. The reason for this can be seen if we examine briefly the nature of moral deliberation. Moral deliberation necessarily involves the moral agent confronting the situation facing him or her, asking the question, "What ought I to do?" This question is not equivalent to the question, "What is the consensus of my colleagues about what to do?" The question "What ought Ito do?" remains to be asked even after finding out what others recommend. That is not, of course, to say that seeking the advice of your colleagues is worthless. Your colleagues' advice may be exactly right My point is that what makes it right has nothing to do with the numbers of people who take it to be so. Therefore, the fact that your professional colleagues have agreed to put a rule in your code does not make this the right action. Moral dilemmas arise when we are faced with incompatible actions, one of which promotes one end, or one person's end, and the other of which promotes some other end, or some other person's end. Thus, we are faced with a trade-off: at most, we can satisfy one of these ends. But this conflict between ends is not all there is to a moral dilemma. This much is common to almost all moral deliberations. The second feature of a dilemma is that we see both ends as ones that we must promote: we recognize a moral loss if we fail to promote either one of these ends, even if the reason we do not promote it is the impossibility of promoting both ends together. In most moral deliberations we encounter, we see the gain from promoting one end as so obviously greater than that to be achieved by any alternative that the sacrifice of the other end is obviously worth it. But in a moral dilemma, we do not see which end to promote: all alternatives seem all to involve ends which we ought not to sacrifice. Now, it is at this point that I shall actually say something controversial. Up until now, I take it that my remarks about moral theories and moral codes have been merely descriptive and would be accepted by philosophers no matter which of the available alternative moral theories they might defend.

It seems to me that there are two approaches to the study of moral dilemmas that we might take. One approach focuses on what it is really like to be in a moral dilemma—on the psychological and phenomenological attributes of the situation: the angst we experience while we are deliberating about what to do and the regret and sense of loss we experience after the decision has been made. This is the approach we find in European philosophy and among those who have been influenced by this tradition, as well as in the continent's literature. The other approach focuses on the method of deliberation itself. According to this tradition, if it really were the case that we had no reason to prefer one option in a moral dilemma to the other, then there is no role for reason to play. It would not matter which option we chose, since ex hypothesi there is nothing to be gained by the choice. So we really have no deliberation problem, even if we do have a human problem—which the novelist is much better equipped to deal with than the philosopher. My position is the second one, though I cannot justify it in this paper. All I have space for here is to point out a consequence of this position. It is only when we think that there really is a reason to prefer one course of action over another, but that we just do not yet know what it is, that we have a problem. All we can do is either try to find empirical facts of the situation which we have not yet seen which might make a morally relevant difference or refine our theories so that this difference will be apparent. Paying attention to our code is of no use in either of these two endeavors. If I am right, we see something as a moral dilemma, rather than as a usual garden-variety moral problem that we handle every day precisely because we have not gathered the facts or have not done enough moral theorizing. Here we have in a nutshell the reason why codes cannot help us resolve moral dilemmas. The articulation of a code always occurs after the real moral deliberation has taken place, after we have put enough thought into a few of our dilemmas to resolve them so that they are now just moral problems. But in a profession which is advancing, there will be new situations which arise of sufficient complexity to tax the individual's moral sense and which have not yet been subjected to the moral deliberation of the entire profession. These will be the moral dilemmas. I think I have now established the point that we should not look to codes of ethics for the resolution of moral dilemmas. But given our comparison between codes of ethics and society's laws, this should come as no surprise. Being a moral person in society as a whole involves much more than simply obeying society's laws: it involves paying attention to moral theory.

Beyerstein: The Functions and Limitations of Professional Codes of Ethics

Similarly, being a moral professional who is capable of resolving the moral dilemmas that arise in one's professional life from time to time involves much more than acting in accordance with a profession's code of ethics. It involves, in addition, reasoning about what one is doing by paying attention to a wider moral theory. That was the "pessimistic" part of this paper. But every paper ought to end on an optimistic note. I actually have two. First, as I have already mentioned, a code of ethics should serve three other functions, and nothing I have said entails that it cannot perform these tasks adequately. But, more importantly, when we find that our professional codes are inadequate to resolve moral dilemmas, we should remember that we have recourse to moral theory. The reason that some people despair about the prospect of resolving ethical dilemmas with moral theories is that they focus only on the most difficult moral dilemmas that have not been resolved and

therefore conclude that moral theory is incapable of resolving moral dilemmas. Such a view is short-sighted. We should remember that professionals make hundreds of moral judgments every day with confidence, without any reason at all to think that these judgments are difficult or controversial. These moral judgments are not moral dilemmas precisely because our common-sense moral theories are adequate to cope with them; we know exactly how to handle the trade-offs between good and bad that arise with them. In these cases, it does not matter whether professional codes of ethics pronounce on what to do when they arise, for the answer is so obvious. On the other hand, genuine moral dilemmas arise infrequently, but these are the ones we remember—for good reason, of course, because these are the ones that do require the bulk of our attention. The optimistic though undefended conclusion is that moral theory will help us resolve these difficult cases as well.

Notes 1.

If this is a necessary condition for a vocation counting as a profession, then it would seem to rule out certain vocations—nursing and the military, to name two—as professions. In both these vocations, practitioners take direction, and even orders, from superiors. However, this is the point of the qualification "simply": nurses have come to realize the problem with professionals simply following orders and have in their codes addressed this issue. The general consensus among nurses is that a nurse ought not to follow an order that violates her professional responsibilities. This does not mean, though, that a nurse must stop to question every order given: it is perfectly consistent with the notion of professional responsibility

to follow orders during an emergency on the general principle(s) that the person giving them is likely to know what is best and/or that when a task requires the cooperation of several people, things will get ruined if one person "breaks rank" or wastes precious time questioning. So my point is not that it is always inconsistent to follow an order. What is inconsistent with professionalism is to agree in advance to follow orders from a superior, no matter what the circumstances. Soldiers come dangerously close to agreeing to this, and for this reason I take the term "professional soldier" to function like "professional athlete," the point of the honorific being to distinguish careerists from amateurs.

4.3 Cases

Case 1

Hassan Rasouli and Medical Futility See the description of this case from the introduction in Chapter 1 (page 2). Of further note is the physicians' justification for wanting to withdraw Mr Rasouli's life support. Their decision was grounded in a belief that their professional ethics, in particular their right to professional autonomy, permitted them to override, or say "no," to the family's wish that life support be continued. The doctors argued before the courts that life support for Mr Rasouli was futile and so did not

constitute treatment. Therefore, they did not need the family's consent to withdraw treatment. Nor did they need to refer the matter to the Ontario Consent and Capacity Board, which was established to adjudicate disputes with patients and families about treatment. The Rasouli case raises acute questions about the nature and boundaries of professional ethics and professional autonomy and how these are to be understood in relation to patient and family rights to autonomy. The physicians' position relied on assumptions about the usefulness of the notion of futility as a ground for asserting professional autonomy (discussed and critiqued

II al

.1111i1

4 Professional Ethics

in the Truog et al. reading, page 176). Mr Rasouli had shown some evidence of being conscious that everyone acknowledged. His family was firmly convinced that he would find his life to be meaningful even at a low level of consciousness or perhaps just because he was alive. If so, his life support seems not to have been futile, at least according to the values of Mr Rasouli and his family. The physicians appeared to place their judgment of what would make for a worthwhile life over Mr Rasouli and his family's. The Rasouli case may also have been an instance, anticipated by Truog et al., of using the notion of futility to avoid discussion about another issue— namely, fair allocation of scarce health care resources. In a split 5-2 decision, the Supreme Court of Canada ruled to refer the dispute to the Ontario Consent and Capacity Board for review.' The majority decision said nothing directly about the legitimacy of decisions not to treat based on futility. However, it did explicitly state that withdrawal of life support was a treatment decision under the law in Ontario, stating: [Withdrawal of life support aims at the healthrelated purpose of preventing suffering and indignity at the end of life, often entails physical interference

with the patient's body, and is closely associated with the provision of palliative care. Withdrawal of life support is inextricably bound up with care that serves health-related purposes . . . By removing medical services that are keeping a patient alive, withdrawal of life support impacts patient autonomy in the most fundamental way. The physicians' attempt to exclude withdrawal of life support from the definition of "treatment" . . . cannot succeed. The dissenting minority took the view that the case did not need to be referred to the Consent and Capacity Board because the relevant legislation did not apply to the withdrawal of treatment. The minority added that patients cannot insist on treatments that are "medically futile, harmful, or contrary to professional medical standards of care." It argued that the case should be referred to the Ontario Superior Court for a determination of such facts. Neither the majority nor the minority took account of discussions challenging the appropriateness of using the concept of futility to refuse treatment. The case was never referred to the Consent and Capacity Board or the Ontario Superior Court, and Mr Rasouli remains on life support.

Note 1.

Cuthbertson v Rasouli, 2013 SCC 53, [2013] 3 S.C.R. 341.

Case 2

SARS in Toronto: A Duty to Care In late 2002 and through the first half of 2003, a viral outbreak of Severe Acute Respiratory Syndrome (SARS) threatened to become a global pandemic. In Canada, the outbreak focused mainly in Toronto where approximately 400 people became ill, 44 died, and some 25,000 people were quarantined before the outbreak was brought under control.' About 40 per cent of those infected were health care workers. Three health care workers died (two nurses and a doctor). The SARS epidemic generally exposed Canada's unpreparedness for responding to a pandemic. In the years immediately following, concerted efforts were made by all levels of government and by health care professional bodies to create such plans. One of the central issues raised by the SARS epidemic concerned health care providers' professional duty to provide care to infected patients at some risk to their own health. Physicians' ethics codes from

the first half of the twentieth century had included explicit mention of duties to care for the sick during epidemic outbreaks, but these had been removed by the middle years of that century. During the SARS outbreak, a small minority of health care providers refused to go to work in hospitals where SARS patients were being treated. Two years after the outbreak, the Joint Centre for Bioethics (JCB) at the University of Toronto produced a report that attempted to provide an ethical framework for advance planning for infectious disease outbreaks. One of its aims was to clarify the rationale for health care workers' duties to provide care during an outbreak of a virulently infectious disease. The JCB stated the following reasons for recognizing a duty of care in such circumstances: 1. The ability of physicians and health care workers to provide care is greater than that of the public, thus increasing their obligation to provide care. 2. By freely choosing a profession devoted to care for the ill, they assume risks.

Cases

3. The profession has a social contract that calls on members to be available in times of emergency. (In addition, they largely work in publicly supported systems in many countries.)2 The JCB called upon health care professions to address the "vacuum" in their ethics codes and policy documents to provide clear guidance to their members about their obligations and the risks they would be expected to take. The Canadian Medical

Association's policy document "Caring in a Crisis: The Ethical Obligations of Physicians and Society during a Pandemic" (see page 188 in this chapter) is partly a response to such requests. It recognizes a qualified duty to provide care. The Canadian Nurses Association developed a guidance on this topic at the same time and has incorporated it into its ethics code.3 To what extent these guidance documents live up to the JCB's proposed framework, and to what extent that framework is justified, remain matters of discussion.

Notes 1.

2.

"SARS Outbreak in Canada," Canadian Environmental

Health Atlas, Social Sciences and Humanities Research Council, http://www.ehatlas.ca/sarssevere-acute-respiratory-syndrome/case-study/ sars-outbreak-canada. Joint Centre for Bioethics, "Stand on Guard for Thee" (2005), http://wwwjcb.utoronto.ca/people/documents/ upshur_stand_guard.pdf.

Case 3

Is It Ever Right to Violate Rules of Rationing? We (two of the editors) had a philosopher friend who was quite ill and required surgery but was on a waiting list and deemed not ill enough to be admitted to hospital under guidelines that control surgical admissions for immediate treatment. The abdominal pains she suffered did, however, significantly interfere with her life. She had a hard time working or sleeping, could

Case 4

BC Physicians "Cherry-Picking" Patients for Ease of Care, Refusing Those in Need In April 2008, British Columbia Health Minister George Abbott made a public announcement condemning a growing trend in which family physicians were turning away new patients who had or who had previously survived cancer. A similar stern

3.

"Ethical Considerations for Nurses in a Natural or HumanMade Disaster, Communicable Disease Outbreak or Pandemic" in Code of Ethics for Registered Nurses, Canadian Nurses Association (2017), https://www.cna-aiic.ca/html/en/ Code-of-Ethics-2017-Edition/files/assets/basic-html/ page-l.html#.

not properly digest her food, and was losing weight. Surgery was scheduled and cancelled three times. Her surgeon finally said, "Enough is enough" and told her to go to emergency and gave her a list of symptoms to tell the physicians there. These symptoms, he said, would give her a bed, and he would then come and perform the necessary surgery. And that is how our friend (who taught ethics) got her surgery. Events like this happen every day in Canada's health care system. Was the surgeon wrong to violate the hospital's rules of rationing? Was our friend wrong to do what he told her to do?

warning was issued by the College of Physicians and Surgeons in its June 2008 bulletin. Dr Morris VanAndel stated that it is "unacceptable for doctors to turn away patients just because their care will be complex, time-consuming, and likely to involve consultations with family members." Dr VanAndel clarified his view that it is acceptable to reject new patients only when the physician does not offer the services the patient requires, such as geriatric services for elderly patients or obstetrics services for pregnant patients.

207

II

iI

111111

4 Professional Ethics

Nicole Adams, a spokesperson for the BC Cancer Agency, confirmed that cancer patients report having a very hard time finding family physicians. Alixe Cormick provides one example, having called nearly 20 physicians before finding one who would attend to her 64-yearold mother, Beatrice Hanson, during the time she was waiting to see an oncologist at the BC Cancer Agency for her pancreatic cancer. Her own family physicianb receptionist refused to book an appointment for her mother, specifically because of the cancer, but did offer an appointment to Ms Cormick for a Botox injection, a cosmetic service the physician had just begun offering. Every one of the nearly 20 offices Ms Cormick contacted turned her away when they found out she was calling for a cancer patient. "I kept being told that they aren't properly compensated for patients who have to be seen so much."' By the time she found a physician willing, as a favour to a friend, to see Ms Cormick's mother, Ms Hanson was in Vancouver General Hospital's palliative care unit, where she subsequently died.

Any trend toward selecting patients on the basis of the ease of providing their care is likely fuelled in part by a continuing shortage of family physicians in British Columbia and by a fee structure that pays $30 for a usual visit for individuals under 80 years old and $45 for those over that age. An estimated 200,000 BC residents did not have a family physician as of December 2015, despite efforts over several years by the provincial government to remedy the situation, including some measures to pay more for complex care patients.2 Given the scarcity of family physicians in BC, is Dr VanAndel correct that there are no circumstances under which a family physician can ethically refuse to see a patient requiring particularly time-consuming care if the physician offers the services the patient requires? For example, if a family physician's practice is almost full and already has a substantial number of complex care patients, is there a professional duty to accept further complex care patients? How do we determine what counts as "cherry-picking" less complicated patients?

Notes 1.

Pamela Fayerman, "Health Minister Upset with Recent Trend of Doctors Refusing to Take New Patients with a History of Cancer," Vancouver Sun 23 April 2008, Section A, 1-2.

Case 5

Helping Residents to Live at Risk Residents in long-term care facilities often wish to engage in behaviours that put their health at risk. These behaviours can include wanting to smoke cigarettes, drink alcohol, or eat food that threatens their health (for example, when there is a choking risk for residents with swallowing difficulties). Residents are often fully competent to make choices to live at risk, and some accommodation is often made for smoking and alcohol. But residents frequently have mobility issues that make it difficult for them to obtain or consume these goods. A question arises over whether health care providers have any responsibility to assist residents in acquiring or consuming these potentially harmful goods or otherwise to facilitate risky behaviour. Health care providers are split on this issue. Those who refuse to help residents are often adamant that their motive is not paternalistic.' Rather, they claim a professional duty of non-maleficence or "to do no harm" prevents them from acting on residents' behalf.

2.

Ron Seymour, "Doctor Shortage Linked to Government Incentive Program," Kelowna Daily Courier 9 December 2015, http://www. kelownadailycourier. ca/news/article_ cf dalObe-9f 00-11 e5-87e7-f f ec8edf76c 1 .htm 1.

Their refusal to help, they say, is an expression of professional autonomy. A response to this position is that, excepting obvious, imminent, and serious risks, residents may be in a better position to judge whether risks of harm are worth it on balance. If so, referring to professional autonomy may be a cover for paternalism after all. It is becoming increasingly recognized that competent residents in care facilities can choose to live at risk. In 2015, the Fraser Health Authority in British Columbia adopted a policy entitled "Supporting Residents to Live at Risk in Residential Care and Assisted Living," which explicitly recognized residents' "right to live at risk" under its residential care and assisted living programs. The policy went further, however, in stating that residents "will be supported to live according to their preferences, values and beliefs to the greatest extent possible—to reflect the level of autonomy they would enjoy outside the facility."2 These words and the title of the policy apparently impose requirements on health care providers to offer assistance to residents when they make choices to live at risk. What this means in practice is unclear, and the boundaries between patient autonomy and

Cases

professional autonomy remain to be worked out. adopted elsewhere and, if so, how they will be reNevertheless, an obligation to help residents live at ceived and applied by health care providers are issues risk is acknowledged. Whether such policies will be for today and coming years. Notes 1.

Alister Browne, "Helping Residents t o Live at Risk," Cambridge Health Quarterly of Healthcare Ethics 12 (2003): 83-90.

Case 6

When Is Treatment Futile? On a cold mid-January morning in 2016, Queen's University physics student Tayyab Jafar attempted to end his life by taking an overdose of sedatives and lying down to die in the snow in Kingston, Ontario. Jafar had been under psychiatric care and had threatened suicide previously. One of his friends found Jafar's "good bye" text message and went looking for him. When he found Jafar, there were no vital signs, and his core body temperature had fallen to 20.8 degrees Celsius, 16 degrees below normal and about 14 degrees below where hypothermia occurs. No one knew how long Jafar had been there. It had been five hours since the text message was sent. There was a slight chance of resuscitation only if Jafar had become hypothermic after he had taken the sedatives. No one knew which happened first. The emergency responders brought Jafar to Kingston General Hospital where it was decided to attempt resuscitation, though it was unknown what the

2.

Fraser Health Authority, "Supporting Residents to Live at Risk in Residential Care and Assisted Living" (2015), http://www.fraserheahh.ca/media/201501PolicySupporting ResidentsLiveAtRisk.pdf.

chances of this working were or if it worked whether the patient would be severely disabled. The resuscitation procedure required intravenous heated saline solutions, about 100 units of rare AB positive type blood, and a cardiac bypass to permit "extracorporeal rewarming" of Jafar's blood. A special ventilator was required to prevent lung damage. Jafar was resuscitated, though the cardiac surgeon noted, "we may just have saved him into a chronic vegetative state." During Jafar's recovery, his body weight went from 140 pounds to about 80 pounds. He was in great pain during his treatment and rehabilitation and asked to be allowed to die. The hospital staff continued treatment. Jafar required extensive rehabilitation, including learning to walk again and to use his hands. He recovered and returned to study at Queen's in the fall. In response to a question about whether he was happy to have a second chance at life, Jafar answered, "I'm happy because I want to do many things . . . I'm pretty much happy at this point." The surgeon commented, "I've been doing this for 25 years and he's the first survivor I had."'

Note 1.

Mary Ormsby, "This Queens Student Froze to Death on a Kingson Pier. Here Is How He Came Back to Life," Toronto Star 19 February 2017, http://www.msn.com/en-ca/news/canada/

this-queen%E2%80%99s-student-froze-to-death-ona-kingston-pier-here%E2%80%99s-how-hecame-back-to-life/ar-AAn6jTr?li=AAadgLE&Tocid= spartandhp.

4.4 Study Questions 1. How is patient autonomy bound u p with patient well-being, and how does this affect the decisions of health care providers' to refuse requested treatments?

209

II

210

iull

111E11

4 Professional Ethics

2. What difficulties arise when health care providers wish to say "no" to a patient's request for a treatment that is harmful to the patient, harmful to others, futile, or uses scarce health care resources? 3. Is a health care provider's "conscientious objection" or "conscientious refusal" to provide a requested treatment an expression of professional autonomy? Why/why not? 4. Is futility ever a reasonable ground on its own for saying "no" to a patient's request for treatment? 5. What professional duties, if any, do health care providers assume to provide care at some risk to themselves—for example, during a pandemic or in dealing with abusive or violent mentally ill patients? 6. How might the fiduciary duty of health care providers to act in their patients' best interests lead to a "tragedy of the medical commons"? How might this "tragedy" be avoided in a way that is consistent with the professional obligations of health care providers? 7. What is the main value of professional ethics codes? How much useful ethical guidance is contained within them?

4.5 Suggested Further Reading Codes of Ethics Canadian Medical Association. 2004. "CMA Code of Ethics." https://www.cma.ca/ Assets/assets-library/document/en/advocacy/policy-research/CMA_Policy_Code_of_ ethics_of the_Canadian_Medical_Association_Update_2004_PD04-06-e.pdf. Canadian Nurses Association. 2017. "Code of Ethics for Registered Nurses." 2017. https:// www.cna-aiic.ca/html/en/Code-of-Ethics-2017-Edition/files/assets/basic-html/page-1. html#.

Professional Ethics Gampel, E. 2006. "Does Professional Autonomy Protect Medical Futility Judgments?" Bioethics (20): 92-104. Hebert, Philip C. 2014. Doing Right: A Practical Guide to Ethics for Medical Trainees and Physicians. 3rd edn. Oxford University Press. Pellegrino, E. 2002. "The Physician's Conscience, Conscience Clauses, and Religious Belief: A Catholic Perspective." Fordham Urban Law Journal (30): 221-44. Reid, Lynette. 2005. "Diminishing Returns: Risk and the Duty to Care in the SARS Epidemic." Bioethics (1): 348-61. Schneiderman, Lawrence J., Nancy S. Jecker, and Albert R. Jonsen. 1996. "Medical Futility: Response to Critics." Annals of Internal Medicine 125 (8): 669-74. Wicclair, Daniel. 2000. "Conscientious Objection in Medicine." Bioethics 14 (3): 205-27. Yeo, Michael, Anne Moorehouse, Pamela Kahn, and Patricia Rodney. 2010. Concepts and Cases in Nursing Ethic. 3rd edn. Broadview Press.

Reproducidve ahiics 5.1 Introduction When people think of ethical issues raised by human reproduction, their attention is often focused on abortion. But human reproduction raises a host of thorny ethical issues both before and after abortion may be contemplated. It is useful therefore to organize discussion of reproductive ethics chronologically. To start, there are ethical issues that arise before reproduction begins. Roughly speaking, these are about controlling the circumstances of reproduction, including when to reproduce, with whom, who makes those decisions, and what support, if any, should be provided to assist people to reproduce. Once reproduction is started, the issues shift to decision-making over whether to stop or continue reproduction. Stopping reproduction is about abortion and the moral claims, if any, the fetus has on decisions to abort it. Continuing reproduction is about obligations an expectant mother and others have to protect and promote the health and well-being of the fetus and future child. The topics in this chapter are presented in this order. We have tried to cover many of the important issues around starting and continuing reproduction, though the issues are more extensive than we have space for (see the suggested readings). The main part of the chapter is devoted to readings on abortion.

Starting Reproduction Christine Overall's "Reproductive Freedom, Autonomy, and Reproductive Rights" (see page 218) addresses the nature and limits of rights to start reproduction. Overall argues that a moral right to reproduction is grounded in both consequentialist and deontological reasoning. She notes that for many people, having children is a profound source of value and not having them a source of great sorrow or regret. It is also important that children be born into families that want them. These are the consequentialist reasons for a right to reproduction. The deontological argument is that reproductive rights respect women's freedom to make choices to start or refuse to start reproduction. Thus, they preserve a woman's autonomy and control over her body. Overall argues that the consequentialist reasons are grounds for providing medical care and assistance to women who have made decisions to reproduce. Such positive assistance should be available without unjust discrimination—for example, regardless of marital status or sexual orientation, as is the case for other medical care. The deontological reasons protect women's and men's choices to control when, where, and with whom to reproduce, including the right to refuse to reproduce and decisions to donate reproductive gametes to assist others to reproduce. These are negative deontological claims, or rights not to be interfered with.

M

111

5 Reproductive Ethics

commercial surrogacy payment to a woman beyond reimbursement of expenses for agreeing to carry a pregnancy for another person or couple who will become the child's parent(s).

However, Overall is clear that these reproductive rights do not guarantee a baby to anyone, nor do they by themselves morally justify a decision to reproduce that includes a broader range of potentially competing moral considerations. A recognition of positive and negative claims or rights to reproduce and their limits is not controversial in modern liberal democratic societies, but they deserve to be carefully stated as a backdrop for discussions of other aspects of starting reproduction. One issue about starting reproduction not addressed in Overall's article is sex selection of children, or "preconception gender selection" of offspring (PGS). This practice is frequently condemned as contributing to sexism toward women where there are cultural preferences for boys. John A. Robertson argues, however, in his article on PGS (see page 226) that, assuming PGS is safe and effective (as it is today), such concerns about sexism do not apply when parents make decisions to have a child of a different gender from the child(ren) they already have. Nor is there anything necessarily problematic about choosing the sex of a child because the parents prefer to have the experience of raising a boy or a girl. Sexism is relevant only when PGS is based on a belief in the superiority or greater value of boys versus girls (or vice versa). Robertson defends a controversial position, but it is difficult to see an argument against it, given deontological assumptions about reproductive freedom stated by Overall. Robertson acknowledges, however, the possibility that such preferences could lead to imbalances in numbers between the sexes that are socially destabilizing. There is no evidence at this time, however, to hold that preferences for different gendered child-raising experiences are common enough that they would lead to imbalances between the sexes in significant numbers in modern industrialized societies. Robertson's position is an implicit criticism of the law in Canada, which prohibits PGS. This law was adopted following the recommendations in 1993 of the report titled Proceed with Care: The Final Report of the Canadian Royal Commission on New Reproductive Technologies. This report held that the practice of PGS "violates respect for human life and dignity, sexual equality, protection of the vulnerable, and appropriate use of resources."' Whether the royal commission adequately answers Robertson's arguments is an interesting question. A further issue (among many) addressed by the royal commission is commercial surrogacy. The relevant material is excerpted in the article "Preconception Arrangements" (see page 233). Again, the royal commission recommended a general prohibition, and this was also adopted in law in Canada. Anyone who offers or pays for or assists commercial surrogacy is criminally liable to fine of up to $500,000 and up to 10 years in prison.2 The royal commission argued that commercial surrogacy should be prohibited on the grounds that it offended human dignity by commodifying women's reproductive capacities and by commodifying children. It also held that commercial surrogacy would have negative psychosocial consequences for the surrogate mother, would cause harms to the children produced, and would facilitate exploitive power imbalances between surrogates and their contracting partners. However, the royal commission did not defend a total ban on surrogacy. It recommended permitting women to volunteer as surrogate mothers as long as they do not accept payment beyond personal expenses related to the pregnancy. The current law is uncertain as to whether this is permitted, but there are businesses and lawyers who will assist such "non-commercial surrogacy" arrangements, including by facilitating offshore international surrogacy contracts, which are not covered by Canadian law. Given the difficulty of establishing a boundary between reimbursement for expenses and payment for services, some forms of commercial surrogacy are undoubtedly being practised in Canada today. From an ethical perspective, the topic remains controversial, since the current criminal prohibition of commercial surrogacy imposes a clear restriction on women's rights to control their bodies and reproductive choices. Moreover, in the over 20 years since the royal commission's report, it has been difficult

Starting Reproduction

to find evidence confirming negative consequences of the practice from First World jurisdictions where it occurs and when there are regulations to address issues around exploitation and the health of the surrogate. A large number of studies from the United States and the United Kingdom show that surrogate mothers are generally happy with their experience.3 Arguments about children's rights and welfare have also been challenged.' As a result, Canadian surrogacy advocates and some researchers have argued that the law is outdated and should be changed. Commercial surrogacy is permitted in many states in the United States but prohibited in many European countries, including the United Kingdom.

Stopping Reproduction—Abortion Most countries in the world regulate abortion. It is generally prohibited outright in South America, predominantly Catholic and Muslim countries, and Africa. Most other countries, including those in Europe, the United States, Australia, and Russia, typically allow abortions up to a certain time and restrict them after that. In sharp contrast, Canada and China are the only countries without any legislative or judicial restrictions whatsoever. In the case of Canada, this was not for want of trying. Canada's abortion law was struck down as unconstitutional by the Supreme Court of Canada in 1988.5 The House of Commons rewrote and approved the law in 1989, but it did not pass the Senate. The government of the day declined to try again, a decision followed by all subsequent governments. The question we take up in this chapter is whether this is ethically acceptable and if not, what restrictions should be placed on access to abortions and why. We begin with the logic of the problem. The standard view is that the central question concerns the rights status of the fetus. Women have an uncontroversial right to control their bodies. Thus, if the fetus has no right to life, there is no case for setting that right aside, and abortion must be available on demand. But if the fetus has a right to life either at conception or some later time, the woman's right to control her body is limited. How limited depends on when that right comes into being and how strong it is. We will call this the "status of the fetus" approach, and it is the view taken by Don Marquis in "Why Abortion Is Immoral" (see page 238), Mary Anne Warren in "On the Moral and Legal Status of Abortion" (see page 248), and L.W. Sumner in "A Third Way" (see page 262). Others, however, think that the status of the fetus is largely or entirely irrelevant to the debate. In our readings, Judith Jarvis Thomson in "A Defence of Abortion" (see page 253) argues that even if the fetus has a full right to life, abortion will still be morally permissible in a wide range of circumstances, and Susan Sherwin in "Abortion through a Feminist Ethics Lens" (see page 268) presents a feminist position according to which abortion must be permissible in all circumstances from conception through birth. We will call this the "priority of the woman" approach. Fetal Development and Terminology It will be useful to start by understanding the nature of fetal development and by introducing some terminology. Human beings come into existence when an egg and sperm unite to form a zygote with a complete genetic code. Within 24 hours, cellular cleavage starts to take place and continues as the zygote moves down the Fallopian tube into the uterus. Upon entering the uterus, the multi-cell zygote—now called a "conceptus"—starts to burrow into the uterine wall. After two weeks, implantation is complete, and the organism is called an "embryo." A number of dramatic changes then rapidly take place. A rudimentary heart starts to beat in three to four weeks; at six weeks the organism is recognizably human; brain activity can be monitored at eight weeks, and spontaneous movement by

fetus (as used in biology) a pre-natal organism in which all organ systems are in place and functioning (approx. 12-36 weeks old).

II

iII

MILLI

5 Reproductive Ethics

fetus (as generally used in philosophic discussions of abortion) a human organism at any stage of fetal development from zygote to (but not including) infant.

abortion termination of pregnancy.

the organism (motility) occurs at around 10 weeks. At 12 weeks, all organ systems are in place and functioning, and the organism is called a "fetus." The woman may feel the fetus moving ("quickening") at 12 to 16 weeks, and it is capable of independent survival ("viable") sometime between 20 and 28 weeks. Birth typically occurs at nine months, and at that time the organism is called an "infant." In the philosophic literature on abortion, it is common to refer to an organism at any stage of fetal development as a "fetus," rather than use that term in the above biological sense, and we will follow that practice. "Abortion" means terminating a pregnancy, and this is to be distinguished from killing the fetus, which is called "feticide." This distinction is important because whereas in many cases pregnancy cannot be terminated without killing the fetus, once the fetus is viable it can be. A woman's right to control her body will (at most) give rise to a right to abortion. Feticide requires a separate argument. We are now ready to consider the morality of abortion.

Status of the Fetus: Approaches feticide killing the fetus.

The Conservative Theory Conservative Theory the fetus has a right to life from the point of conception onwards.

The Conservative Theory holds that the fetus has a right to life from the point of conception onwards. There are two versions of the theory. One holds that the fetus acquires a small right to life at the point of conception, and as the organism develops the right becomes stronger, and a woman accordingly requires increasingly weighty reasons to have an abortion. The problem facing this version is that of giving the view operational significance. To have practical application, the theory must tell us when abortions are permissible and when not. But it is not clear what non-arbitrary rationale can be provided for saying (even roughly) at what point in fetal development a woman may have an abortion for what reason. The second version of the Conservative Theory holds that the fetus acquires a full right to life (i.e., a right indistinguishable in strength from that of a full-fledged human being) at conception. This entails that that abortions that kill the fetus are justified only in the circumstances in which it would be justified to kill a full-fledged innocent human being. Since that is seldom, abortions are seldom if ever justified. This is the view Don Marquis holds in "Why Abortion Is Immoral" (see page 238). He argues that what makes it prima facie wrong (i.e., wrong in the absence of special circumstances) to kill a fullfledged human being is that the killing takes away that beings future. But killing the fetus likewise takes away its future, and thus it must also be prima facie wrong to kill it. Thus, the fetus and a full-fledged human being have the same right to life. A standard objection to the Conservative view is that the same reasoning makes celibacy and contraception wrong as well, for engaging in those things likewise cancel a potential future. Marquis's reply is that this objection does not face his version of the Conservative Theory. He argues that in celibacy and contraception, there is no identifiable single being whose future is taken away. A possible future is cancelled, to be sure, but no one now exists who has a right to that future. If Marquis had argued (as other Conservatives do) that the fetus has a right to life because it is a potential full-fledged human being, he arguably would be committed to opposing celibacy and contraception. This is because an egg and sperm are objects that together constitute a potential full-fledged human being and thus to prevent their union must be as wrong as permitting it and then killing the resulting organism. He would also be vulnerable to the objection that potential Xs do not have the rights of actual Xs. For example, medical students and citizens are potential doctors and prime ministers but do not have the rights of each. There is no reason, then, to recognize the fetus as having a full right to life in virtue of its potentiality. These are serious problems for other Conservative views, but Marquis's account escapes them.

Stopping Reproduction—Abortion

The Liberal Theory

The Liberal Theory holds that the fetus lacks a right to life sufficient to compete with a woman right to control her body at all stages of its development. This view is represented in our readings by Mary Anne Warren in "On the Moral and Legal Status of Abortion" (see page 248). For Warren, the key question of abortion is "Is the fetus a person?" where a person is characterized by five traits: (1) consciousness, (2) reasoning, (3) self-motivated activity, (4) the capacity to communicate, and (5) the presence of self-consciousness and self-awareness. Warren does not claim that these traits are individually necessary or any subset of them jointly sufficient to be a person but merely that a being that had none of them is not a person. Warren further claims that all and only persons have full moral rights and acknowledges that many genetic human beings will thereby lack full moral rights. These include the permanently unconscious, those without appreciable mental capacity, and fetuses. In her view, it is plain that the fetus is not a person in the sense described above and that "to ascribe full moral rights to an entity which is not a person is as absurd as to ascribe moral obligations and responsibilities to such an entity." Warren anticipates the objection that while these beings may not have full moral rights by virtue of being persons, they may be ascribed such rights because they are like persons or are potential persons. In reply, Warren contends that even if a right to life could be ascribed to the fetus on this basis, "such a right could not possibly outweigh the right of a woman to obtain an abortion, since the rights of an actual person invariably outweigh" those of pseudo-persons or potential persons. A much-discussed alternative Liberal Theory is that developed by Michael Tooley in "In Defense of Abortion and Infanticide" (see Suggested Further Reading on page 286). This makes a valuable addition to Warren's account because whereas Warren treats it as self-evident that only persons have a right to life, Tooley provides a reason for thinking this. In Tooley's view, the key element in ascribing a right to life to an organism is self-consciousness (Warren's fifth characteristic of a person). This is because (he argues) an entity cannot have a right to life unless it is capable of having an interest in its own continued existence, and this requires that it has a concept of a continuing self as a subject of experiences and other mental states. Thus, an entity that has no concept of self cannot have a right to life, and since it is plain that a fetus does not have such a concept, a fetus does not have a right to life. Both Warren's and Tooley's theories have the consequence, common to all Liberal Theories, that since newborn infants do not have a right to life, infanticide must be morally permissible. Both also embrace that view on roughly the same grounds, and Warren sets these out in the postscript to her article.

Liberal Theory the fetus lacks a right to life at all stages of fetal development.

infanticide killing an infant.

The Middle Theory

For those who find the Conservative and Liberal theories unacceptable, the Middle Theory beckons. The standard Middle Theory holds that the fetus lacks a right to life up to a particular point and after that point has a right to life. Sometimes it is held that the right to life starts out very weak and grows stronger as the fetus matures (just like in the gradualist version of the Conservative Theory), and sometimes that a full right to life suddenly springs into existence. Middle theories also differ on what that point is. Traditional candidates include when it is recognizably human, acquires brain activity, makes spontaneous movements, and becomes viable. Two problems, however, face selecting the magic moment. The first is to show that any characteristic selected is significant enough to make such a momentous moral change. The second is to identify at what point that characteristic comes into being in a process of development that does not have sharp discontinuities anywhere. These difficulties have made it seem to many that we are forced to choose between the extreme positions. L.W. Sumner in "A Third Way" (see page 262), however, argues that a middle ground can be found and proposes "sentience" as the criterion of moral standing. Sentience

Middle Theory the fetus lacks a right to life up to some point or threshold stage in fetal development and has a right to life after that time.

sentience the ability to have experiences.

II

ill

I111.1.1

5 Reproductive Ethics

refers to the ability to have experiences. These include primitive sensations like pleasure and pain as well as more sophisticated mental states such as wants, aims, tastes, emotions, and sentiments. The moral relevance of sentience comes (for Sumner) from the fact that the ability to have any of the above experiences gives an organism interests and morality has to do with the protection and promotion of interests. Since it is plausible to suppose the fetus can experience pain at some point in fetal development, Sumner concludes that the fetus has moral standing at that point. It does not follow that abortion is always wrong from that point on Exceptions for special considerations such as the health of the mother or fetus are possible. But on the assumption that abortion will cause the fetus pain, it does follow that some regulation of the practice is apt, and thus Canada is wrong to allow abortion on demand. Sumner acknowledges that the criterion of sentience does not allow us to identify a precise point at which abortion should be regulated, since sentience gradually phases into existence. But he argues that it is possible to identify a threshold period before which we can be morally certain that the fetus cannot experience pain and after which we can be morally certain it can. On the assumption that the capacity to feel pain tracks physiological development, this period can be located in the second trimester around the time of viability. The more precise dating of when abortions are not permissible required by regulation (Sumner argues) can be arbitrarily identified within this non-arbitrarily identified period. Priority of the Woman: Approaches Judith Jarvis Thomson, in "A Defense of Abortion" (see page 253), argues that ascribing a right to life to the fetus will not serve those who oppose abortion in the clear simple way they often think it will. Even if the fetus has a full right to life, she argues, abortion will be justifiable in a wide range of circumstances. To show this, Thomson invents one of the bestknown thought experiments in philosophy. Suppose that a famous unconscious violinist has failing kidneys, and to save him the Society of Music Lovers attach him to a sleeping woman whose kidneys can service both. This will save his life, but the woman will have to live in this compromised state for nine months. She awakes and wants to detach him. Can she do so? Thomson says it is absurd to say "no" and that this immediately shows that it is not wrong to kill an innocent human being with a full right to life. To be at all plausible, that claim must be amended to "it is always wrong to kill such a human being unjustly." The question then becomes when would such a killing be unjust, and Thomson argues that in the case of the violinist, it would be only if he had a right to the continued use of the woman's body. Thomson further contends that he would have such a right only if the woman had explicitly or implicitly granted it to him. Thus (to return to the case of abortion), abortion in the case of rape is not wrong as the violinist case immediately shows, nor is it wrong in cases of pregnancy when the woman has taken sufficient precautions to make it implausible to say she has consented to letting the fetus share her body. Thomson's view thus does not show that abortion is never wrong but only that its wrongness cannot be straightaway inferred from an alleged right to life of the fetus. To conclude that abortion would be wrong in any given case requires showing that in that case, the woman gave the fetus a right to use her body, and Thomson leaves it open as to whether that can ever be shown. In "Abortion through a Feminist Ethics Lens," Susan Sherwin (see page 268) takes a stronger view. Whereas Thomson holds that there may be occasions on which abortion is "unjust killing," Sherwin does not. According to Sherwin, the fetus is morally significant only through its relational status as part of a pregnancy occurring within a particular woman's body. As such, the fetus is not a moral person and cannot be ascribed a right to life. In this Sherwin and the Liberal agree that women must have a right to abortion on

Stopping Reproduction—Abortion

demand. But their views differ on the permissibility of infanticide. The Liberal is committed to endorsing infanticide because, on the Liberal view, the right to abortion follows from the claim that the fetus does not have a right to life. Sherwin, by contrast, holds that a woman right to abortion follows from the right of a woman to control her pregnancy, and this leaves open the question as to whether the infant has a right to life once the pregnancy is terminated.

Continuing Reproduction Women sometimes do things (e.g., smoke, sniff glue) or fail to do things (e.g., follow a diet, get bed rest) that put their fetus at risk. The possible effects on the fetus and any child that may eventuate from a fetus carried to term (which we will call a "child-to-be") range from unlikely, mild, and transitory to highly probable, devastating, and lifelong. What obligation, if any, does the woman have to avoid these risks? If she does not avoid them voluntarily should the law be used to force her to? These are the questions we take up in this section with two selections from the Royal Commission on New Reproductive Technologies. In the first selection, "Judicial Interventions in Pregnancy and Birth" (see page 277), the commission gives the official recommendation of the report. The commission argues that women have an ethical obligation to care for their fetus and child-to-be, but if they do not, judicial intervention should not be invoked to force them to. In On Liberty, John Stuart Mill argued that if an act does or threatens harm to the interests of others, it falls in principle under the scope of the law.6 Mill did not, to be sure, think that the law should always or even often be used. In his view, it should be used only if making the conduct illegal would: (1) be effective in curbing the harm; (2) not generate other and greater harms than it prevents; (3) be the mildest means to curb the evil; and (4) except as a last resort, be thought justifiable by those subjected to it. For Mill, any intervention that fails to meet this last condition (to use his purple phrase) "partakes of the degradation of slavery."' (See the account of Mill's theory of liberty in the Introduction to this text, page 18.) The commission argued (in effect; it did not mention Mill) that judicial intervention in obstetrical cases would fail on every one of Mill's points. Thus it firmly and unequivocally concluded that judicial intervention is never permissible in either the interest of the fetus or the child-to-be. This absolutist position on issues that are filled with ethical ambiguity is striking, and Suzanne Rozell Scorsone, in the second selection from the Royal Commission, "Dissent on Judicial Intervention in Pregnancy and Birth" (see page 281), disagrees. She argues that while society should be reluctant to appeal to the law in obstetrical matters, its use cannot be absolutely ruled out.

Notes 1. 2. 3.

4. 5. 6. 7.

Proceed with Care: The Final Report of the Royal Commission on New Reproductive Technologies, vol. 2, p. 916 (Minister of Government Services Canada, 1993). The Assisted Human Reproduction Act (S.C. 2004 c. 2), section 6. http://laws-lois.justice.gc.ca/eng/ acts/A-13.4/page-1.html. Karen Busby and Delaney Vun, "Revisiting The Handmaid's Tale: Feminist Theory Meets Empirical Research on Surrogate Mothers," Canadian Journal of Family Law 26 (2010), 13-94; Karen Busby, "Is It Time to Legalize Commercial Surrogacy in Canada?" Impact Ethics (3 February 2015), https•!/ impactethics.ca/2015/02/03/is-it-time-to-legalize-commercial-surrogacy-in-canada. Jason K.M. Hanna, "Revisiting Child-Based Objections to Commercial Surrogacy," Bioethics 24 (2010), 342-7. R v Morgentaler, 11988] 1 S.C.R. 30 J.S. Mill, On Liberty (1858), ch. 1, para. 11. J.S. Mill, Principles of Political Economy (1848), bk. 5, ch. 11, sec. 2.

I I .111

111141

5 Reproductive Ethics

5.2 Starting Reproduction

Reproductive Freedom, Autonomy, and Reproductive Rights Christine Overall

The ground of the right to become a parent is indeed the interests of the potential parents. Becoming a parent is something that lends shape and meaning to one's life and often to a life that one shares with another parent; and evidence suggests that the interest is one that is very widely shared. So it is a natural candidate to ground a right. That right cannot be absolute, for two reasons: first, the standard feasibility test may in some cases be a problem, and it would be unreasonable to demand that unlimited medical resources be devoted to solving it; second, there are issues about the suitability of potential parents that worried (for example) J.S. Mill, who believed that the responsibilities of parenthood were beyond the capacity of some potential parents. So not everyone can, or should, become a parent. —Sarah Hannan and Richard Vernon, Parental Rights There are three main reasons for starting this exploration of the ethics of procreation with an examination of rights. First, moral rights are fundamental to much ethical debate, and reproductive rights have typically been emphasized in traditional discussions of procreative issues such as abortion and IVF. But as the quotation from Sarah Hannan and Richard Vernon suggests, reproductive rights are more complex than is often recognized. It is therefore necessary to delineate the scope and limits of reproductive rights; this discussion provides background for the rest of the book. Second, some people might assume that deciding to have a child is easily justified on the grounds that doing so is simply an expression of reproductive rights.' I examine whether this assumption is correct. Third, like many other rights, reproductive rights may also play a protective role, providing a moral defense against social and legal demands for certain kinds of procreative behavior. If a person has a right to do X, then it is morally wrong to prevent the person from doing X. If a person has a right not to do Y, then it is morally wrong to force

the person to do Y. The protective function of reproductive rights must be clarified. In this chapter, I develop a taxonomy of reproductive rights—what they are, how they are justified, and what their limits might be. My focus here is on moral rights, not legal rights. A moral right is an entitlement that we have good reason to accept, an entitlement that is an expression of one's humanity and that belongs to an individual by virtue of her or his being a human person. A moral right may or may not be legally recognized; that is, a moral entitlement may or may not be encoded within the laws of the state in which an individual lives, but it is no less real and morally significant even if it is not recognized by the state (Kates 2004). For example, the 1994 United Nations International Conference on Population and Development (uNicPD) affirmed "the basic right of all couples and individuals to decide freely and responsibly the number, spacing and timing of their children and to have the information and means to do so, and the right to attain the highest standard of sexual and reproductive health. It also includes the right of all to make decisions concerning reproduction free of discrimination, coercion and violence as expressed in human rights documents" (quoted in Kates 2004, 57). Even in the most repressive regimes where there are few or no reproductive services or protections, individuals nonetheless still possess moral reproductive rights, despite their not being legally recognized. I both identify our moral reproductive rights and discuss the kinds of state resources, services, and protections that those rights may indicate. A state that provides the resources, services, and protections supported by our moral reproductive rights is in fact according to its citizens legal reproductive rights, even if those legal rights are not formally stated and acknowledged. Carol A. Kates claims that there is no need to recognize any special fundamental reproductive rights; we can be content with "a general right to liberty" (2004, 63). But subsuming reproductive rights under a general right to liberty may not be adequate both because of the profound value of reproduction to most individuals and societies as well as because of the far-reaching impact of procreation, particularly on women. We might therefore say that reproductive rights have both a consequentialist and a deontological foundation.

Source: From Why Have Children? The Ethical Debate, Chapter 2 (The MIT Press, 2012), 19-33. Reprinted with permission.

Overall: Reproductive Freedom, Autonomy, and Reproductive Rights

First, having children is, for many people, deeply definitive of their identity and their life's value. For others, remaining childless is equally essential. Failing to have a child when one wants to be a parent can be a source of immense sorrow and regret. Becoming a parent against one's wish can be a lifelong burden. Hence, the protection of procreative choices by means of the recognition of reproductive rights is necessary both to ensure that peoples lives go well and to prevent the misery deprivation, and even oppression that results when people have little or no control over their procreative behavior. In addition, the well-being of children is dependent on the recognition of reproductive rights. Children may suffer if they are unwanted. Within families, unwanted children may be handicapped by scarce material resources, lack of attention, and stressed parenting. Within society at large, as we know from the past, unwanted children are likely to face inadequate provisions for their education, health care, and eventual employment. There are thus strong consequentialist reasons for articulating and protecting reproductive rights. Second, women are particularly vulnerable with respect to procreation. Most women are fertile from their early teens until their late forties. Women conceive, gestate, deliver, and breastfeed their babies. Women are still expected to be the primary caregivers for their children. As a feminist, I therefore take women's bodily freedom (the absence of physical, legal, or social constraints on one's decisions about one's body) and autonomy (the capacity to be self-determining, especially with respect to one's body) to be the sine qua non for women's equality and full citizenship. The deontological basis for reproductive rights is that they are indispensable to protecting women's personhood. Without moral recognition and legal protection of their bodily freedom and autonomy, women are little more than procreative slaves. It is essential to respect women's bodily freedom and autonomy because it is simply wrong to subject women to forced reproduction; it is wrong to use women as a means to others' reproductive goals. Such treatment violates their personhood. (A similar claim might also be made about men with regard to the use of men's gametes, but the fact that gestation is a female condition makes respect for bodily freedom and autonomy particularly significant for women.) Thus, there are both consequentialist and deontological justifications for reproductive rights. Those rights, properly delineated and understood, provide the foundation for the ethics of procreation. They are a necessary (though not sufficient) criterion for evaluating procreative decisions. And, as I argue in later chapters,

because reproductive rights are foundational, they cannot be disregarded or voided. Nonetheless, to say that essential reproductive rights must be recognized does not imply that there is an unlimited right to reproduce. Respect for bodily freedom and autonomy does not constitute a license to procreate. For example, as Onora O'Neill points out, if, like Jean JacquesRousseau and his mistress, individuals simply produced infants and then abandoned them at a foundling hospital, we would not be inclined to say such persons were justifiably exercising their right to procreate (1979, 25). I propose that the general idea of reproductive rights should be analyzed in terms of two distinct prima facie rights—that is, rights that are "conditional on not being overridden by other relevant moral principles" (Frankena and Granrose 1974, 80). These prima facie rights are the right to reproduce and the right not to reproduce. Prima facie rights are strong entitlements that make legitimate calls on the behavior of others—either individuals or the state (or both)—but they are not defeasible. They can sometimes, though perhaps only rarely, be superseded by other moral requirements—for example, when rights are in conflict with each other. Let's look first at the right to reproduce. In my early work on reproductive ethics and social policy (Overall 1987, 1993), I distinguished between two different kinds of right to reproduce: the positive or welfare right and the negative or liberty right. The Right to Reproduce in the Positive or Welfare Sense

The prima facie right to have children in the positive or welfare sense is an entitlement to have a child by any means one may choose and to be provided with all possible assistance in reproduction. Although there arguably is such a right, and it provides important protections, it is nonetheless limited in some important ways. Reproductive services should be seen for what they are: a category of health care. As a category of health care, reproductive services ought to be available to and accessible by potential patients on medical grounds. Social discrimination in the provision of reproductive services is just as unfair and unjustified as social discrimination in the provision of other forms of health care, and anyone who opposes social discrimination for the latter should oppose it for the former. Hence, an individual patienth gender identity, race, sexual orientation, marital status, and other identity characteristics are not justified criteria either for providing or

220

5 Reproductive Ethics

for refusing medical treatment because they are not relevant to health care for the individual's medical condition. For example, if a married heterosexual woman is entitled to reproductive treatment X, then so is a single woman or a lesbian woman. The single woman and the lesbian woman should not be denied access to donor insemination or IVF just because of their marital status or sexuality. In general, the social identity of potential patients is irrelevant to whether they should be permitted reproductive services. For a number of reasons, it is morally wrong for the medical profession to use social identity criteria to determine who is entitled to become a parent, who is competent to become a parent, or, for that matter, who lacks the characteristics for parenthood. First, physicians are not trained or qualified to determine who is and is not competent to become a parent. A physician's role is to provide medical care, not to serve as a gatekeeper to parenthood. There is no reason to believe that physicians have a special ability to discern parenting capacities. Moreover, attempting to be such a gatekeeper would compromise physicians' moral responsibilities to serve their patients' best interests because it would make reference to standards that are independent of and even incompatible with the patient's health care. Second, individuals who do not seek reproductive services are not subjected to any test or qualification for parenting. If there is no screening system for prospective parents who do not need reproductive services, it seems unjust to subject to screening those who have the misfortune of needing medical help. Although some philosophers have argued that there should be a licensing system for prospective parents (see, for example, Tittle 2004), such a system would be a severe imposition on people's bodily freedom and autonomy, would be grotesquely unwieldy, would consume enormous resources that can better be used to support good health and good parenting, and would not likely improve the care of children. Third, there is no clear empirical method for predicting which individuals will be good parents and which will not. Certain factors no doubt make parenting easier (perhaps training about children's development and needs) or harder (serious socioeconomic deprivation). If so, then instead of requiring physicians to make ad hoc predictions in individual cases, it would be better for a society that is concerned about the quality of parenting to provide good education, health care, and socioeconomic support for all its citizens, including children, parents, and prospective parents.

Fourth, there is no empirical evidence that social identity in itself either compromises or enhances parenting abilities. It is unjustified to generalize about all persons with a particular identity. Contrary to social conservatives' protests, single women have been successfully rearing children in greater and greater numbers for the past half-century. The evidence shows that lesbians' children turn out fine (Gartrell and Bos 2010). And, in contrast, some married, heterosexual parents physically and psychologically abuse their children. One's age, impairments, and health history are similarly in principle relevant to one's eligibility for reproductive services only on medical grounds— that is, if these characteristics demonstrably make such services either ineffective or dangerous to the individual seeking them. It is not up to physicians to make judgments about whether a patient's age or impairment renders her ineligible for motherhood. (In chapter 7, however, I discuss the significance of one's age for the moral justification of one's procreative choices, and in chapter 8 I say more about parenting decisions by persons with impairments.) Thus, if a person is simply not healthy enough to undergo a particular medical treatment, if there is little or no prospect that the medical treatment will help the individual, or if it in fact will even harm the individual, then it is legitimate not to provide the treatment. For example, if the body of an older woman or a woman with an impairment does not respond to hormonal hyperstimulation, then it is medically unjustified to persist in attempting to extract ova from her. The cessation of treatment would be based on empirical evidence of likely ineffectiveness or harm, not on the mere fact that the patient is a certain age or has a particular impairment. Hence, the right to reproduce in the positive sense can be construed as protecting individuals against unjustified discrimination in access to reproductive services, which ought not to be denied to individuals on any grounds but medical ones. However, because the positive right to reproduce draws upon others' material resources and services, it cannot be boundless. To what sorts of reproductive services might a positive right to reproduce allow access? My answer here should be considered somewhat tentative and suggestive only, for much depends on two factors: a society's general level of affluence and the resulting availability of money and resources for health care, and the kinds of decisions the society may make about health-care goals and priorities. For example, societies with scarce resources are entitled to decide whether they will provide IVF services. In some

Overall: Reproductive Freedom, Autonomy, and Reproductive Rights

societies, other services may legitimately be considered more immediately valuable, including preventive health services, pregnancy care, infant and maternal care, medical treatments for various impairments, and so on. I do not attempt here to determine how such priorities might be evaluated. The issue is complex and difficult; debating the relative value of different health-care goals and priorities would require a book of its own. My discussion in the rest of this section is intended simply to indicate the kinds of considerations that would be relevant to determining the health-care implications of the positive right to reproduce, given adequate resources and a careful assessment of priorities. I suggest that the right to reproduce requires minimally the provision of health-care services for healthy pregnancy and delivery, including well-supported home birth and midwifery when these options are chosen. Depending on the society's assets and resources, the positive right to reproduce may also entitle patients to reproductive services that alleviate infertility, including the surgical repair of damaged fallopian tubes in women, the reversal of a vasectomy in men, and, for women, intrauterine sperm injection or trials of IVF with a willing, competent, informed, and autonomous partner.2 All of these procedures increase the likelihood that an individual may become a parent, although of course they do not guarantee anyone a baby, especially in the case of IVF, for which success rates are still much less than 50 percent .3 A society may nonetheless be entitled to place appropriate limits on how some reproductive services are provided. IVF is a good example. Having access to IVF does not automatically entitle a woman to determine how the procedure will be carried out, for that issue is a medical matter. Being provided with IVF does not give a woman patient the entitlement to have as many embryos as she wishes inserted into her uterus. Although inserting a large number of embryos increases the woman's chances of becoming pregnant, it also exacerbates the probability of higher-order multiples (HOM)—triplets, quadruplets, quintuplets, sextuplets, and more. Such pregnancies are dangerous for the woman and for the fetuses; Homs require vastly more health-care services—during gestation and birth, in the neonatal period, and often beyond—than do singletons. (Think of the case of American "Octomom" Nadya Suleman, who after IVF ended up gestating and delivering eight fetuses.) Hence, a hospital is entitled to set a policy limiting the number of embryos implanted—perhaps to two or three. Second, access to IVF does not include an automatic entitlement to use the procedure at any age. Despite the

fact that some women have become pregnant and given birth, thanks to IVF, at the age of sixty or older, there are good medical reasons, related to the woman's health risks, her ability to sustain a pregnancy, and perhaps her offsprings prospective physical health, not to permit IVF for every postmenopausal woman who may want it. I am not saying that no "older" woman should ever have access to IVF; I am simply reiterating my earlier point that health-care professionals are entitled to use medical criteria to determine the patient's suitability for the procedure. The situation must be decided on a case-bycase basis according to the woman's medical condition. Third, there is no automatic entitlement to have as many individual IVF treatments as the patient may wish for. In some cases, a large number of treatments may turn out to be medically futile. If so, then there is a significant question of health-care resource allocation. If a society does determine that it will offer IVF, as I have argued, it must not discriminate on irrelevant grounds in providing access to such services, but it is entitled to place limits on the number of treatments an individual woman may have. Doing so is in part a matter of medical effectiveness but in part also a matter of straightforward fairness with respect to a limited and expensive service. A woman may have no more than x treatments simply because other women also need IVF treatments, and if one woman has many treatments, then others will have fewer. Thus, issues of fairness among individuals with similar needs dictate that no woman can just demand a large number of IVF treatments as part of her right to reproduce and expect automatic compliance. So far, then, I have suggested that there is a positive right to reproduce and that in a society with good health-care resources, exercising that right might include access to a variety of treatments intended to support pregnancy and birth and to enhance, repair, or restore fertility. Access to these services should not be denied based on social identity characteristics such as marital status or sexual orientation, nor should it be denied on the basis of age, health status, or impairment unless the latter are medically relevant to the provision of services. But the right to reproduce in the positive or welfare sense has one additional important and indefeasible limit: it does not and cannot include an entitlement to consumerlike activities such as buying and selling gametes or purchasing the services of a contract procreator, because no one has a right to the use of others' sperm or ova or a right to have another woman gestate and give birth to a baby for them. There can be no entitlement to such activities because to satisfy such supposed entitlements would require that one or more other people must provide bodily products or services. There cannot

11 .111

1111.11

5 Reproductive Ethics

be a moral requirement for anyone to provide sperm or eggs or to serve as a gestator, such a requirement would violate other women's or men's bodily freedom and autonomy.4 Respect for women's and men's bodily freedom and autonomy obviates the possibility of compelling some women to provide ova or to undergo pregnancy or of compelling men to provide sperm, even to facilitate procreation by other women. Some individuals may (for whatever reasons, including perhaps the dubious desire for genetic "immortality") wish to donate their gametes (sperm or ova). Such donations can be very helpful to people who are struggling with infertility. But the donator cannot expect a guarantee that the gametes will actually be used or even that a health-care facility should have to take the gametes if there are medical reasons not to. If a particular jurisdiction offers sperm or egg banks, donors to which have given willing, competent, informed, and autonomous consent, then women who are candidates for using donated sperm or eggs ought not to be discriminated against on irrelevant grounds if they seek access to them. But no individual has an obligation to donate or sell their gametes; hence, no other individual has a right to be given them or buy them. If a particular society does not offer gamete donation, or if it does offer the service and no one chooses to supply gametes, then, nonetheless, the rights of those who may want or need others' gametes have not been violated. There is never a guarantee that an individual will be able to have a baby, and it is a mistake to say that a woman (or a man) has a "right to a baby" or a "right to a family" This claim is true for several reasons. There is no guarantee of obtaining a baby biologically even through the use of one's own gametes and pregnancy, for some women cannot conceive, some pregnancies end prematurely, and some births fail to yield a live infant. But there are also no guarantees morally and politically, for people are not entitled to keep a child, even a child that came out of their body, whom they abuse. There is no guarantee through adoption because no one has a right to someone else's baby, although at times some women are generous enough to give up their infants for adoption. There is no guarantee through the use of someone else's gametes because no one has a right to someone else's gametes and no one owes an in-need person their gametes. There is no guarantee through contract pregnancy because no one has a right that a woman will provide gestational services and no woman owes another person the child of her pregnancy. Insofar as reproduction for some people requires the use of and access to other individuals' bodies and bodily products, there is no positive

right to reproduce in that manner.5 Saying this does not mean that all of these activities are necessarily wrong, only that they cannot be justified as the exercise of a moral right, for no one has an obligation to provide gametes or gestational services. An additional limit to the prima facie right to reproduce in the positive sense is that individuals do not have a right to the best possible baby or even to the kind of baby they most want. Although women certainly are entitled to good prenatal and neonatal care, there can be no guarantees about the characteristics of the resulting infant. Ben Bova claims, "An individuaN desire to produce offspring that are as close to the ideal that the parent can envision seems well within the rights of any citizen" (1998, 194). He makes this claim by contrast to the idea that a government would be entitled to run a eugenics program, an idea that he rejects. But individuals don't have an entitlement to engage in their own private eugenics program either. Although there is a strong case for helping people with infertility to increase their chances of procreating, the case for helping people to obtain a "designer baby" is much weaker. People are, of course, free to choose among consenting adults with whom to reproduce—or they should be. Such decisions in themselves have an effect on the kind of child they will have. But whether they are entitled to the complex genetic services—including IVF and embryo screening—that would enable them to make more precise choices is a further issue and a complicated one. Much depends on the costs of such services compared to other health-care costs. Societies have to make policy decisions as to which kinds of health care should be provided. Consider, for example, the possible funding of hearing aids for the hearing impaired, prostheses for those missing all or part of a limb, and wheelchairs for those whose mobility is reduced. It is not at all obvious that these items are less important than funding the capacity to choose the best baby possible. Even in a nation where health care is not publicly funded, there will still be questions as to what sorts of medical research, education, and training should be supported and encouraged; as I pointed out earlier, choices about health-care priorities have to be made. The potential justification of seeking to produce the best possible baby also depends on the reason that screening is needed. It would be far more controversial to screen embryos for intelligence, beauty, or athletic ability than it would ever be to screen them for the presence of diseases, such as Tay-Sachs, that will cause unrelieved suffering. In chapter 7, I discuss whether parents have any obligation to enhance their offspring by choosing the best available embryo, and in chapter 8

Overall: Reproductive Freedom, Autonomy, and Reproductive Rights

I return to the topic of prenatal screening when I explore whether there is an obligation in some circumstances not to procreate because of the fetus's condition.

The Right to Reproduce in the Negative or Liberty Sense Although there are several actual and potential limits on the right to reproduce in the positive sense, there are no comparable limits on the prima facie right to reproduce in the negative or liberty sense. This right is the entitlement not to be interfered with in procreation, and it is primarily this reproductive right that is asserted in the consensus statement from the UNICPD quoted earlier. The right to reproduce in this sense means the freedom to decide when, where, and with whom one will have one's biological children and how many. Of course, no man is entitled to have his female partner produce a baby for him, any more than a woman has an entitlement to be given her male partnerb sperm. But if two people agree to reproduce together using their own gametes, they are entitled not to be interfered with or prevented by third parties. What, then, about the right to reproduce in the negative sense with respect to women in a same-sex relationship? They have the same reproductive rights as persons in heterosexual relationships; the state has no right to prevent them from having children if they choose. But their situation is, of course, reproductively more complex. They need the collaboration of one more person. As I explained earlier, no one has an entitlement to another personb reproductive gametes. But if a third person agrees willingly, competently, and autonomously to collaborate by providing his sperm, then women in a same-sex relationship are entitled not to be interfered with in their decisions about when, where, and with whom to procreate or about how many children to have. Honoring the right to reproduce in the negative sense would also require the provision of at least a minimally healthy physical environment and working conditions that do not damage or compromise one's procreative capacities. People are entitled to be protected from environments and employments that are dangerous to their health, including their reproductive health. It would also mean no interference in or limitation of one's biological procreative behavior. Thus, respecting the negative right to reproduce requires that there can be no forced sterilization, no coercive contraception,6 no forced marriage or prostitution, no racist marriage or domestic partnership laws, no forced abortions, and no forced caesarians (all of which are, of course, wrong on other grounds, too).

The Right Not to Reproduce Women and men also have a clear right not to reproduce. The right not to reproduce means that human beings are entitled not to be compelled to reproduce against their will. No one should be compelled to give away or sell their gametes or embryos against their will. No woman should be compelled to undergo pregnancy or continue it against her will. This right is implicit in what I said earlier, in discussing limits on the right to reproduce, about the absence of entitlement to other people's reproductive resources and services. Indeed, I suggest that the right not to reproduce is morally more basic than the right to reproduce, at least in the sense that it is strongly unethical to violate one individualb right not to reproduce in order to serve another individualb right to reproduce. As I noted in chapter 1, pregnancy sometimes results because contraception is simply not used, whether because it is not available or not chosen or because the individuals are unaware of it or the need for it. In other cases, heterosexual intercourse is voluntary, but although children are not wanted, there is no effective contraception or contraceptive methods are inadequate or are incorrectly used. And conception is sometimes unchosen because it is the result of rape. Therefore, respecting the right not to reproduce requires comprehensive education about sex and procreation, access to safe and effective contraception and to timely and effective abortion services, and protection from compulsory heterosexual intercourse, whether by assault or by forced marriage. Of course, just as there is no guarantee that one will receive a baby after pregnancy and birth, so also there is no guarantee that contraception will work. That is a fact that women must always take into account. It is also a fact that men must accept, as I show in chapter 3. The justification for recognizing the negative right to reproduce is simply that if it is not respected, then women especially, but men too, have no choice of or control over their biological reproductive behavior, and therefore their bodily freedom and autonomy are violated. All women, like men, have a right not to reproduce—that is, a right not to be forced to reproduce, whether through sexual slavery, denial of access to contraception, or lack of access to abortion.' Because some women have historically been forced against their will into marriage or prostitution, women's right not to reproduce is important whatever their sexual orientation may be. Part of what is morally wrong about forced marriage and prostitution is that it makes women vulnerable to unwanted procreation; it violates their right not to reproduce. Any violation of the right not to reproduce is serious. No individual should be forced to provide her or his gametes to another person, whether for procreation

II .111

224

11141/1

5 Reproductive Ethics

or for research. No woman should ever be forced to undergo pregnancy against her will. In the words of D.S. Hutchinson, in unwanted pregnancy "the body is being altered for a purpose that does not belong to the person. But this is a case where the body is not just used, but transformed. The transformation is different in kind from the bodily changes involved in ordinary actions, for it is a transformation of a very special nature. It results in two people where there was previously only one." In unwanted pregnancy, the woman's body "is treated as a means to a project that is alien to [her;] it involves an alienation of the body from the person." It thereby undermines her integrity because "a commitment to one's body is given in the human condition" (1982, 71). If the right not to reproduce is not respected, then the result is reproductive slavery: the compulsory and unwilled use of people's bodies for procreative purposes—whether they are other individuals' or the states purposes. (Reproductive coercion— the production of babies through the forced, often violent use of women's bodies—is a standard feature of enslaved societies and is part of what makes enslavement morally execrable.) People need to be able to protect their reproductive capacities and not have them exploited or coerced to serve purposes that are not their own. As I argued at the beginning of this chapter, genuine reproductive freedom is one of the essential building blocks of a just and flourishing society. Conclusion

What does this discussion o f reproductive rights tell us about the "why have children?" ethical decision? The right not to reproduce and the right to reproduce in the negative sense (the entitlement to freedom from interference) are grounded in general human interests; people need protection from compulsory procreation as much as (or perhaps even more than) they need protection from denied procreation. The appeal to the right not to reproduce also provides important protections to women (as well as to men) insofar as it provides a strong prima facie argument against any obligation to procreate. If a woman has a right not to reproduce, then she is prima facie not obligated to reproduce. In chapters 4 and 5, I investigate whether there are any factors that can create such an obligation despite the prima facie right not to reproduce. Moreover, the appeal to the right to reproduce in the negative sense is inadequate to serve as a complete ethical justification for choosing to have children because, as subsequent chapters show, many other factors are morally relevant to procreative decisions. The fact that interfering in an individual's procreative behavior violates the individual's right to reproduce does not by

itself morally justify the individual's decision to reproduce. At the same time, the right to reproduce in the positive sense is limited by a variety of factors and does not guarantee a baby to anyone. In other words, a general appeal to reproductive rights cannot be used to vindicate a wholesale entitlement to procreate, to be given a baby, or to parent. Among the reasons for this claim, the most obvious is that in making the decision to procreate, a new human being will be brought into existence. As S.L. Floyd and D. Pomerantz put it in regard to a putatively unlimited right to self-determination, "It does not take into account the impossibility of a being's actual or hypothetical consent to becoming a child" (2004, 232). Of course, by the very nature of the situation, the prospective child cannot consent. And the impossibility of the child's consent does not make all procreation wrong. At the very least, however, the prospective well-being of the child-to-be is surely relevant to the ethics of choosing to have children. That issue is one to which I return several times. A second reason that the right to reproduce, whether positive or negative, is not by itself sufficient to justify procreation has to do, broadly speaking, with resource issues, including the constraints generated by the growing human population and the planet's limited carrying capacity. I explore this latter set of issues in the penultimate chapter. A third reason that the right to reproduce is not by itself sufficient to justify procreation is provided by the need for a procreative partner. It still takes two to make a baby, if not by sexual intercourse, then by other means of combining egg and sperm. Several times in this chapter, I have added the caveat that one must have a willing, informed, competent, and autonomous reproductive partner. No one has a right to another person's sexual services or procreative capacities, and no one has a right of access to another person's gametes. No one is entitled to obtain them by force or by subterfuge. Consent to providing reproductive capacities and gametes is essential. The content and structure of reproductive rights leave open the possibility that there might sometimes be an obligation not to reproduce. Acting on one's rights, including the right to reproduce, is not always morally justified. One may have the right to do something that is nonetheless not morally justified. Reproductive rights are a necessary but not sufficient condition for justifying choosing to have children. Therefore, the next topic to be considered is the situation where potential parents disagree—in particular, where they disagree as to whether to continue a pregnancy or not. In such cases, one or the other partner does not consent to procreate. How might such disputes be resolved?

Overall: Reproductive Freedom, Autonomy, and Reproductive Rights

Notes 1. 2.

3.

4.

5.

For a history of the development of the concept of reproductive rights, see Kates 2004. How such services should be funded is a separate issue that I cannot explore here. Nonetheless, it seems plausible to me that if these services are part of the health-care system and especially if prospective patients are not wealthy, then the state should pay for them. (It is also often argued that public financing of IVF, in particular, will reduce the incidence of gestation of Homs (triplets, quadruplets, and so on], and hence of all the medical problems (including blindness, cerebral palsy, and death] and resulting costs borne by the public that are associated with such infants (Barwin 2009] because if women do not have to pay for IVF themselves, then they will be less likely to expect that, to increase the odds of becoming pregnant, large numbers of fertilized eggs be implanted at each attempt but will be content with the implantation of only one or two at a time.) In Canada, for example, the overall live birth rate with IVF is 27 percent per cycle (Bouzayen and Eggertson 2009, 243). Another moral reason for rejecting commercial contract pregnancy is that it is the sale of babies (Overall 1993). However, I set this argument aside. Philosopher Elisabeth Gedge has suggested that my position may compromise an analogous claim on behalf of a positive right to general health care (personal communication, June 19, 2008) and such an implication is morally problematic. However, there are always limits that must be placed on the availability of services, even if those services are essential to life itself. For example, a system of blood and organ donation is highly desirable,

6.

7.

but no individual, no matter how needy, has a right to the use of the blood or organs from another person if the latter person does not want to donate them. If there is a system of blood and organ donation in which donors participate willingly, knowledgeably, and consensually, then a needy patient is of course entitled to be served by that system if he or she meets the criteria of medical eligibility. But the system cannot compel donors to participate, and there is not a right to the use of other persons' body parts. Note, however, that I am not saying that women (or men, for that matter) are not entitled to set, as a condition for their participation in heterosexual sexual activity, the requirement that their partner use contraception. We are entitled to do so in order to protect (insofar as the effectiveness of contraception permits) our procreative future. In such cases, the use of contraception is not coerced: because each participating partner is entitled to say "no" to the sexual activity if he or she does not want to use contraception, there is no compulsion. In this book, I do not discuss the general arguments for and against the justification of abortion. First, there is a huge body of literature about the abortion issue, so attempting to summarize it here would take me too far from my original topic. Second, I think the question whether to abort an embryo or fetus that has already come into existence is different from the question primarily at issue in this book, which is whether to conceive that embryo or fetus in the first place. However, in chapter 7 I make a few comments about the possibility that there may be an obligation to have an abortion.

References Barwin, Norman. 2009. A Private Issue in Need of a Public Solution. The Globe and Mail May 21, IC2. Bouzayan, Renda, and Laura Eggerston. 2009. In Vitro Fertilization: A Private Matter Becomes Public. Canadian Medical Association Journal 181 (5):243. Floyd, S.L, and D. Pomerantz. 2004. Is There a Natural Right to Have Children? in Should Parents Be Licensed? Debating the Issues, ed. Peg Tittle, 230-32. Amherst, NY: Prometheus Books. Frankena, William K., and John T. Granrose, eds. 1974. Introductory Readings in Ethics. Englewood Cliffs, NJ: Prentice Hall. Gartrell, Nanette, and Henry Bos. 2010. US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17 year old Adolescents. Pediatrics 126 (10):28-36.

Hannan, Sarah, and Richard Vernon. 2008. Parental Rights: A Role-Based Approach. Theory and Research in Education 6 (2):173-89. Hutcheson, D.S. 1982. Utilitarianism and Children. Canadian Journal of Philosophy 12 (1):61-73. Kates, Carol A. 2004. Reproductive Liberty and Overpopulation. Environmental Values 13:51-79. O'Neill, Onora. 1979. Begetting, Bearing, and Rearing. In

Having Children: Philosophical and Legal Reflections on Parenthood, ed. Onora O'Neill and William Ruddick, 25-38. New York: Oxford University Press. Overall, Christine. 1987. Ethics and Human Reproduction: A Feminist Analysis. Boston: Allen and Unwin. Overall, Christine. 1993. Human Reproduction: Principles, Practices, and Policies. Toronto: Oxford University Press.

u

1

II

11

al

III1Jj

5 Reproductive Ethics

Preconception Gender Selection' John A. Robertson

Advances in genetics and reproductive technology present prospective parents with an increasing number of choices about the genetic makeup of their children. Those choices now involve the use of carrier and prenatal screening techniques to avoid the birth of children with serious genetic disease, but techniques to choose nonmedical characteristics will eventually be available. One nonmedical characteristic that may soon be within reach is the selection of offspring gender by pre-conception gender selection (PGS). Gender selection through prenatal diagnosis and abortion has existed since the 1970s. More recently, preimplantation sexing of embryos for transfer has been developed (Tann and Handyside 1993; The Ethics Committee of the American Society of Reproductive Medicine 1999). Yet prenatal or preimplantation methods of gender selection are unattractive because they require abortion or a costly, intrusive cycle of in vitro fertilization OW and embryo discard. Attempts to separate X- and Y-bearing sperm for preconception gender selection by sperm swim-up or swim-through techniques have not shown consistent X- and Y-sperm cell separation or success in producing offspring of the desired gender. The use of flow cytometry to separate X- and Y-bearing sperm may turn out to be a much more reliable method of enriching sperm populations for insemination. Laser beams passed across a flowing array of specially dyed sperm can separate most of the 2.8% heavier X- from Y-bearing sperm, thus producing an X-enriched sperm sample for insemination.2 Flow cytometry has been used successfully in over 400 sex selections in rabbit, swine, ovine, and bovine species, including successive generations in swine and rabbit (Fugger et al. 1998). A human pregnancy was reported in 1995 (Levinson, Keyvanfar, and Wu 1995). The United States Department of Agriculture (USDA), which holds a patent on the flow cytometry separation process, has licensed the Genetics and IVF Institute in Fairfax, Virginia, to study the safety and efficacy of the technique for medical and "family balancing" reasons in an institutional review board—approved clinical trial? In 1998 researchers at the Institute reported a 92.9% success rate for selection of females in 27 patients, with most fertilizations occurring after

intrauterine insemination (Fugger et al. 1998). A lower success rate (72%) was reported for male selection.4 At this early stage of development much more research is needed to establish the high degree of safety and efficacy of flow cytometry methods of PGS that would justify widespread use. With only one published study of outcomes to date, it is too soon to say whether the 92% success rate in determining female gender will hold for other patients, much less that male selection will reach that level of efficacy. Animal safety data have shown no adverse effect of the dye or laser used in the technique on offspring, but that is no substitute for more extensive human studies (Vidal et al. 1999). In addition, if flow cytometry instruments are to be used for sperm separation purposes, they may be classified as medical devices that require US Food and Drug Administration (FDA) approval. Finally, the holder of the process patent—the USDA—will have to agree to license the process for human uses. If further research establishes that flow cytometry is a safe and effective technique for both male and female PGS, and regulatory and licensing barriers are overcome, then a couple wishing to choose the gender of their child would need only provide a sperm sample and undergo one or more cycles of intrauterine insemination with separated sperm.5 A clinic or physician that offers assisted reproductive technologies (ART) and invests in the flow cytometry equipment could run the separation and prepare the X- or Y-enriched sperm for insemination, or it could have the sperm processed by a clinic or firm that has made that investment. Flow cytometry separation would not be as cheap and easy as determining gender by taking a pill before intercourse, but it would be within reach of most couples who have gender preferences in offspring.6 Demand for Preconception Gender Selection

Unknown at present is the number of people who have offspring gender preferences robust enough to incur the costs and inconvenience of PGS. Although polls have often shown a preference for firstborn males, they have not shown that a large number of couples would be willing to forego coital conception in order to select the gender of their children. If PGS proves to be safe and effective, however, it may be sought by two groups of persons with gender preferences. One group would seek PGS in order to have a child of a gender different from that of a previous child or children. A preference for gender variety in offspring

Source: From The American Journal of Bioethics 1 (2001), 1, 2-9. Reprinted with permission.

Robertson: Preconception Gender Selection

would be strongest in families that have already had several children of one gender. They may want an additional child only if they can be sure that it will be of the gender opposite to their existing children.' Couples who wish to have only two children might use PGS for the second child to ensure that they have one child of each gender. If social preferences for two-child families remain strong, some families may use PGS to choose the gender of the second child. A second group of PGS users would be those persons who have strong preferences for the gender of the first child. The most likely candidates here are persons with strong religious or cultural beliefs about the role or importance of children with a particular gender. Some Asian cultures have belief systems that strongly prefer that the firstborn child be a male. In some cases the preference reflects religious beliefs or traditions that require a firstborn son to perform funeral rituals to assure his parents' entrance into heaven (for a discussion of son preferences in India and China, see Macklin 1999, 148-51). In others it simply reflects a deeply embedded social preference for males over females. The first-child preference will be all the stronger if a one-child-per-family policy is in effect, as occurred for a while in China (Greenlagh and Li 1995, 627). While the demand for PGS for firstborn children is likely to be strongest in those countries, there has been a sizable migration of those groups to the United States, Canada, and Europe.8 Until they are more fully assimilated, immigrant groups in Western countries may retain the same gender preferences that they would have held in their homelands. Other persons with strong gender preferences for firstborn children would be those who prize the different rearing or relational experiences they think they would have with children of a particular gender. They may place special value on having their firstborn be male or female because of personal experiences or beliefs. Numerous scenarios are likely here, from the father who very much wants a son because of a desire to provide his child with what he lacked growing up to the woman who wants a girl because of the special closeness that she thinks she will have with a daughter (Belkin 1999). The Ethical Dilemma of Preconception Gender Selection The prospect o f preconception gender selection appears to pose the conflict—long present in other bioethical issues—between individual desires and the larger common good. Acceding to individual desires about the makeup of children seems to be required by individual

autonomy. Yet doing so leads to the risk that children will be treated as vehicles of parental satisfaction rather than as ends in themselves and could accelerate the trend toward negative and even positive selection of offspring characteristics. The dilemma of reconciling procreative liberty with the welfare of offspring and families will only intensify as genetic technology is further integrated with assisted reproduction and couples seek greater control over the genes of offspring. Arguments for Preconception Gender Selection The strongest argument for preconception gender selection is that it serves the needs of couples who have strong preferences about the gender of their offspring and would not reproduce unless they could realize those preferences. Because of the importance of reproduction in an individual's life, the freedom to make reproductive decisions has long been recognized as a fundamental moral and legal right that should not be denied to a person unless exercise of that right would cause significant harm to others (Robertson 1994, 22-42). A corollary of this right, which is now reflected in carrier and prenatal screening practices to prevent the birth of children with genetic disease, is that prospective parents have the right to obtain preconception or prenatal information about the genetic characteristics of offspring so that they may decide in a particular case whether or not to reproduce (Robertson 1996, 424-35).9 Although offspring gender is not a genetic disease, a couple's willingness to reproduce might well depend on the gender of expected offspring. Some couples with one or more children of a particular gender might refuse to reproduce if they cannot use PGS to provide gender variety in their offspring or to have additional children of the same gender (E.E Fugger, personal communication to author). In other cases they might have such strong rearing preferences for their firstborn child that they might choose not to reproduce at all if they cannot choose that child's gender. Few persons contemplating reproduction may fall into either group, but for persons who strongly hold those preferences, the ability to choose gender may determine whether they reproduce. In cases where the gender of offspring is essential to a couple's decision to reproduce, the freedom to choose offspring gender would arguably be part of their procreative liberty (Robertson 1996, 434). Since respect for a right is not dependent on the number of persons asserting the right, they should be free to use a technique essential to their reproductive decision

II

41/1

111611

5 Reproductive Ethics

unless the technique would cause the serious harm to others that overcomes the strong presumption that exists against government interference in reproductive choice. Until there is a substantial basis for thinking that a particular use of PGS would cause such harms, couples should be free to use the technique in constituting their families. The right they claim is a right against government restriction or prohibition of PGS. It is not a claim that society or insurers are obligated to fund PGS or that particular physicians must provide it. Arguments against Preconception Gender Selection

There are several arguments against preconception gender selection. Although such methods do not harm embryos and fetuses or intrude on a woman's body as prenatal gender selection does, they do raise other important issues. One concern is the potential of such techniques to increase or reinforce sexism, either by allowing more males to be produced as first or later children or by paying greater attention to gender itself. A second concern is the welfare of children born as a result of PGS whose parents may expect them to act in certain gender-specific ways when the technique succeeds but who may be disappointed if the technique fails. A third concern is societal. Widely practiced, PGS could lead to sex-ratio imbalances, as have occurred in some parts of India and China due to female infanticide, gender-driven abortions, and a one-childper-family policy (Sen 1990). Finally, the spread of PGS would be another incremental step in the growing technologization of reproduction and genetic control of offspring. While each step alone may appear to be justified, together they could constitute a threat to the values of care and concern that have traditionally informed norms of parenting and the rearing of children. Evaluation of Ethical and Social Issues

Concerns about sex-ratio imbalances, welfare of offspring, and technologizing reproduction may be less central to debates over PGS than whether such practices would be sexist or contribute to sexism. If the number of persons choosing PGS is small or the technique is used solely for offspring gender variety, sex-ratio imbalances should not be a problem. If use patterns did produce drastic changes in sex ratios, self-correcting or regulatory mechanisms might come into play. For example, an overabundance of males would mean fewer females to marry, which would make being male less desirable and provide incentives to increase the number of female births. Alternatively, laws or policies

that required providers of PGS to select for males and females in equal numbers would prevent such imbalances.1° A serious threat of a sex-ratio imbalance would surely constitute the compelling harm necessary to justify limits on reproductive choice. It may also be difficult to show that children born after PGS were harmed by use of the technique. Parents who use PGS may indeed have specific gender role expectations of their children, but so will parents who have a child of a preferred gender through coitus. Children born with the desired gender after PGS will presumably be wanted and loved by the parents who sought this technique. Parents who choose PGS should be informed of the risk that the technique will not succeed and counseled about what steps they will take if a child of the undesired gender is born.' If they commit themselves in advance to the well-being of the child, whatever its gender, the risk to children should be slight. However, it is possible that some couples will abort if the fetus is of the undesired gender. PGS might thus inadvertently increase the number of genderselection abortions. Finally, technological assistance in reproduction is now so prevalent and entrenched that a ban on PGS would probably have little effect on the use of genetic and reproductive technologies in other situations. With some form of prenatal screening of fetuses occurring in over 80% of United States pregnancies, genetic selection by negative exclusion is already well installed in contemporary reproductive practice. Although there are valid concerns about whether positive forms of selection, including nonmedical genetic alteration of offspring genes, should also occur, drawing the line at all uses of PGS will not stop the larger social and technological forces that lead parents to use genetic knowledge to have healthy wanted offspring. If a particular technique can be justified on its own terms, it should not be barred because of speculation of a slippery slope toward genetic engineering of offspring traits (for an analysis of the slippery-slope problem with genetic selection, see Robertson 1994, 162-5. Is Gender Selection Inherently Sexist?

A central ethical concern with PGS is the effect of such practices on women, who in most societies have been subject to disadvantage and discrimination because of their gender. Some ethicists have argued that any attention to gender, male or female, is per se sexist and should be discouraged, regardless of whether one can show actual harmful consequences for women (see Grubb and Walsh 1994; and Wertz and Fletcher 1989). Others have argued that there are real differences

Robertson: Preconception Gender Selection

between male and female children that affect parental rearing experiences and thus legitimate nonsexist reasons for some couples to prefer to rear a girl rather than a boy or vice versa, either as a single child or after they have had a child of the opposite gender. To assess whether PGS is sexist, we must first be clear about what we mean by sexism. The Compact OED (1991, 1727) defines sexism as "the assumption that one sex is superior to the other and the resultant discrimination practised against members of the supposed inferior sex, especially by men against women." By this definition, sexism is wrong because it denies the essential moral, legal, and political equality between men and women. Under this definition, if a practice is not motivated by judgments or evaluations that one gender is superior to the other or does not lead to discrimination against one gender, it is not sexist. Professor Mary Mahowald, an American bioethicist writing from an egalitarian feminist perspective, makes the same point with a consequentialist twist: Selection of either males or females is justifiable on medical grounds and morally defensible in other situations [emphasis added] so long as the intention and the consequences of the practice are not sexist. Sexist intentions are those based on the notion that one sex is inferior to the other, sexist consequences are those that disadvantage or advantage one sex vis-à-vis the other (2000, 121). In my view, the OED definition, modified by Mahowald's attention to consequences, is a persuasive account of the concept of sexism. If that account is correct, then not all attention to the biologic, social, cultural, or psychological differences between the sexes would necessarily be sexist or disadvantage females. That is, one could recognize that males and females have different experiences and identities because of their gender and have a preference for rearing a child of one gender over another, without disadvantaging the dispreferred gender or denying it the equal rights, opportunities, or value as a person that constitutes sexism. If this conjecture is correct, it would follow that some uses of PGS would clearly be sexist, while others would clearly not be. It would be sexist to use PGS to produce males because of a parental belief that males are superior to females. It would be nonsexist to use PGS to produce a girl because of a parental recognition that the experience of having and rearing a girl will be different than having a boy. In the latter case, PGS would not rest on a notion of the greater superiority of one gender over another, nor, if it occurred in countries that legally recognize the equal rights of women, would it likely

contribute to sexism or further disadvantage women. As Christine Overall, a British feminist bioethicist, has put it, "sexual similarity or sexual complementarity are morally acceptable reasons for wanting a child of a certain sex" (1987, 27; quoted in Mahowald 2000, 117). Psychological research seems to support this position. It has long been established that there are differences between boys and girls in a variety of domains, such as (but not limited to) aggression, activity, toy preference, psychopathology, and spatial ability (Maccoby and Jacklin 1974; Gilligan 1980; Kimura and Hampson 1994; Feingold 1994; Collaer and Hines 1995; and Halpern 1997). Whether these differences are primarily in-born or learned, they are facts that might rationally lead people to prefer rearing a child of one gender rather than another, particularly if one has already had one or more children of a particular gender. Indeed, Supreme Court Justice Ruth Bader Ginsburg, a noted activist for women's rights before her appointment to the Supreme Court, in her opinion striking down a male-only admissions policy at the Virginia Military Institute (United States v. Virginia, 116 S. Ct . 2264 [1996]), noted that: Physical differences between men and women . . . are enduring " [T] he two sexes are not fungible; a community made up exclusively of one [sex] is different from a community composed of both." . . . "Inherent differences" between men and women, we have come to appreciate, remain cause for celebration. Some persons will strongly disagree with this account of sexism and argue that any attention to gender difference is inherently sexist because perceptions of gender difference are themselves rooted in sexist stereotypes. They would argue that any offspring gender preference is necessarily sexist because it values gender difference and thus reinforces sexism by accepting the gendered stereotypes that have systematically harmed women (Grubb and Walsh 1994; and Wertz and Fletcher 1989, 2112). According to them, a couple with three boys who use PGS to have a girl are likely to be acting on the basis of deeply engrained stereotypes that harm women. Similarly, a couple's wish to have only a girl might contribute to unjustified gender discrimination against both men and women, even if the couple especially valued females and would insist that their daughter receive every benefit and opportunity accorded males. Resolution of this controversy depends ultimately on one's view of what constitutes sexism and what actions are likely to harm women. Although any recognition of gender difference must be treated cautiously,

U I

H 3u

11111I

5 Reproductive Ethics

I submit that recognizing and preferring one type of childrearing experience over the other can occur without disadvantaging women generally or denying them equal rights and respect. On this view, sexism arises not from the recognition or acceptance of difference but from unjustified reactions to it. Given the biological and psychological differences between male and female children, parents with a child of one gender might without being sexist prefer that their next child be of the opposite gender. Similarly, some parents might also prefer that their firstborn or only child be of a particular gender because they desire a specific rearing and companionship experience. If it is correct that using PGS for offspring diversity is sexist, then those who deny that biological gender differences exist, or who assume that any recognition of them always reinforces sexism or disadvantages women, will not have carried the burden of showing that a couple's use of PGS for offspring gender variety or other nonintentionally sexist uses is so harmful to women that it justifies restricting procreative choice. Until a clearer ethical argument emerges, or there is stronger empirical evidence that most choices to select the gender of offspring would be harmful, policies to prohibit or condemn as unethical all uses of nonmedically indicated PGS would not be justified. The matter is further complicated by the need to respect a woman's autonomy in determining whether a practice is sexist. If a woman is freely choosing to engage in gender selection, even gender-selection abortion, she is exercising procreative autonomy. One might argue in response that the woman choosing PGS or abortion for gender selection is not freely choosing if her actions are influenced by strong cultural mores that prefer males over females. Others, however, would argue that the straighter path to equal rights is to respect female reproductive autonomy whenever it is exercised, even if particular exercises of autonomy are strongly influenced by the sexist norms of her community (Mahowald 2000, 188). Public Policy and Preconception Gender Selection

Because of the newness of PGS and uncertainties about its effects, the best societal approach would, of course, be to proceed slowly, first requiring extensive studies of safety and efficacy and then at first only permitting PGS for increasing the gender variety of offspring in particular families.13 Only after the demographic and other effects of PGS for gender variety have been found acceptable should PGS be available for firstborn children. However, given the close connection between parental gender preferences for offspring and reproductive

choice, public policies that bar all nonmedical uses of PGS or that restrict it to choosing gender variety in off-

spring alone could be found unconstitutional or illegal. If there are physical, social, and cultural differences between girls and boys that affect the rearing or relational experiences of parents, individuals and couples would have the right to implement those preferences as part of their fundamental procreative liberty. The risk that exercising rights of procreative liberty would hurt offspring or women—or contribute to sexism generally— is too speculative and uncertain to justify infringement of those rights. The claim of a right to choose offspring gender is clearest in the case of PGS for gender variety. If flow cytometry or other methods of PGS are found to be safe and effective, there would be no compelling reason to ban or restrict their nonmedical use by persons seeking gender variety in the children they rear. Couples with one child or several children of a particular gender might, without being sexist or disadvantaging a particular gender, prefer to have an additional child of the opposite gender. ART clinics should be free to proceed with PGS for offspring variety in cases where couples are aware of the risk of failure and have undergone counseling that indicates that they will accept and love children of the dispref erred gender if PGS fails. Clinics providing PGS should also ask couples to participate in research to track and assess the effects of PGS on children and families. The use of PGS to determine the gender of firstborn children is a more complicated question. The choice to have one's first or only child be female has the least risk of being sexist, because it is privileging or giving first place to females, who have traditionally been disfavored.'4 The use of PGS to select firstborn males is more problematic because of the greater risk that this choice reflects sexist notions that males are more highly valued. It is also more likely to entrench male dominance. The danger of sexism is probably highest in those ethnic communities that place a high premium on male offspring, but it could exist independently of those settings. Yet restricting PGS to offspring gender variety and firstborn females may be difficult to justify. Given that individuals could prefer to have a boy rather than a girl because of the relational and rearing experiences he will provide, just as they might prefer a girl for those reasons, it might be difficult to show that all preferences for firstborn males are sexist. Nor could one easily distinguish firstborn male preferences when the couple demanding them is of a particular ethnic origin. Although the risk that firstborn male preferences would be sexist is greatest if the PGS occurred in a country

Robertson: Preconception Gender Selection

in which those beliefs prevailed, the chance that PGS would contribute to societal sexism lessens greatly if the child is reared in a country that legally protects the equal status of women and men. If prohibitions on some or all nonmedical uses of PGS could not be justified and might even be unconstitutional, regulation would have to take different forms. One form would be to deny public or private insurance funding of PGS procedures, which would mean that only those willing to pay out-of-pocket would utilize them. Another form would be for the physicians who control access to PGS techniques to take steps to assure that it is used wisely. If they comply with laws banning discrimination, physician organizations or ART clinics could set guidelines concerning access to PGS. They might, for example, limit its use to offspring gender variety or firstborn female preferences only. As a condition of providing services, they might also require that any couple or individual seeking PGS receive counseling about the risks of failure and commit to rear a child even if its gender is other than that sought through PGS.15 Although such guidelines would not have the force of statutory law, they could affect the eligibility of ART clinics to list their ART success rates in national registries and could help define the standard of care in malpractice cases. Conclusion

The successful development o f flow cytometry separation of X- and Y-bearing sperm would make safe, effective, and relatively inexpensive means of

nonmedical preconception gender selection available for selecting female, if not also for male, offspring. The nonmedical use of PGS raises important ethical, legal, and social issues, including the charge that any or most uses of PGS would be sexist and should therefore be banned or discouraged. Assessment of this charge, however, shows that the use of PGS to achieve offspring gender variety and (in some cases) even firstborn gender preference may not be inherently sexist or disadvantaging of women. Although it would be desirable to have extensive experience using PGS to increase the variety of offspring gender before extending it to firstborn gender preferences, it may not be legally possible to restrict the technique in this way. However, practitioners offering PGS should restrict their PGS practice to offspring gender variety until further debate and analysis of the issues has occurred. In any event, physicians offering PGS should screen and counsel prospective users to assure that persons using PGS are committed to the well-being of their children, whatever their gender. A policy solution that gives practitioners and patients primary control without direct legal or social oversight, although not ideal, may be the best way to deal with new reprogenetic techniques. Society should not prohibit or substantially burden reproductive decisions without stronger evidence of harm than PGS now appears to present. Ultimately, the use of PGS and other reprogenetic procedures will depend on whether they satisfy ethical norms of care and concern for children while meeting the needs of prospective parents.

References Belkin, L. 1999. Getting the girl. New York Times Magazine, 25 July, 26. Chen, D.W. 1999. Asian middle class alters a rural enclave. New York Times, 28 December, Al. Collaer, M.L., and M. Hines. 1995. Human behavioral sex differences: A role for gonadal hormones during early development? Psychological Bulletin 118(1): 55-107. Compact OED, The new ed. 1991. New York: Oxford University Press. Ethics Committee of the American Society of Reproductive Medicine, The. 1999. Sex selection and preimplantation genetic diagnosis. Fertility and Sterility 72(4): 595. Feingold, A. 1994. Gender differences in personality: A metaanalysis. Psychological Bulletin 116(3): 429-56. Fugger, E.F., S.H. Black, K. Keyvanfar, and J.D. Schulman. 1998. Births of normal daughters after microsort sperm separation and intrauterine insemination, in-vitro

fertilization, or intracytoplasmic sperm injection. Human Reproduction 13:2367. Gilligan, C. 1980. In a different voice. Cambridge, MA: Harvard University Press. Glover, J. 1989. Ethics of new reproductive technologies: The Glover report to the European Commission. DeKalb, IL: Northern Illinois University Press. . 1994. Comments on some ethical issues in sex selection 6. Paper presented at International Symposium on Ethics in Medicine and Reproductive Biology, July. [Paper on file with author.] Greenhalgh, S., and J. Li 1995. Engendering reproductive policy and practice in peasant China: For a feminist demography of reproduction. Signs 20:601. Grubb, A., and P Walsh. 1994. Gender-vending II. Dispatches 1 (summer). Halpern, D.F. 1997. Sex differences in intelligence Implications for education. American Psychologist 52(10): 1091-102.

II A

11111.1

5 Reproductive Ethics

Jones, O.D. 1992. Sex selection: Regulating technology enabling predetermination of a child's gender. Harvard Journal of Law and Technology 6:51. Kimura, D., and E. Hampson. 1994. Cognitive pattern in men and women is influenced by fluctuations in sex hormones. Current Directions in Psychological Science 3(2): 57-61. Levinson, G., K. Keyvanfar, and J.C. Wu. 1995. DNA based X-enriched sperm separation as an adjunct to preimplantation genetic testing for the prevention of Xlinked disease. Human Reproduction 10:979-82. Maccoby, E.E., and C.N. Jacklin. 1974. The psychology of sex differences. Palo Alto: Stanford University Press. Macklin, R. 1999. Against relativism. New York: Oxford University Press. Mahowald, M.B. 2000. Genes, women, equality. New York: Oxford University Press.

Overall, C 1987. Ethics and human reproduction. Boston: Allen and Unwin. Robertson, J.A. 1994. Children of choice: Freedom and the new reproductive technologies. Princeton, NJ: Princeton University Press. . 1996. Genetic selection of offspring characteristics. Boston University Law Review 76:421. Sen, A. 1990. More than 100 million women are missing. New York Review of Books 20 December, 61-6. Tarin, J.J., and A.H. Handyside. 1993. Embryo biopsy strategies for preimplantation diagnosis. Fertility and Sterility 59:943. Vidal, E, J. Blanco, E.F. Fugger, et al. 1999. Preliminary study of the incidence of disomy in sperm fractions after microsort flow cytometry. Human Reproduction 14:2987. Wertz, D.C., and J.C. Fletcher. 1989. Fatal knowledge: Prenatal diagnosis and sex selection. Hastings Center Report 19:21.

Notes 1. "Preconception gender selection" (PGs) rather than "preconception sex selection" (Pss) is used throughout this article to convey the importance of the social and psychological meanings with which biologic sex is invested for prospective parents and society generally. Because earlier versions of this article used "sex" rather than "gender," commentators may not have had the opportunity to revise their comments in response to the change. For the discussion at hand, either "sex" or "gender," "Pss" or "PGS" is acceptable. 2. When combined with the X-chromosomes of oocytes, X-bearing sperm can produce only XX or female offspring. Similarly, Y-bearing sperm combined with the X chromosome of oocytes can produce only XY or male offspring. 3. See the study's web page, http://www.microsortnet. This article uses the term "gender variety" rather than "family balancing" to avoid the misconception that a family is "unbalanced" if it has many or only children of one gender. (I am grateful to George Annas for this suggestion.) 4. See the study's web page, http://www.microsortnet. Because Y-bearing sperm are smaller and contain less DNA, there is more chance that the sorting machine will fail to distinguish X's and Y's and thus provide samples that are insufficiently enriched with Y-sperm to give a high chance of having a male child. 5. Presumably, flow cytometry separation of sperm could occur with donor as well as couple sperm. It could also be requested by couples undergoing IVF or intracytoplasmic sperm injection (ics0 who request that the sperm provided be enriched or chosen to effect the gender of choice. 6. The current cost of $1500 per insemination cycle should decrease as further progress in the field occurs. 7. Persons requesting PGS for gender balancing in the Fairfax study had an average of 3.4 children of the same gender

8.

9.

10.

11.

12.

13.

14. 15.

and sought boys and girls in roughly equal numbers (Edward Fugger, personal communication to author, 23 February 2000). See Chen (1999). The article describes immigration of a middle class family from the Indian state of Gujarat to Bridgewater, New Jersey, a suburb 40 miles from New York City It should be emphasized that the right claimed here is a negative right against government interference, not an obligation of a particular provider or public or private insurers to provide those services. See Glover (1994). Professor Glover has apparently changed his position from the more negative one he took in his earlier Ethics of New Reproductive Technologies: The Glover Report to the European Commission (1989, 141-4). See also Jones (1992). The risk arises because flow cytometry separation can only provide a greatly enriched sample of X- or Y-bearing sperm for insemination. It cannot guarantee that every sperm in the sample is either X or Y. Wertz and Fletcher overlook how one could have gender preferences based on perceptions of experiential and rearing differences, rather than on differences in the worth or rights of women, when they assert that any form of gender selection violates the principle of equality between the genders "because it is premised upon a belief in sexual inequality" FDA approval of the safety and efficacy of flow cytometry methods of PGS would also be required before widespread use. Persons taking a more purist approach to sexism would, of course, differ with this assessment. They might also require that consumers agree to participate in research so that policymakers will have reliable information about the uses of PGS and the problems it presents.

Royal Commission on New Reproductive Technologies: Preconception Arrangements

Preconception Arrangements Royal Commission on New Reproductive Technologies [. . .] [P]reconception arrangements raise ethical and legal issues that are neither straightforward nor easy to deal with. As Commissioners listened to the continuing debate about this practice, one conclusion became evident: proponents and opponents are not likely to change each other's minds about the ethical and social dimensions of preconception arrangements. Views on preconception arrangements are based on fundamentally different convictions about human nature and about how the world works or ought to work; therefore, assessments of the actual or potential implications of preconception arrangements for women, for children, for couples, and for our evolution as a society also differ. Commercial Arrangements Using our ethical framework and standards, the Commission finds commercial preconception arrangements offensive on several grounds. First, they offend human dignity by commodifying women's reproductive capacities and commodifying children; they contradict the principle that human reproduction should not be commercialized in any way. Second, we see actual and potential harms for families, for individual women and children, and for specific groups within society. Finally, we believe that public policy that condones or supports the establishment of adversarial relationships is fundamentally flawed; public policy should seek instead to support and encourage humane, non-conflictual family and social relationships. Any attempt to legitimize or support commercial preconception arrangements through public policy would represent the antithesis of this goal. Commodification of Children and Reproduction The fundamentally repugnant aspect of preconception arrangements is that they instrumentalize human beings through the deliberate act of creating a child for the express purpose of giving it up, usually in exchange for money. The premise of commercial preconception contracts is that a child is a product that can be bought and. sold on the market. The moral point of view requires that people be treated as ends in themselves, not as a means to the ends of others. We must therefore

uphold the value of children in and of themselves. Children are not a commodity, nor are they instruments to be used to serve the purposes of others. The commodification of children entailed by preconception arrangements ignores these essential values. Moreover, commercial preconception arrangements commodify women's reproductive functions and place women in the situation of alienating aspects of themselves that should be inherently inalienable. A preconception contract obliges the gestational mother to sell an intimate aspect of her human functioning to provide someone else with a genetically related child; the capacity to become pregnant and bear a child is reduced to a marketable service. We do not allow people to give up their freedom and become slaves, even if they make a choice to do so, because of our collective conviction that this would negate the value we attach to human dignity and the inalienability of the person. Similarly, assigning a commercial value to the human function of reproduction would result eventually in a new and, in our view, undesirable social understanding of the value and dignity of women, their reproductive capacity, and their bodily integrity. Commercial preconception contracts by their nature—the exchange of money for a child—contradict one of the fundamental tenets of the Commission's ethical framework. On these grounds alone, we could recommend prohibition of such arrangements, since we believe that all public policy in this field should be based on the principle of non-commercialization of reproduction. The evidence is clear that in commercial preconception contracts the principal motivation of both the broker and the gestational woman is money. Far from being the idyllic situation portrayed by brokers— gestational woman as "altruistic angel" giving the gift of a child to a couple who is happy but infertile— commercial preconception contracts are business transactions. The child is a product being sold by one party and bought by the other. Harms to Individuals The Commission heard strong arguments that preconception arrangements are detrimental to the autonomy of gestational mothers. We concur. Far from enhancing the gestational woman's autonomy, as some have argued, the practice circumscribes and dictates the gestational woman's behaviour by specifying contractual obligations, including the obligation to be treated

Source: From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 661-93. Reprinted with permission.

X NI I

11

I I ill

Illlll

5 Reproductive Ethics

by medical personnel selected by the commissioning implications of their ability to bear children. We reject couple, to have an abortion if the commissioning this argument. Although they are strongly held values, couple so decides, and to renounce her maternal feel- freedom and autonomy do not include the right to ings and rights even before conception. Again, we be- engage in activities that will result in harms to others, lieve there is clear evidence of the potential for coercion particularly, as in this case, to the child that is evenand exploitation of gestational women because of the tually born; the limits of autonomy become apparent disparities in power and resources between gestational when its exercise will harm others, as the commodificawomen and commissioning couples. tion of children most certainly does. A commissioning couple uses a woman as a vehicle Harm to children born as a result of these arrangeto serve their own ends. As the New York State Task ments cannot be ignored. Commissioners reject the Force on Life and the Law observed, they seek the birth argument that these harms could be outweighed by components of gestation from the gestational woman the opportunity for life, as this argument assumes the while denying the personal, emotional, and psycho- very factor under deliberation—the child's conception logical dimensions of her experiences and self. If she and birth. We concur with the assessment of the New succeeds in denying her emotional responses during York State Task Force on Life and the Law: "The asthis profound experience, she is dehumanized in the sessment for public policy occurs prior to conception process. If she fails, her attachment to the child produ- when the . . . arrangements are made. The issue then ces a conflict that cannot be resolved without anguish is not whether a particular child should be denied life, for all involved. but whether children should be conceived in circumMoreover, informed choice is a necessary compon- stances that would place them at risk." Nor do we see ent of autonomy. As we have seen, however, brokers any practical way of protecting the interests of the child arranging preconception arrangements are not neutral that will eventually be born or even of ensuring that intermediaries; they act in the interests of the commis- they are taken into account in negotiating and concludsioning couple. If the gestational mother has her own ing a preconception arrangement. lawyer, it is often one recommended by the broker. Commercial preconception arrangements do proFurthermore, since so little is known about the psycho- duce benefits—but the benefits are to brokers and social effects of these arrangements on the participants commissioning couples at the expense of the interests and the resulting child, the woman's decision cannot of vulnerable women and of children who had no part be made in light of all the information that might influ- in the arrangement. ence it. All these factors undoubtedly undermine her capacity to exercise informed choice in deciding to Harms to Society enter a preconception arrangement. We agree with those who argued that commercial We also conclude that concerns about negative preconception arrangements have potentially negapsychosocial consequences for the gestational mother tive consequences not only for individuals but also are well founded, particularly because it is impossible for women collectively and for other groups in society. for her to predict, at the time she signs the contract, These arrangements reinforce social attitudes about how she will feel about fulfilling its terms after the motherhood as the role that defines women's status and child has been born. value in society. Furthermore, preconception arrangeEven if fully informed choice were possible, soci- ments could create broad social harms by diminishing ety has established certain limits on what people are the dignity of reproduction and undermining society's free to make choices about. Such situations are rare but commitment to the inherent value of children. central to our definition of the kind of society we want Even if the number of commercial preconception to live in. Thus, in a caring society, personal autonomy arrangements to date has been relatively small, over is not a value that trumps all others, and society may time such arrangements would be bound to have a see fit to place limits on the exercise of free choice detrimental effect on the way society perceives women, when the choice concerns an activity that society re- children, and reproduction generally. gards as fundamentally incompatible with values such In short, we reject the notion that public policy as respect for human dignity and the inalienability of can be based on a description of procreation in terms of a market production model—which is, essentially, the person. We heard the view that preconception arrange- that such arrangements should be permissible and ments enhance women's autonomy by giving substance legally enforceable because commissioning couples to their right to control their own bodies—by allowing are willing to pay and gestational women are free to them to decide for themselves the meaning and social sell their labour. Second, we do not accept that the

Royal Commission on New Reproductive Technologies: Preconception Arrangements

freedom to procreate automatically assumes a right to state support—whether in the form of enforceable contracts or publicly supported medical services—for the exercise of that right. Finally, we reject the arguments of proponents because they are premised on incomplete or inaccurate depictions of preconception arrangements, making them an inappropriate basis for public policy, which must take into account not only the interests of the participants but also the other interests affected, including those of the resulting child, as well as the potential for individual and/or social harm. We also reject the argument of medical necessity. We find it unacceptable that one party—the gestational woman—should be called upon to bear all the medical risks of pregnancy and birth, and possibly those of in vitro fertilization and zygote transfer, while all benefits accrue to the commissioning couple. In no other circumstances does society accept that a healthy person be placed at medical risk for the benefit of someone else for money, even when that condition is life-threatening. In this case, the commissioning woman's infertility is not life-threatening. The commonly understood motivation for commissioning couples to enter into a preconception arrangement is because the commissioning woman is unable to conceive and/or carry a fetus to term. This makes preconception arrangements appear as a last resort when infertility is untreatable or treatment has been unsuccessful. As we have seen, however, this is not always the case; not everyone who seeks a preconception arrangement is involuntarily infertile, a member of a couple, or even of childbearing age. But even in cases where it is true, preconception arrangements are not an acceptable remedy. Goals of Public Policy

As we have argued elsewhere in this report, one goal of public policy in the field of new reproductive technologies should be to seek to prevent conflict—or, at the very least, to avoid knowingly setting up situations where conflict is bound to result. Instead, we would seek to foster healthy family and social relationships through such means, for example, as promoting greater social acceptance of family ties based on other than a genetic component. The Commission recognizes the value of public policy that supports people's attempts to establish families, and we uphold women's right to autonomy. We believe, nevertheless, that preconception arrangements can cause damage to children, families, and society as a whole and can actually limit, rather than enhance, women's autonomy. We recognize the

genuine and legitimate desire of couples who are infertile to have children, but this should not take precedence over the best interests of children, which lie in not being the object of a contract, agreement, or paid transaction. We believe further that preconception arrangements contradict the ethic of care, as they result— inevitably and intentionally—in the breaking of parental bonds and in strain on family relationships. They can also result in long, acrimonious conflict, in court and in the media, between the gestational woman and the commissioning couple; far from preventing conflict, preconception arrangements actually make conflict more likely. In reaching our conclusions, we also took into account the Commission's public hearings, consultations, and surveys, which illustrated the ambivalence of Canadians' attitudes toward commercial arrangements. Canadians have seen what has happened in the United States, where several cases have come to public attention through bitter custody disputes, and do not want to see these events repeated here. At the same time, as we saw in our survey of people across the country, Canadians attach great value to having children and appear reluctant to deny others the opportunity to have a child, even if it means permitting a practice they do not condone or would not engage in themselves. It is not clear, however, that public opinion about preconception arrangements is based on a full understanding of their nature and implications because much of the public information about them comes from brokers or others with an interest in portraying the practice in a positive light. In some circumstances a preconception arrangement may seem a reasonable response to a particular situation. For example, where a woman has a serious health problem that prevents her from carrying a pregnancy, she might seek a gestational woman to carry a fetus conceived using the commissioning woman's eggs and her partner's sperm. Given the broader social harms we have described, however, we do not believe that using another woman's reproductive capacity is justifiable even in this situation, as difficult as it might be for the individuals involved to accept their inability to have a genetically related child. We do not deny that a public response in the form of regulation could help to control some of the pressures and abuses identified with respect to preconception arrangements—for example, by requiring the provision of independent legal advice for gestational women or by making counselling mandatory for all parties to an arrangement. We are sceptical, however, that any regulatory scheme could ensure that all parties

uu

I

II ill

111111

5 Reproductive Ethics

were able to make free and informed choices. Moreover, a regulatory approach by its nature would invite disputes and conflict. Even if a regulatory system could be designed to overcome these obstacles, the deepest and most serious harms of preconception arrangements would remain. No regulatory system could remedy the basic affront to human dignity occasioned by the commodification of children and the commercialization of reproduction. Non-Commercial Arrangements The Commissionb conclusions with respect to non-

commercial preconception arrangements between family members or close friends are similar. We do not believe such arrangements should be undertaken, sanctioned, or encouraged. The motivation might be sincere and generous, but the arrangement still results in the commodification of a child and the reproductive process. Even if no money is involved, no one should have the right to make a "gift" of another human being; this is offensive to the human dignity of the child. Non-commercial arrangements present the potential for coercion in the form of family pressure to participate. They also give rise to the possibility of damage to family relationships before or after the child is born, as well as even greater potential for confusion on the part of the child, because of continuing contact between the birth mother and the commissioning couple. Moreover, the arrangement still results in a healthy woman being placed at medical risk for the benefit of someone else. At the same time, we recognize that the practice may continue to some degree and so demands a public policy response, particularly because of the uncertainties surrounding the legal status of a child born as a result of such an arrangement and the need to ensure that the child's best interests are served. We wish to make it clear, however, that our recommendations in this latter regard are not intended to sanction the practice but simply to recognize that it is probably going to occur and that, in the absence of public policy, significant harm to children could result. Recommendations

Commissioners are strongly of the view that preconception arrangements are unacceptable and do not warrant state support in any form that would signal acceptance or encouragement of them. We do not advise sanctions with respect to gestational mothers, however, as this would simply compound their vulnerability. While we

recognize the vulnerability of couples who are infertile and their emotional needs, we believe that making payment for such arrangements should be prohibited. We also believe it is essential, in particular, to prohibit others from assisting in such arrangements—for example, brokers and physicians—by criminalizing the knowing provision of assistance. With these principles in mind, the Commission reviewed the options available and came to the following recommendations to prohibit commercial preconception arrangements. Our first goal is to ensure that the status of this practice is uniform across the country, to discourage people from travelling to parts of the country where it is permitted. Evidence before the Commission shows that arrangements can take place across provincial/ territorial borders. Thus, prohibition only at the provincial level would not be effective with respect to such arrangements. Hence, we sought a comprehensive, uniform, and effective approach to preconception arrangements across the country. This can be achieved by prohibiting certain activities aimed at facilitating such arrangements for gain. Accordingly, the Commission recommends that 199. The federal government legislate to prohibit advertising for or acting as an intermediary to bring about a preconception arrangement; and to prohibit receiving payment or any financial or commercial benefit for acting as an intermediary, under threat of criminal sanction. It should also legislate to prohibit making payment for a preconception arrangement, under threat of criminal sanction. This proposed criminal prohibition will serve as an effective deterrent to commercial preconception arrangements. Given our recommendations with respect to donor insemination and in vitro fertilization (donor insemination restricted to anonymously donated sperm collected by a licensed facility, in vitro fertilization restricted to licensed facilities and offered only for diagnosed fallopian tube blockage), physicians too would be barred from participating in any such arrangement. Second, statutory provisions making preconception contracts unenforceable would operate as a strong deterrent to the practice, because they would generate uncertainty for the commissioning couple, whether or not a broker has been involved. Such provisions would ensure that the gestational woman could not be compelled by a court to give up custody of a child born as a result of a preconception agreement. The Commission therefore recommends that

Royal Commission on New Reproductive Technologies: Preconception Arrangements

200. Provinces/territories amend their family law legislation to specify that all preconception agreements, whether or not they involve payment, are unenforceable against the gestational woman. Commissioners do not wish to leave the impression that we consider non-commercial arrangements acceptable or to imply that non-commercial arrangements have no potential to harm individual women, the status of women generally, children and families, or society at large. However, we believe that the most effective way to deter non-commercial arrangements is to provide for penalties for third parties who facilitate preconception arrangements. The Commission recommends that 201. Self-regulating professional bodies, such as provincial colleges of physicians and surgeons and provincial law societies, adopt strict codes of conduct, disciplinary measures, and severe penalties, including loss of licence to practise, against members involved in brokering or performing assisted insemination, in vitro fertilization, or zygote/ embryo transfer to facilitate a preconception arrangement. and that 202. Any facility knowingly providing assisted conception procedures in support of a preconception arrangement lose its licence to provide assisted conception services. It is important to ensure that the interests of any resulting children are protected. In particular, establishing their legal parentage is vital for the children and for the other participants, as it affects the rights and obligations of the parties with respect to custody, access, support, and inheritance. Without clarification of legal parentage, children could be deprived of the support they are owed and become subject to the trauma of protracted litigation. For these reasons the Commission recommends that 203. All provinces/territories that have not already done so amend their family law legislation to ensure that a woman who gives birth to a child is considered the legal mother of the child, regardless of the source of the egg 204. As in the case of adoption, the birth mother be allowed to relinquish her maternal rights only

after a minimum waiting period following the birth of the child. and that 205. In any dispute over custody arising from a preconception arrangement, the best interests of the child prevail over the interests of the adults involved. Finally, in support of our international obligations, Commissioners believe that Canada should demonstrate leadership by supporting policies aimed at achieving an international ban on preconception arrangements. Given that Canadians could go to other countries, particularly the United States, to seek arrangements not permitted in this country, we believe that such a step is needed on the part of the international community to prevent the exploitation of women and the commodification of children. Adopting a domestic policy would be the first step toward this goal; encouraging other countries to adopt similar measures would reinforce and extend it. The extent to which Commissioners deplore the practice of preconception arrangements is evident in our determination to recommend strong measures to discourage these arrangements and to penalize those who would seek to benefit financially from them. Our goal is to halt commercial practices entirely and to discourage others from participating in these arrangements. We recognize the value Canadians attach to having a genetically related child. In our view, however, this value cannot be upheld in the face of the other values that would have to be sacrificed. A caring society has an obligation to ensure that individual actions—even those intended to benefit others—do not generate greater social harms and that public policy works to support and foster healthy family and social connections, not to undermine them or set them up to fail. Preconception arrangements illustrate the ethical dilemmas posed by situations where both benefits and harms can result from a practice. In this case, however, the benefits to a few individuals are far outweighed by the harms to others and to society that are likely to result. This is why we have adopted such a strong stand against preconception arrangements. Taken together, Commissioners believe, the measures we propose will have a strong deterrent effect on preconception arrangements and, in particular, on third-party activities in this area, but without compounding the vulnerability of participants. [. . .1

LI I

I I ill

111111

5 Reproductive Ethics

5.3 Stopping Reproduction

Why Abortion Is Immoral Don Marquis The view that abortion is, with rare exceptions, seriously immoral has received little support in the recent philosophical literature. No doubt most philosophers affiliated with secular institutions of higher education believe that the anti-abortion position is either a symptom of irrational religious dogma or a conclusion generated by seriously confused philosophical argument. The purpose of this essay is to undermine this general belief. This essay sets out an argument that purports to show, as well as any argument in ethics can show, that abortion is, except possibly in rare cases, seriously immoral, that it is in the same moral category as killing an innocent adult human being. The argument is based on a major assumption. Many of the most insightful and careful writers on the ethics of abortion—such as Joel Feinberg, Michael Tooley, Mary Anne Warren, H. Tristram Engelhardt, Jr, L.W. Sumner, John T. Noonan, Jr, and Philip Devine— believe that whether or not abortion is morally permissible stands or falls on whether or not a fetus is the sort of being whose life it is seriously wrong to end. The argument of this essay will assume, but not argue, that they are correct. Also, this essay will neglect issues of great importance to a complete ethics of abortion. Some antiabortionists will allow that certain abortions, such as abortion before implantation or abortion when the life of a woman is threatened by a pregnancy or abortion after rape, may be morally permissible. This essay will not explore the casuistry of these hard cases. The purpose of this essay is to develop a general argument for the claim that the overwhelming majority of deliberate abortions are seriously immoral.

A sketch of standard anti-abortion and pro-choice arguments exhibits how those arguments possess certain symmetries that explain why partisans of those positions are so convinced of the correctness of their own positions, why they are not successful in convincing their opponents, and why, to others, this issue

seems to be unresolvable. An analysis of the nature of this standoff suggests a strategy for surmounting it. Consider the way a typical anti-abortionist argues. She will argue or assert that life is present from the moment of conception or that fetuses look like babies or that fetuses possess a characteristic such as a genetic code that is both necessary and sufficient for being human. Anti-abortionists seem to believe that (1) the truth of all of these claims is quite obvious, and (2) establishing any of these claims is sufficient to show that abortion is morally akin to murder. A standard pro-choice strategy exhibits similarities. The pro-choicer will argue or assert that fetuses are not persons or that fetuses are not rational agents or that fetuses are not social beings. Pro-choicers seem to believe that (1) the truth of any of these claims is quite obvious, and (2) establishing any of these claims is sufficient to show that an abortion is not a wrongful killing. In fact, both the pro-choice and the anti-abortion claims do seem to be true, although the "it looks like a baby" claim is more difficult to establish the earlier the pregnancy. We seem to have a standoff. How can it be resolved? As everyone who has taken a bit of logic knows, if any of these arguments concerning abortion is a good argument, it requires not only some claim characterizing fetuses, but also some general moral principle that ties a characteristic of fetuses to having or not having the right to life or to some other moral characteristic that will generate the obligation or the lack of obligation not to end the life of a fetus. Accordingly, the arguments of the anti-abortionist and the prochoicer need a bit of filling in to be regarded as adequate. Note what each partisan will say. The antiabortionist will claim that her position is supported by such generally accepted moral principles as "It is always prima facie seriously wrong to take a human life," or "It is always prima facie seriously wrong to end the life of a baby." Since these are generally accepted moral principles, her position is certainly not obviously wrong. The prochoicer will claim that her position is supported by such plausible moral principles as "Being a person is what gives an individual intrinsic moral worth," or "It is only seriously prima facie wrong to take the life of a member of the human community." Since these are generally accepted moral principles, the

Source: From Journal of Philosophy 86 (1989) (4), 183-202. Reprinted by permission of the author and the Journal of Philosophy.

Marquis: Why Abortion Is Immoral

pro-choice position is certainly not obviously wrong. Unfortunately, we have again arrived at a standoff. Now, how might one deal with this standoff? The standard approach is to try to show how the moral principles of one's opponent lose their plausibility under analysis. It is easy to see how this is possible. On the one hand, the anti-abortionist will defend a moral principle concerning the wrongness of killing which tends to be broad in scope in order that even fetuses at an early stage of pregnancy will fall under it. The problem with broad principles is that they often embrace too much. In this particular instance, the principle "It is always prima facie wrong to take a human life" seems to entail that it is wrong to end the existence of a living human cancer-cell culture, on the grounds that the culture is both living and human. Therefore, it seems that the anti-abortionist's favoured principle is too broad. On the other hand, the prochoicer wants to find a moral principle concerning the wrongness of killing which tends to be narrow in scope in order that fetuses will not fall under it. The problem with narrow principles is that they often do not embrace enough. Hence, the needed principles such as "It is prima facie seriously wrong to kill only persons" or "It is prima facie wrong to kill only rational agents' do not explain why it is wrong to kill infants or young children or the severely retarded or even perhaps the severely mentally ill. Therefore, we seem again to have a standoff. The anti-abortionist charges, not unreasonably, that prochoice principles concerning killing are too narrow to be acceptable; the prochoicer charges, not unreasonably, that anti-abortionist principles concerning killing are too broad to be acceptable. Attempts by both sides to patch up the difficulties in their positions run into further difficulties. The antiabortionist will try to remove the problem in her position by reformulating her principle concerning killing in terms of human beings. Now we end up with: "It is always prima fade seriously wrong to end the life of a human being." This principle has the advantage of avoiding the problem of the human cancer-cell culture counterexample. But this advantage is purchased at a high price. For although it is clear that a fetus is both human and alive, it is not at all clear that a fetus is a human being. There is at least something to be said for the view that something becomes a human being only after a process of development, and that therefore first-trimester fetuses and perhaps all fetuses are not yet human beings. Hence, the anti-abortionist, by this move, has merely exchanged one problem for another.' The prochoicer fares no better. She may attempt to find reasons why killing infants, young children, and the severely retarded is wrong which are independent

of her major principle that is supposed to explain the wrongness of taking human life, but which will not also make abortion immoral. This is no easy task. Appeals to social utility will seem satisfactory only to those who resolve not to think of the enormous difficulties with a utilitarian account of the wrongness of killing and the significant social costs of preserving the lives of the unproductive.' A pro-choice strategy that extends the definition of "person" to infants or even to young children seems just as arbitrary as an anti-abortion strategy that extends the definition of "human being" to fetuses. Again, we find symmetries in the two positions and we arrive at a standoff. There are even further problems that reflect symmetries in the two positions. In addition to counterexample problems, or the arbitrary application problems that can be exchanged for them, the standard anti-abortionist principle "It is prima facie seriously wrong to kill a human being," or one of its variants, can be objected to on the grounds of ambiguity. If "human being" is taken to be a biological category, then the anti-abortionist is left with the problem of explaining why a merely biological category should make a moral difference. Why, it is asked, is it any more reasonable to base a moral conclusion on the number of chromosomes in one's cells than on the colour of one's skin?4 If "human being," on the other hand, is taken to be a moral category, then the claim that a fetus is a human being cannot be taken to be a premise in the anti-abortion argument, for it is precisely what needs to be established. Hence, either the anti-abortionist's main category is a morally irrelevant, merely biological category, or it is of no use to the anti-abortionist in establishing (non-circularly, of course) that abortion is wrong. Although this problem with the anti-abortionist position is often noticed, it is less often noticed that the pro-choice position suffers from an analogous problem. The principle "Only persons have the right to life" also suffers from an ambiguity. The term "person" is typically defined in terms of psychological characteristics, although there will certainly be disagreement concerning which characteristics are most important. Supposing that this matter can be settled, the prochoicer is left with the problem of explaining why psychological characteristics should make a moral difference. If the prochoicer should attempt to deal with this problem by claiming that an explanation is not necessary, that in fact we do treat such a cluster of psychological properties as having moral significance, the sharp-witted antiabortionist should have a ready response. We do treat being both living and human as having moral significance. If it is legitimate for the prochoicer to demand that the anti-abortionist provide an explanation of the

xi .11

I

.111

II

ill

1111.1.1

5 Reproductive Ethics

connection between the biological character of being a human being and the wrongness of being killed (even though people accept this connection), then it is legitimate for the anti-abortionist to demand that the prochoicer provide an explanation of the connection between psychological criteria for being a person and the wrongness of being killed (even though that connection is accepted).5 Feinberg has attempted to meet this objection (he calls psychological personhood "commonsense personhood"): The characteristics that confer commonsense personhood are not arbitrary bases for rights and duties, such as race, sex or species membership; rather they are traits that make sense out of rights and duties and without which those moral attributes would have no point or function. It is because people are conscious; have a sense of their personal identities; have plans, goals, and projects; experience emotions; are liable to pains, anxieties, and frustrations; can reason and bargain, and so on—it is because of these attributes that people have values and interests, desires and expectations of their own, including a stake in their own futures, and a personal well-being of a sort we cannot ascribe to unconscious or non-rational beings. Because of their developed capacities they can assume duties and responsibilities and can have and make claims on one another. Only because of their sense of self, their life plans, their value hierarchies, and their stakes in their own futures can they be ascribed fundamental rights. There is nothing arbitrary about these linkages. (op. cit., p. 270) The plausible aspects of this attempt should not be taken to obscure its implausible features. There is a great deal to be said for the view that being a psychological person under some description is a necessary condition for having duties. One cannot have a duty unless one is capable of behaving morally, and a being's capability of behaving morally will require having a certain psychology. It is far from obvious, however, that having rights entails consciousness or rationality, as Feinberg suggests. We speak of the rights of the severely retarded or the severely mentally ill, yet some of these persons are not rational. We speak of the rights of the temporarily unconscious. The New Jersey Supreme Court based their decision in the Quinlan case on Karen Ann Quinlan's right to privacy, and she was known to be permanently unconscious at that time. Hence, Feinberg's claim that having rights entails being conscious is, on its face, obviously false.

Of course, it might not make sense to attribute rights to a being that would never in its natural history have certain psychological traits. This modest connection between psychological personhood and moral personhood will create a place for Karen Ann Quinlan and the temporarily unconscious. But then it makes a place for fetuses also. Hence, it does not serve Feinberg's pro-choice purposes. Accordingly, it seems that the prochoicer will have as much difficulty bridging the gap between psychological personhood and personhood in the moral sense as the anti-abortionist has bridging the gap between being a biological human being and being a human being in the moral sense. Furthermore, the prochoicer cannot any more escape her problem by making "person" a purely moral category than the anti-abortionist could escape by the analogous move. For if "person" is a moral category, then the prochoicer is left without the resources for establishing (non-circularly, of course) the claim that a fetus is not a person, which is an essential premise in her argument. Again, we have both a symmetry and a standoff between pro-choice and anti-abortion views. Passions in the abortion debate run high. There are both plausibilities and difficulties with the standard positions. Accordingly, it is hardly surprising that partisans of either side embrace with fervor the moral generalizations that support the conclusions they preanalytically favour, and reject with disdain the moral generalizations of their opponents as being subject to inescapable difficulties. It is easy to believe that the counterexamples to one's own moral principles are merely temporary difficulties that will dissolve in the wake of further philosophical research, and that the counterexamples to the principles of one's opponents are as straightforward as the contradiction between A and 0 propositions in traditional logic. This might suggest to an impartial observer (if there are any) that the abortion issue is unresolvable. There is a way out of this apparent dialectical quandary. The moral generalizations of both sides are not quite correct. The generalizations hold for the most part, for the usual cases. This suggests that they are all accidental generalizations that the moral claims made by those on both sides of the dispute do not touch on the essence of the matter. This use of the distinction between essence and accident is not meant to invoke obscure metaphysical categories. Rather, it is intended to reflect the rather atheoretical nature of the abortion discussion. If the generalization a partisan in the abortion dispute adopts were derived from the reason why ending the life of a human being is wrong, then there could not be exceptions to that generalization unless some special

Marquis: Why Abortion Is Immoral

case obtains in which there are even more powerful countervailing reasons. Such generalizations would not be merely accidental generalizations; they would point to, or be based upon, the essence of the wrongness of killing, what it is that makes killing wrong. All this suggests that a necessary condition of resolving the abortion controversy is a more theoretical account of the wrongness of killing. After all, if we merely believe, but do not understand, why killing adult human beings such as ourselves is wrong, how could we conceivably show that abortion is either immoral or permissible? II In order t o develop such an account, w e can start from the following unproblematic assumption concerning our own case: it is wrong to kill us. Why is it wrong? Some answers can be easily eliminated. It might be said that what makes killing us wrong is that a killing brutalizes the one who kills. But the brutalization consists of being inured to the performance of an act that is hideously immoral; hence, the brutalization does not explain the immorality It might be said that what makes killing us wrong is the great loss others would experience due to our absence. Although such hubris is understandable, such an explanation does not account for the wrongness of killing hermits, or those whose lives are relatively independent and whose friends find it easy to make new friends. A more obvious answer is better. What primarily makes killing wrong is neither its effect on the murderer nor its effect on the victim's friends and relatives, but its effect on the victim. The loss of one's life is one of the greatest losses one can suffer. The loss of one's life deprives one of all the experiences, activities, projects, and enjoyments that would otherwise have constituted one future. Therefore, killing someone is wrong, primarily because the killing inflicts (one of) the greatest possible losses on the victim. To describe this as the loss of life can be misleading, however. The change in my biological state does not by itself make killing me wrong. The effect of the loss of my biological life is the loss to me of all those activities, projects, experiences, and enjoyments which would otherwise have constituted my future personal life. These activities, projects, experiences, and enjoyments are either valuable for their own sakes or are means to something else that is valuable for its own sake. Some parts of my future are not valued by me now, but will come to be valued by me as I grow older and as my values and capacities change. When I am killed, I am deprived both of what I now value which would have been part of my future personal life, but also what I would come to value.

Therefore, when I die, I am deprived of all of the value of my future. Inflicting this loss on me is ultimately what makes killing me wrong. This being the case, it would seem that what makes killing any adult human being prima facie seriously wrong is the loss of his or her future.° How should this rudimentary theory of the wrongness of killing be evaluated? It cannot be faulted for deriving an "ought" from an "is," for it does not. The analysis assumes that killing me (or you, reader) is prima fade seriously wrong. The point of the analysis is to establish which natural property ultimately explains the wrongness of the killing, given that it is wrong. A natural property will ultimately explain the wrongness of killing, only if (1) the explanation fits with our intuitions about the matter and (2) there is no other natural property that provides the basis for a better explanation of the wrongness of killing. This analysis rests on the intuition that what makes killing a particular human or animal wrong is what it does to that particular human or animal. What makes killing wrong is some natural effect or other of the killing. Some would deny this. For instance, a divine command theorist in ethics would deny it. Surely this denial is, however, one of those features of divinecommand theory which renders it so implausible. The claim that what makes killing wrong is the loss of the victim's future is directly supported by two considerations. In the first place, this theory explains why we regard killing as one of the worst of crimes. Killing is especially wrong, because it deprives the victim of more than perhaps any other crime. In the second place, people with AIDS or cancer who know they are dying believe, of course, that dying is a very bad thing for them. They believe that the loss of a future to them that they would otherwise have experienced is what makes their premature death a very bad thing for them. A better theory of the wrongness of killing would require a different natural property associated with killing which better fits with the attitudes of the dying. What could it be? The view that what makes killing wrong is the loss to the victim of the value of the victim's future gains additional support when some of its implications are examined. In the first place, it is incompatible with the view that it is wrong to kill only beings who are biologically human. It is possible that there exists a different species from another planet whose members have a future like ours. Since having a future like that is what makes killing someone wrong, this theory entails that it would be wrong to kill members of such a species. Hence, this theory is opposed to the claim that only life that is biologically human has great moral

3

II

11 31

11311

5 Reproductive Ethics

worth, a claim which many anti-abortionists have seemed to adopt. This opposition, which this theory has in common with personhood theories, seems to be a merit of the theory. In the second place, the claim that the loss of one future is the wrong-making feature of one's being killed entails the possibility that the futures of some actual non-human mammals on our own planet are sufficiently like ours that it is seriously wrong to kill them also. Whether some animals do have the same right to life as human beings depends on adding to the account of the wrongness of killing some additional account of just what it is about my future or the futures of other adult human beings which makes it wrong to kill us. No such additional account will be offered in this essay. Undoubtedly, the provision of such an account would be a very difficult matter. Undoubtedly, any such account would be quite controversial. Hence, it surely should not reflect badly on this sketch of an elementary theory of the wrongness of killing that it is indeterminate with respect to some very difficult issues regarding animal rights. In the third place, the claim that the loss of one's future is the wrong-making feature of one's being killed does not entail, as sanctity of human life theories do, that active euthanasia is wrong. Persons who are severely and incurably ill, who face a future of pain and despair, and who wish to die will not have suffered a loss if they are killed. It is, strictly speaking, the value of a humans future which makes killing wrong in this theory. This being so, killing does not necessarily wrong some persons who are sick and dying. Of course, there may be other reasons for a prohibition of active euthanasia, but that is another matter. Sanctity-of-human-life theories seem to hold that active euthanasia is seriously wrong even in an individual case where there seems to be good reason for it independently of public policy considerations. This consequence is most implausible, and it is a plus for the claim that the loss of a future of value is what makes killing wrong that it does not share this consequence. In the fourth place, the account of the wrongness of killing defended in this essay does straightforwardly entail that it is prima facie seriously wrong to kill children and infants, for we do presume that they have futures of value. Since we do believe that it is wrong to kill defenceless little babies, it is important that a theory of the wrongness of killing easily account for this. Personhood theories of the wrongness of killing, on the other hand, cannot straightforwardly account for the wrongness of killing infants and young children.' Hence, such theories must add special ad hoc accounts of the wrongness of killing the young. The plausibility

of such ad hoc theories seems to be a function of how desperately one wants such theories to work. The claim that the primary wrong-making feature of a killing is the loss to the victim of the value of its future accounts for the wrongness of killing young children and infants directly; it makes the wrongness of such acts as obvious as we actually think it is. This is a further merit of this theory. Accordingly, it seems that this value of a future-like-ours theory of the wrongness of killing shares strengths of both sanctity-of-life and personhood accounts while avoiding weaknesses of both. In addition, it meshes with a central intuition concerning what makes killing wrong. The claim that the primary wrong-making feature of a killing is the loss to the victim of the value of its future has obvious consequences for the ethics of abortion. The future of a standard fetus includes a set of experiences, projects, activities, and such which are identical with the futures of adult human beings and are identical with the futures of young children. Since the reason that is sufficient to explain why it is wrong to kill human beings after the time of birth is a reason that also applies to fetuses, it follows that abortion is prima facie seriously morally wrong. This argument does not rely on the invalid inference that, since it is wrong to kill persons, it is wrong to kill potential persons also. The category that is morally central to this analysis is the category of having a valuable future like ours; it is not the category of personhood. The argument to the conclusion that abortion is prima facie seriously morally wrong proceeded independently of the notion of person or potential person or any equivalent. Someone may wish to start with this analysis in terms of the value of a human future, conclude that abortion is, except perhaps in rare circumstances, seriously morally wrong, infer that fetuses have the right to life, and then call fetuses "persons" as a result of their having the right to life. Clearly in this case, the category of person is being used to state the conclusion of the analysis rather than to generate the argument of the analysis. The structure of this anti-abortion argument can be both illuminated and defended by comparing it to what appears to be the best argument for the wrongness of the wanton infliction of pain on animals. This latter argument is based on the assumption that it is prima facie wrong to inflict pain on me (or you, reader). What is the natural property associated with the infliction of pain which makes such infliction wrong? The obvious answer seems to be that the infliction of pain causes suffering and that suffering is a misfortune. The suffering caused by the infliction of pain is what makes the wanton infliction of pain on me

Marquis: Why Abortion Is Immoral

wrong. The wanton infliction of pain on other adult humans causes suffering. The wanton infliction of pain on animals causes suffering. Since causing suffering is what makes the wanton infliction of pain wrong and since the wanton infliction of pain on animals causes suffering, it follows that the wanton infliction of pain on animals is wrong. This argument for the wrongness of the wanton infliction of pain on animals shares a number of structural features with the argument for the serious prima facie wrongness of abortion. Both arguments start with an obvious assumption concerning what it is wrong to do to me (or you, reader). Both then look for the characteristic or the consequence of the wrong action which makes the action wrong. Both recognize that the wrong-making feature of these immoral actions is a property of actions sometimes directed at individuals other than postnatal human beings. If the structure of the argument for the wrongness of the wanton infliction of pain on animals is sound, then the structure of the argument for the prima facie serious wrongness of abortion is also sound, for the structure of the two arguments is the same. The structure common to both is the key to the explanation of how the wrongness of abortion can be demonstrated without recourse to the category of person. In neither argument is that category crucial. This defence of an argument for the wrongness of abortion in terms of a structurally similar argument for the wrongness of the wanton infliction of pain on animals succeeds only if the account regarding animals is the correct account. Is it? In the first place, it seems plausible. In the second place, its major competition is Kant's account. Kant believed that we do not have direct duties to animals at all, because they are not persons. Hence, Kant had to explain and justify the wrongness of inflicting pain on animals on the grounds that "he who is hard in his dealings with animals becomes hard also in his dealing with men."8 The problem with Kant's account is that there seems to be no reason for accepting this latter claim unless Kant's account is rejected. If the alternative to Kant's account is accepted, then it is easy to understand why someone who is indifferent to inflicting pain on animals is also indifferent to inflicting pain on humans, for one is indifferent to what makes inflicting pain wrong in both cases. But, if Kant's account is accepted, there is no intelligible reason why one who is hard in his dealings with animals (or crabgrass or stones) should also be hard in his dealings with men. After all, men are persons: animals are no more persons than crabgrass or stones. Persons are Kant's crucial moral category. Why, in short, should a Kantian accept the basic claim in Kant's argument?

Hence, Kant's argument for the wrongness of inflicting pain on animals rests on a claim that, in a world of Kantian moral agents, is demonstrably false. Therefore, the alternative analysis, being more plausible anyway, should be accepted. Since this alternative analysis has the same structure as the anti-abortion argument being defended here, we have further support for the argument for the immorality of abortion being defended in this essay. Of course, this value of a future-like-ours argument, if sound, shows only that abortion is prima facie wrong, not that it is wrong in any and all circumstances. Since the loss of the future to a standard fetus, if killed, is, however, at least as great a loss as the loss of the future to a standard adult human being who is killed, abortion, like ordinary killing, could be justified only by the most compelling reasons. The loss of one's life is almost the greatest misfortune that can happen to one. Presumably abortion could be justified in some circumstances, only if the loss consequent on failing to abort would be at least as great. Accordingly, morally permissible abortions will be rare indeed unless, perhaps, they occur so early in pregnancy that a fetus is not yet definitely an individual. Hence, this argument should be taken as showing that abortion is presumptively very seriously wrong, where the presumption is very strong as strong as the presumption that killing another adult human being is wrong. III

How complete an account o f the wrongness o f killing does the value of a future-like-ours account have to be in order that the wrongness of abortion is a consequence? This account does not have to be an account of the necessary conditions for the wrongness of killing. Some persons in nursing homes may lack valuable human futures, yet it may be wrong to kill them for other reasons. Furthermore, this account does not obviously have to be the sole reason killing is wrong where the victim did have a valuable future. This analysis claims only that, for any killing where the victim did have a valuable future like ours, having that future by itself is sufficient to create the strong presumption that the killing is seriously wrong. One way to overturn the value of a future-like-ours argument would be to find some account of the wrongness of killing which is at least as intelligible and which has different implications for the ethics of abortion. Two rival accounts possess at least some degree of plausibility. One account is based on the obvious fact that people value the experience of living and wish for that valuable experience to continue. Therefore, it might be

II ill

111111

244 5 Reproductive Ethics said, what makes killing wrong is the discontinuation of that experience for the victim. Let us call this the discontinuation account.9 Another rival account is based upon the obvious fact that people strongly desire to continue to live. This suggests that what makes killing us so wrong is that it interferes with the fulfillment of a strong and fundamental desire, the fulfillment of which is necessary for the fulfillment of any other desires we might have. Let us call this the desire account.'° Consider first the desire account as a rival account of the ethics of killing which would provide the basis for rejecting the anti-abortion position. Such an account will have to be stronger than the value of a future-like-ours account of the wrongness of abortion if it is to do the job expected of it. To entail the wrongness of abortion, the value of a future-like-ours account has only to provide a sufficient, but not a necessary, condition for the wrongness of killing. The desire account, on the other hand, must provide us also with a necessary condition for the wrongness of killing in order to generate a prochoice conclusion on abortion. The reason for this is that presumably the argument from the desire account moves from the claim that what makes killing wrong is interference with a very strong desire to the claim that abortion is not wrong because the fetus lacks a strong desire to live. Obviously, this inference fails if someone's having the desire to live is not a necessary condition of its being wrong to kill that individual. One problem with the desire account is that we do regard it as seriously wrong to kill persons who have little desire to live or who have no desire to live or, indeed, have a desire not to live. We believe it is seriously wrong to kill the unconscious, the sleeping, those who are tired of life, and those who are suicidal. The value-of-a-human-future account renders standard morality intelligible in these cases; these cases appear to be incompatible with the desire account. The desire account is subject to a deeper difficulty. We desire life, because we value the goods of this life. The goodness of life is not secondly to our desire for it. If this were not so, the pain of one's own premature death could be done away with merely by an appropriate alteration in the configuration of one's desires. This is absurd. Hence, it would seem that it is the loss of the goods of one's future, not the interference with the fulfillment of a strong desire to live, which accounts ultimately for the wrongness of killing. It is worth noting that, if the desire account is modified so that it does not provide a necessary, but only a sufficient, condition for the wrongness of killing, the desire account is compatible with the value of a future-like-ours account. The combined accounts will yield an anti-abortion ethic.

This suggests that one can retain what is intuitively plausible about the desire account without a challenge to the basic argument of this paper. It is also worth noting that, if future desires have moral force in a modified desire account of the wrongness of killing, one can find support for an anti-abortion ethic even in the absence of a value of a future-like-ours account. If one decides that a morally relevant property, the possession of which is sufficient to make it wrong to kill some individual, is the desire at some future time to live one might decide to justify one's refusal to kill suicidal teenagers on these grounds, for example then, since typical fetuses will have the desire in the future to live, it is wrong to kill typical fetuses. Accordingly, it does not seem that a desire account of the wrongness of killing can provide a justification of a pro-choice ethic of abortion which is nearly as adequate as the value of a human-future justification of an anti-abortion ethic. The discontinuation account looks more promising as an account of the wrongness of killing. It seems just as intelligible as the value of a future-like-ours account, but it does not justify an anti-abortion position. Obviously, if it is the continuation of one's activities, experiences, and projects, the loss of which makes killing wrong, then it is not wrong to kill fetuses for that reason, for fetuses do not have experiences, activities, and projects to be continued or discontinued. Accordingly, the discontinuation account does not have the anti-abortion consequences that the value of a futurelike-ours account has. Yet, it seems as intelligible as the value of a future-like-ours account, for when we think of what would be wrong with our being killed, it does seem as if it is the discontinuation of what makes our lives worthwhile which makes killing us wrong. Is the discontinuation account just as good an account as the value of a future-like-ours account? The discontinuation account will not be adequate at all, if it does not refer to the value of the experience that may be discontinued. One does not want the discontinuation account to make it wrong to kill a patient who begs for death and who is in severe pain that cannot be relieved short of killing. (I leave open the question of whether it is wrong for other reasons.) Accordingly, the discontinuation account must be more than a bare discontinuation account. It must make some reference to the positive value of the patient's experiences. But, by the same token, the value of a future-like-ours account cannot be a bare future account either. Just having a future surely does not itself rule out killing the above patient. This account must make some reference to the value of the patient's future experiences and projects also. Hence, both accounts involve the value of

Marquis: Why Abortion Is Immoral 245

experiences, projects, and activities. So far we still have symmetry between the accounts. The symmetry fades, however, when we focus on the time period of the value of the experiences, etc., which has moral consequences. Although both accounts leave open the possibility that the patient in our example may be killed, this possibility is left open only in virtue of the utterly bleak future for the patient. It makes no difference whether the patient's immediate past contains intolerable pain, or consists in being in a coma (which we can imagine is a situation of indifference), or consists in a life of value. If the patient's future is a future of value, we want our account to make it wrong to kill the patient. If the patient's future is intolerable, whatever his or her immediate past, we want our account to allow killing the patient. Obviously, then, it is the value of that patient's future which is doing the work in rendering the morality of killing the patient intelligible. This being the case, it seems clear that whether one has immediate past experiences or not does no work in the explanation of what makes killing wrong. The addition the discontinuation account makes to the value of a human future account is otiose. Its addition to the value-of-a-future account plays no role at all in rendering intelligible the wrongness of killing. Therefore, it can be discarded with the discontinuation account of which it is a part. IV The analysis of the previous section suggests that alternative general accounts of the wrongness of killing are either inadequate or unsuccessful in getting around the anti-abortion consequences of the value of a futurelike-ours argument. A different strategy for avoiding these anti-abortion consequences involves limiting the scope of the value of a future argument. More precisely, the strategy involves arguing that fetuses lack a property that is essential for the value-of-a-future argument (or for any anti-abortion argument) to apply to them. One move of this sort is based upon the claim that a necessary condition of one's future being valuable is that one values it. Value implies a valuer. Given this one might argue that, since fetuses cannot value their futures, their futures are not valuable to them. Hence, it does not seriously wrong them deliberately to end their lives. This move fails, however, because of some ambiguities. Let us assume that something cannot be of value unless it is valued by someone. This does not entail that my life is of no value unless it is valued by me. I may think, in a period of despair, that my future is of no worth

whatsoever, but I may be wrong because others rightly see value even great value in it. Furthermore, my future can be valuable to me even if I do not value it. This is the case when a young person attempts suicide, but is rescued and goes on to significant human achievements. Such young people's futures are ultimately valuable to them, even though such futures do not seem to be valuable to them at the moment of attempted suicide. A fetus's future can be valuable to it in the same way. Accordingly, this attempt to limit the anti-abortion argument fails. Another similar attempt to reject the anti-abortion position is based on Tooley's claim that an entity cannot possess the right to life unless it has the capacity to desire its continued existence. It follows that, since fetuses lack the conceptual capacity to desire to continue to live, they lack the right to life. Accordingly, Tooley concludes that abortion cannot be seriously prima facie wrong (op. cit., pp. 46-7). What could b e the evidence for Tooley's basic claim? Tooley once argued that individuals have a prima facie right to what they desire and that the lack of the capacity to desire something undercuts the basis of one's right to it (op. cit., pp. 44-5). This argument plainly will not succeed in the context of the analysis of this essay, however, since the point here is to establish the fetus's right to life on other grounds. Tooley's argument assumes that the right to life cannot be established in general on some basis other than the desire for life. This position was considered and rejected in the preceding section of this paper. One might attempt to defend Tooley's basic claim on the grounds that, because a fetus cannot apprehend continued life as a benefit, its continued life cannot be a benefit or cannot be something it has a right to or cannot be something that is in its interest. This might be defended in terms of the general proposition that, if an individual is literally incapable of caring about or taking an interest in some x, then one does not have a right to x or x is not a benefit or xis not something that is in one's interest.° Each member o f this family o f claims seems to be open to objections. As John C. Stevens12 has pointed out, one may have a right to be treated with a certain medical procedure (because of a health insurance policy one has purchased), even though one cannot conceive of the nature of the procedure. And, as Tooley himself has pointed out, persons who have been indoctrinated, or drugged, or rendered temporarily unconscious may be literally incapable of caring about or taking an interest in something that is in their interest or is something to which they have a right, or is something that benefits them. Hence, the Tooley claim that would restrict the scope of the value of a future-like-ours argument is undermined by counterexamples.°

I I ill

246

111111

5 Reproductive Ethics

Finally, Paul Bassenm has argued that, even though the prospects of an embryo might seem to be a basis for the wrongness of abortion, an embryo cannot be a victim and therefore cannot be wronged. An embryo cannot be a victim, he says, because it lacks sentience. His central argument for this seems to be that, even though plants and the permanently unconscious are alive, they clearly cannot be victims. What is the explanation of this? Bassen claims that the explanation is that their lives consist of mere metabolism and mere metabolism is not enough to ground victimizability. Mentation is required. The problem with this attempt to establish the absence of victimizability is that both plants and the permanently unconscious clearly lack what Bassen calls "prospects" or what I have called "a future life like ours." Hence, it is surely open to one to argue that the real reason we believe plants and the permanently unconscious cannot be victims is that killing them cannot deprive them of a future life like ours; the real reason is not their absence of present mentation. Bassen recognizes that his view is subject to this difficulty, and he recognizes that the case of children seems to support this difficulty, for "much of what we do for children is based on prospects." He argues, however, that, in the case of children and in other such cases, "potentiality comes into play only where victimizability has been secured on other grounds" (ibid., p. 333). Bassen's defence of his view is patently questionbegging, since what is adequate to secure victimizability is exactly what is at issue. His examples do not support his own view against the thesis of this essay. Of course, embryos can be victims: when their lives are deliberately terminated, they are deprived of their futures of value, their prospects. This makes them victims, for it directly wrongs them. The seeming plausibility of Bassen's view stems from the fact that paradigmatic cases of imagining someone as a victim involve empathy, and empathy requires mentation of the victim. The victims of flood, famine, rape, or child abuse are all persons with whom we can empathize. That empathy seems to be part of seeing them as victims.15 In spite of the strength of these examples, the attractive intuition that a situation in which there is victimization requires the possibility of empathy is subject to counterexamples. Consider a case that Bassen himself offers: "Posthumous obliteration of an author's work constitutes a misfortune for him only if he had wished his work to endure" (op cit., p. 318). The conditions Bassen wishes to impose upon the possibility of being victimized here seem far too strong. Perhaps this author, due to his unrealistic standards of excellence

and his low self-esteem, regarded his work as unworthy of survival, even though it possessed genuine literary merit. Destruction of such work would surely victimize its author. In such a case, empathy with the victim concerning the loss is clearly impossible. Of course, Bassen does not make the possibility of empathy a necessary condition of victimizability; he requires only mentation. Hence, on Bassen's actual view, this author, as I have described him, can be a victim. The problem is that the basic intuition that renders Bassen's view plausible is missing in the author'S case. In order to attempt to avoid counterexamples, Bassen has made his thesis too weak to be supported by the intuitions that suggested it. Even so, the mentation requirement on victimizability is still subject to counterexamples. Suppose a severe accident renders me totally unconscious for a month, after which I recover. Surely killing me while I am unconscious victimizes me, even though I am incapable of mentation during that time. It follows that Bassen's thesis fails. Apparently, attempts to restrict the value of a future-like-ours argument so that fetuses do not fall within its scope do not succeed. V In this essay, it has been argued that the correct ethic of the wrongness of killing can be extended to fetal life and used to show that there is a strong presumption that any abortion is morally impermissible. If the ethic of killing adopted here entails, however, that contraception is also seriously immoral, then there would appear to be a difficulty with the analysis of this essay. But this analysis does not entail that contraception is wrong. Of course, contraception prevents the actualization of a possible future of value. Hence, it follows from the claim that futures of value should be maximized that contraception is prima facie immoral. This obligation to maximize does not exist, however, furthermore, nothing in the ethics of killing in this paper entails that it does. The ethics of killing in this essay would entail that contraception is wrong only if something were denied a human future of value by contraception. Nothing at all is denied such a future by contraception, however. Candidates for a subject of harm by contraception fall into four categories: (1) some sperm or other, (2) some ovum or other, (3) a sperm and an ovum separately, and (4) a sperm and an ovum together. Assigning the harm to some sperm is utterly arbitrary, for no reason can be given for making a sperm the subject of harm rather than an ovum. Assigning the harm to some ovum is utterly arbitrary, for no reason can be given for making an ovum the subject of harm rather than a sperm. One might attempt

Marquis: Why Abortion Is Immoral

to avoid these problems by insisting that contraception deprives both the sperm and the ovum separately of a valuable future like ours. On this alternative, too many futures are lost. Contraception was supposed to be wrong, because it deprived us of one future of value, not two. One might attempt to avoid this problem by holding that contraception deprives the combination of sperm and ovum of a valuable future like ours. But here the definite article misleads. At the time of contraception, there are hundreds of millions of sperm, one (released) ovum and millions of possible combinations of all of these. There is no actual combination at all. Is the subject of the loss to be a merely possible combination? Which one? This alternative does not yield an actual subject of harm either. Accordingly, the immorality of contraception is not entailed by the loss of a future-like-ours argument simply because there is no non-arbitrarily identifiable subject of the loss in the case of contraception. VI The purpose of this essay has been to set out an argument for the serious presumptive wrongness of abortion subject to the assumption that the moral permissibility of abortion stands or falls on the moral status of the

fetus. Since a fetus possesses a property, the possession of which in adult human beings is sufficient to make killing an adult human being wrong, abortion is wrong. This way of dealing with the problem of abortion seems superior to other approaches to the ethics of abortion, because it rests on an ethics of killing which is close to self-evident, because the crucial morally relevant property clearly applies to fetuses, and because the argument' avoids the usual equivocations on "human life," "human being," or "person." The argument rests neither on religious claims nor on Papal dogma. It is not subject to the objection of "speciesism." Its soundness is compatible with the moral permissibility of euthanasia and contraception. It deals with our intuitions concerning young children. Finally, this analysis can be viewed as resolving a standard problem indeed, the standard problem concerning the ethics of abortion. Clearly, it is wrong to kill adult human beings. Clearly, it is not wrong to end the life of some arbitrarily chosen single human cell. Fetuses seem to be like arbitrarily chosen human cells in some respects and like adult humans in other respects. The problem of the ethics of abortion is the problem of determining the fetal property that settles this moral controversy. The thesis of this essay is that the problem of the ethics of abortion, so understood, is solvable.

Notes 1.

2.

3.

4.

Feinberg, (1986), "Abortion," i n Matters of Life and Death: New Introductory Essays in Moral Philosophy, ed. Tom Regan (Random House: New York), 256-93; Tooley, (1972), "Abortion and Infanticide," in Philosophy and Public Affairs 11(1): 37-65; Tooley, (1984), Abortion and Infanticide (Oxford: New York); Warren, (1973), "On the Moral and Legal Status of Abortion," in The Monist 1 (11): 43-61; Engelhardt, (1974), "The Ontology of Abortion," in Ethics, 34 (3): 217-34; Sumner, (1981), Abortion and Moral Theory (Princeton University Press: Princeton); Noonan, (1970), "An Almost Absolute Value in History," in The Morality of Abortion: Legal and Historical Perspectives, Noonan, ed. (Harvard University Press: Cambridge); and Devine, (1978), The Ethics of Homicide (Cornell: Ithaca, NY). For interesting discussions of this issue, see Warren Quinn, (1984), "Abortion: Identity and Loss," in Philosophy and Public Affairs 13 (1): 24-54; and Lawrence C Becker, (1975), "Human Being: The Boundaries of the Concept," in Philosophy and Public Affairs 4 (4): 334-59. For example, see my "Ethics and The Elderly: Some Problems," (1978), in Aging and the Elderly: Humanistic Perspectives in Gerontology, ed. Stuart Spicker, Kathleen Woodward, and David Van Tassel (Humanities: Atlantic Highlands, NJ), 341-55. See Warren, op. cit., and Tooley, "Abortion and Infanticide."

5. This seems to be the fatal flaw in Warren's treatment of this issue. 6. I have been most influenced on this matter by Jonathan Glover, (1977), Causing Death and Saving Lives (Penguin: New York), ch. 3; and Robert Young, (1979), "What Is So Wrong with Killing People?," in Philosophy, 225 (210): 515-28. 7. Feinberg, Tooley, Warren, and Engelhardt have all dealt with this problem. 8. "Duties to Animals and Spirits," (1963), in Lectures on Ethics, trans. Louis Infeld (Harper: New York), 239. 9. I am indebted to Jack Bricke for raising this objection. 10. Presumably a preference utilitarian would press such an objection. Tooley once suggested that his account has such a theoretical underpinning. See his "Abortion and Infanticide," 44-5. 11. Donald VanDeVeer seems to think this is self-evident. See his "Whither Baby Doe?" in Matters of Life and Death, 233. 12. "Must the Bearer of a Right Have the Concept of That to Which He Has a Right?," (1984), in Ethics 95 (1): 68-74. 13. See Tooley again in "Abortion and Infanticide," 47-9. 14. "Present Sakes and Future Prospects: The Status of Early Abortion," (1982), in Philosophy and Public Affairs 11 (4): 322-6. 15. Note carefully the reasons he gives on the bottom of p. 316.

111

II ill

111111

5 Reproductive Ethics

On the Moral and Legal Status of Abortion Mary Anne Warren

The question which we must answer in order to produce a satisfactory solution to the problem of the moral status of abortion is this: How are we to define the moral community, the set of beings with full and equal moral rights, such that we can decide whether a human fetus is a member of this community or not? What sort of entity, exactly, has the inalienable rights to life, liberty, and the pursuit of happiness? Jefferson attributed these rights to all men, and it may or may not be fair to suggest that he intended to attribute them only to men. Perhaps he ought to have attributed them to all human beings. If so, then we arrive, first, at [John] Noonan's problem of defining what makes a being human, and, second, at the equally vital question which Noonan does not consider, namely, what reason is there for identifying the moral community with the set of all human beings, in whatever way we have chosen to define that term?

1. On the Definition of"Human" One reason why this vital second question is so frequently overlooked in the debate over the moral status of abortion is that the term "human" has two distinct, but not often distinguished, senses. This fact results in a slide of meaning, which serves to conceal the fallaciousness of the traditional argument that since (1) it is wrong to kill innocent human beings, and (2) fetuses are innocent human beings, then (3) it is wrong to kill fetuses. For if "human" is used in the same sense in both (1) and (2) then, whichever of the two senses is meant, one of these premises is question-begging. And if it is used in two different senses then of course the conclusion doesn't follow Thus, (1) is a self-evident moral truth,' and avoids begging the question about abortion, only if "human being" is used to mean something like "a full-fledged member of the moral community" (It may or may not also be meant to refer exclusively to members of the species Homo sapiens.) We may call this the moral sense of "human." It is not to be conf used with what we call the genetic sense, i.e., the sense in which any member of the species is a human being, and no member of any

other species could be. If (1) is acceptable only if the moral sense is intended, (2) is non-question-begging only if what is intended is the genetic sense. In "Deciding Who is Human," Noonan argues for the classification of fetuses with human beings by pointing to the presence of the full genetic code, and the potential capacity for rational thought.2 It is clear that what he needs to show, for his version of the traditional argument to be valid, is that fetuses are human in the moral sense, the sense in which it is analytically true that all human beings have full moral rights. But, in the absence of any argument showing that whatever is genetically human is also morally human, and he gives none, nothing more than genetic humanity can be demonstrated by the presence of the human genetic code. And, as we will see, the potential capacity for rational thought can at most show that an entity has the potential for becoming human in the moral sense.

2. Defining the Moral Community Can it be established that genetic humanity is sufficient for moral humanity? I think that there are very good reasons for not defining the moral community in this way. I would like to suggest an alternative way of defining the moral community, which I will argue for only to the extent of explaining why it is, or should be, self-evident. The suggestion is simply that the moral community consists of all and only people, rather than all and only human beings,3 and probably the best way of demonstrating its self-evidence is by considering the concept of personhood, to see what sorts of entity are and are not persons, and what the decision that a being is or is not a person implies about its moral rights. What characteristics entitle an entity to be considered a person? This is obviously not the place to attempt a complete analysis of the concept of personhood, but we do not need such a fully adequate analysis just to determine whether and why a fetus is or isn't a person. All we need is a rough and approximate list of the most basic criteria of personhood, and some idea of which, or how many, of these an entity must satisfy in order to properly be considered a person. In searching for such criteria, it is useful to look beyond the set of people with whom we are acquainted, and ask how we would decide whether a totally alien being was a person or not. (For we have no right to assume that genetic humanity is necessary for personhood.) Imagine a space traveller who lands on an

Source From The Monist: An International Quarterly Journal of General Philosophical Inquiry 57 (1973), 43-61. Copyright © 1973 The MONIST: An International Quarterly Journal of General Philosophical Inquiry Peru, Illinois, USA 61354. Reprinted with permission.

Warren: On the Moral and Legal Status of Abortion

unknown planet and encounters a race of beings utterly unlike any he has ever seen or heard of. If he wants to be sure of behaving morally toward these beings, he has to somehow decide whether they are people, and hence have full moral rights, or whether they are the sort of thing which he need not feel guilty about treating as, for example, a source of food. How should he go about making this decision? If he has some anthropological background, he might look for such things as religion, art, and the manufacturing of tools, weapons, or shelters, since these factors have been used to distinguish our human from our prehuman ancestors, in what seems to be closer to the moral than the genetic sense of "human." And no doubt he would be right to consider the presence of such factors as good evidence that the alien beings were people, and morally human. It would, however, be overly anthropocentric of him to take the absence of these things as adequate evidence that they were not, since we can imagine people who have progressed beyond, or evolved without ever developing, these cultural characteristics. I suggest that the traits which are most central to the concept of personhood, or humanity in the moral sense, are, very roughly the following: 1. consciousness (of objects and events external and/ or internal to the being), and in particular the capacity to feel pain; 2. reasoning (the developed capacity to solve new and relatively complex problems); 3. self-motivated activity (activity which is relatively independent of either genetic or direct external control); 4. the capacity to communicate, by whatever means, messages of an indefinite variety of types, that is, not just with an indefinite number of possible contents, but on indefinitely many possible topics; 5. the presence of self-concepts, and self-awareness, either individual or racial, or both. Admittedly, there are apt to be a great many problems involved in formulating precise definitions of these criteria, let alone in developing universally valid behavioural criteria for deciding when they apply. But I will assume that both we and our explorer know approximately what (1)—(5) mean, and that he is also able to determine whether or not they apply. How, then, should he use his findings to decide whether or not the alien beings are people? We needn't suppose that an entity must have all of these attributes to be properly considered a person; (1) and (2) alone may well be sufficient for personhood, and quite probably (1)—(3) are

sufficient. Neither do we need to insist that any one of these criteria is necessary for personhood, although once again (1) and (2) look like fairly good candidates for necessary conditions, as does (3), if "activity" is construed so as to include the activity of reasoning. All we need to claim, to demonstrate that a fetus is not a person, is that any being which satisfies none of (1)—(5) is certainly not a person. I consider this claim to be so obvious that I think anyone who denied it, and claimed that a being which satisfied none of (1)—(5) was a person all the same, would thereby demonstrate that he had no notion at all of what a person is—perhaps because he had confused the concept of a person with that of genetic humanity. If the opponents of abortion were to deny the appropriateness of these five criteria, I do not know what further arguments would convince them. We would probably have to admit that our conceptual schemes were indeed irreconcilably different, and that our dispute could not be settled objectively. I do not expect this to happen, however, since I think that the concept of a person is one which is very nearly universal (to people), and that it is common to both pro-abortionists and anti-abortionists, even though neither group has fully realized the relevance of this concept to the resolution of their dispute. Furthermore, I think that on reflection even the antiabortionists ought to agree not only that (1)—(5) are central to the concept of personhood, but also that it is a part of this concept that all and only people have full moral rights. The concept of a person is in part a moral concept; once we have admitted that x is a person we have recognized, even if we have not agreed to respect, x's right to be treated as a member of the moral community. It is true that the claim that x is a human being is more commonly voiced as part of an appeal to treat x decently than is the claim that x is a person, but this is either because "human being" is here used in the sense which implies personhood, or because the genetic and moral sense of "human" have been confused. Now if (1)—(5) are indeed the primary criteria of personhood, then it is clear that genetic humanity is neither necessary nor sufficient for establishing that an entity is a person. Some human beings are not people, and there may well be people who are not human beings. A man or woman whose consciousness has been permanently obliterated but who remains alive is a human being which is no longer a person; defective human beings, with no appreciable mental capacity, are not and presumably never will be people; and a fetus is a human being which is not yet a person, and which therefore cannot coherently be said to have full moral rights. Citizens of the next century should be

II

2

111111

250 5 Reproductive Ethics

prepared to recognize highly advanced, self-aware robots or computers, should such be developed, and intelligent inhabitants of other worlds, should such be found, as people in the fullest sense, and to respect their moral rights. But to ascribe full moral rights to an entity which is not a person is as absurd as to ascribe moral obligations and responsibilities to such an entity. 3. Fetal Development and the Right to Life

Two problems arise in the application of these suggestions for the definition of the moral community to the determination of the precise moral status of a human fetus. Given that the paradigm example of a person is a normal adult human being, then (1) how like this paradigm, in particular how far advanced since conception, does a human being need to be before it begins to have a right to life by virtue, not of being fully a person as of yet, but of being like a person?, and (2) to what extent, if any, does the fact that a fetus has the potential for becoming a person endow it with some of the same rights? Each of these questions requires some comment. In answering the first question, we need not attempt a detailed consideration of the moral rights of organisms which are not developed enough, aware enough, intelligent enough, etc., to be considered people, but which resemble people in some respects. It does seem reasonable to suggest that the more like a person, in the relevant respects, a being is, the stronger is the case for regarding it as having a right to life, and indeed the stronger its right to life is. Thus we ought to take seriously the suggestion that, insofar as "the human individual develops biologically in a continuous fashion . . . the rights of a human person might develop in the same way."4 But we must keep in mind that the attributes which are relevant in determining whether or not an entity is enough like a person to be regarded as having some of the same moral rights are no different from those which are relevant to determining whether or not it is fully a person—i.e., are no different from (1)—(5)—and that being genetically human, or having recognizable human facial and other physical features, or detectable brain activity, or the capacity to survive outside the uterus, are simply not among these relevant attributes. Thus it is clear that even though a seven- or eightmonth fetus has features which make it apt to arouse in us almost the same powerful protective instinct as is commonly aroused by a small infant, nevertheless it is not significantly more personlike than is a very small embryo. It is somewhat more personlike; it can apparently feel and respond to pain, and it may even have a rudimentary form of consciousness, insofar as its brain is quite active. Nevertheless, it seems safe to say that it

is not fully conscious, in the way that an infant of a few months is, and that it cannot reason, or communicate messages of indefinitely many sorts, does not engage in self-motivated activity and has no self-awareness. Thus, in the relevant respects, a fetus, even a fully developed one, is considerably less personlike than is the average mature mammal, indeed the average fish. And I think that a rational person must conclude that if the right to life of a fetus is to be based upon its resemblance to a person, then it cannot be said to have any more right to life than, let us say, a newborn guppy (which also seems to be capable of feeling pain), and that a right of that magnitude could never override a woman's right to obtain an abortion, at any stage of her pregnancy. There may, of course, be other arguments in favour of placing legal limits upon the stage of pregnancy in which an abortion may be performed. Given the relative safety of the new techniques of artificially inducing labour during the third trimester, the danger to the woman's life or health is no longer such an argument. Neither is the fact that people tend to respond to the thought of abortion in the later stages of pregnancy with emotional repulsion, since mere emotional responses cannot take the place of moral reasoning in determining what ought to be permitted. Nor, finally is the frequently heard argument that legalizing abortion, especially late in the pregnancy, may erode the level of respect for human life, leading, perhaps, to an increase in unjustified euthanasia and other crimes. For this threat, if it is a threat, can be better met by educating people to the kinds of moral distinctions which we are making here than by limiting access to abortion (which limitation may in its disregard for the rights of women, be just as damaging to the level of respect for human rights). Thus, since the fact that even a fully developed fetus is not personlike enough to have any significant right to life on the basis of its personlikeness shows that no legal restrictions upon the stage of pregnancy in which an abortion may be performed can be justified on the grounds that we should protect the rights of the older fetus; and since there is no other apparent justification for such restrictions, we may conclude that they are entirely unjustified. Whether or not it would be indecent (whatever that means) for a woman in her seventh month to obtain an abortion just to avoid having to postpone a trip to Europe, it would not, in itself, be immoral, and therefore it ought to be permitted. 4. Potential Personhood and the Right to Life

We have seen that a fetus does not resemble a person in any way which can support the claim that it has even some of the same rights. But what about its potential,

Warren: On the Moral and Legal Status of Abortion

the fact that if nurtured and allowed to develop naturally it will very probably become a person? Doesn't that alone give it at least some right to life? It is hard to deny that the fact that an entity is a potential person is a strong prima facie reason for not destroying it; but we need not conclude from this that a potential person has a right to life, by virtue of that potential. It may be that our feeling that it is better, other things being equal, not to destroy a potential person is better explained by the fact that potential people are still (felt to be) an invaluable resource, not to be lightly squandered. Surely, if every speck of dust were a potential person, we would be much less apt to conclude that every potential person has a right to become actual. Still, we do not need to insist that a potential person has no right to life whatever. There may well be something immoral, and not just imprudent, about wantonly destroying potential people, when doing so isn't necessary to protect anyone's rights. But even if a potential person does have some prima facie right to life, such a right could not possibly outweigh the right of a woman to obtain an abortion, since the rights of any actual person invariably outweigh those of any potential person, whenever the two conflict. Since this may not be immediately obvious in the case of a human fetus, let us look at another case. Suppose that our space explorer falls into the hands of an alien culture, whose scientists decide to create a few hundred thousand or more human beings, by breaking his body into its component cells, and using these to create fully developed human beings, with, of course, his genetic code. We may imagine that each of these newly created men will have all of the original man's abilities, skills, knowledge, and so on, and also have an individual self-concept—in short, that each of them will be a bona fide (though hardly unique) person. Imagine that the whole project will take only seconds, and that its chances of success are extremely high, and that our explorer knows all of this, and also knows that these people will be treated fairly. I maintain that in such a situation he would have every right to escape if he could, and thus to deprive all of these potential people of their potential lives; for his right to life outweighs all of theirs together, in spite of the fact that they are all genetically human, all innocent, and all have a very high probability of becoming people very soon, if only he refrains from acting. Indeed, I think he would have a right to escape even if it were not his life which the alien scientists planned to take, but only a year of his freedom, or, indeed, only a day. Nor would he be obligated to stay if he had gotten captured (thus bringing all these people-potentials into existence) because of his own

carelessness, or even if he had done so deliberately, knowing the consequences. Regardless of how he got captured, he is not morally obligated to remain in captivity for any period of time for the sake of permitting any number of potential people to come into actuality, so great is the margin by which one actual person's right to liberty outweighs whatever right to life even a hundred thousand potential people have. And it seems reasonable to conclude that the rights of a woman will outweigh by a similar margin whatever right to life a fetus may have by virtue of its potential personhood. Thus, neither a fetus resemblance to a person, nor its potential for becoming a person provides any basis whatever for the claim that it has any significant right to life. Consequently, a woman right to protect her health, happiness, freedom, and even her life,' by terminating an unwanted pregnancy, will always override whatever right to life it may be appropriate to ascribe to a fetus, even a fully developed one. And thus, in the absence of any overwhelming social need for every possible child, the laws which restrict the right to obtain an abortion, or limit the period of pregnancy during which an abortion may be performed, are a wholly unjustified violation of a woman's most basic moral and constitutional rights.6

Postscript on Infanticide, 26 February 1982 One of the most troubling objections to the argument presented in this article is that it may appear to justify not only abortion but infanticide as well. A newborn infant is not a great deal more personlike than a nine-month fetus, and thus it might seem that if late-term abortion is sometimes justified, then infanticide must also be sometimes justified. Yet most people consider that infanticide is a form of murder, and thus never justified. While it is important to appreciate the emotional force of this objection, its logical force is far less than it may seem at first glance. There are many reasons why infanticide is much more difficult to justify than abortion, even though if my argument is correct neither constitutes the killing of a person. In this country, and in this period of history, the deliberate killing of viable newborns is virtually never justified. This is in part because neonates are so very close to being persons that to kill them requires a very strong moral justification—as does the killing of dolphins, whales, chimpanzees, and other highly personlike creatures. It is certainly wrong to kill such beings just for the sake of convenience, or financial profit, or "sport." Another reason why infanticide is usually wrong, in our society, is that if the newborn's parents do not want it, or are unable to care for it, there are (in most

.11

II

.111

111111

5 Reproductive Ethics

cases) people who are able and eager to adopt it and to provide a good home for it. Many people wait years for the opportunity to adopt a child, and some are unable to do so even though there is every reason to believe that they would be good parents. The needless destruction of a viable infant inevitably deprives some person or persons of a source of great pleasure and satisfaction, perhaps severely impoverishing their lives. Furthermore, even if an infant is considered to be unadoptable (e.g., because of some extremely severe mental or physical handicap) it is still wrong in most cases to kill it. For most of us value the lives of infants, and would prefer to pay taxes to support orphanages and state institutions for the handicapped rather than to allow unwanted infants to be killed. So long as most people feel this way, and so long as our society can afford to provide care for infants which are unwanted or which have special needs that preclude home care, it is wrong to destroy any infant which has a chance of living a reasonably satisfactory life. If these arguments show that infanticide is wrong, at least in this society, then why don't they also show that late-term abortion is wrong? After all, third-trimester fetuses are also highly personlike, and many people value them and would much prefer that they be preserved, even at some cost to themselves. As a potential source of pleasure to some family, a viable fetus is just as valuable as a viable infant. But there is an obvious and crucial difference between the two cases once the infant is born, its continued life cannot (except, perhaps, in very exceptional cases) pose any serious threat to the woman's life or health, since she is free to put it up for adoption, or, where this is impossible, to place it in a state-supported institution. While she might prefer that it die, rather than being raised by others, it is not clear that such a preference would constitute a right on her part. True, she may suffer greatly from the knowledge that her child will be thrown into the lottery of the adoption system, and that she will be unable to ensure its well-being, or even to know whether it is healthy, happy, doing well in school, etc.: for the law generally does not permit natural parents to remain in contact with their children, once they are adopted by another family. But there are surely better ways of dealing with these problems than by permitting infanticide in such cases. (It might help, for instance, if the natural parents of adopted children could at least receive some information about their progress, without necessarily being informed of the identity of the adopting family.) In contrast, a pregnant woman's right to protect her own life and health clearly outweighs other people's desire that the fetus be preserved—just as, when a persong life or limb is threatened by some wild animal, and

when the threat cannot be removed without killing the animal, the person's right to self-protection outweighs the desires of those who would prefer that the animal not be harmed. Thus, while the moment of birth may not mark any sharp discontinuity in the degree to which an infant possesses a right to life, it does mark the end of the mother's absolute right to determine its fate. Indeed, if and when a late-term abortion could be safely performed without killing the fetus, she would have no absolute right to insist on its death (e.g., if others wish to adopt it or pay for its care), for the same reason that she does not have a right to insist that a viable infant be killed. It remains true that according to my argument neither abortion nor the killing of neonates is properly considered a form of murder. Perhaps it is understandable that the law should classify infanticide as murder or homicide, since there is no other existing legal category which adequately or conveniently expresses the force of our society's disapproval of this action. But the moral distinction remains, and it has several important consequences. In the first place, it implies that when an infant is born into a society which—unlike ours—is so impoverished that it simply cannot care for it adequately without endangering the survival of existing persons, killing it or allowing it to die is not necessarily wrong—provided that there is no other society which is willing and able to provide such care. Most human societies, from those at the hunting and gathering stage of economic development to the highly civilized Greeks and Romans, have permitted the practice of infanticide under such unfortunate circumstances, and I would argue that it shows a serious lack of understanding to condemn them as morally backward for this reason alone. In the second place, the argument implies that when an infant is born with such severe physical anomalies that its life would predictably be a very short and/or very miserable one, even with the most heroic of medical treatment, and where its parents do not choose to bear the often crushing emotional, financial, and other burdens attendant upon the artificial prolongation of such a tragic life, it is not morally wrong to cease or withhold treatment, thus allowing the infant a painless death. It is wrong (and sometimes a form of murder) to practice involuntary euthanasia on persons, since they have the right to decide for themselves whether or not they wish to continue to live. But terminally ill neonates cannot make this decision for themselves, and thus it is incumbent upon responsible persons to make the decision for them, as best they can. The mistaken belief that infanticide is always tantamount to murder is responsible for a great deal of unnecessary suffering, not just on the part of infants which are made to endure needlessly prolonged and painful deaths, but also on the part of parents, nurses, and other involved

Warren: On the Moral and Legal Status of Abortion

persons, who must watch infants suffering needlessly, helpless to end that suffering in the most humane way. I am well await that these conclusions, however modest and reasonable they may seem to some people, strike other people as morally monstrous, and that some people might even prefer to abandon their previous support for women's right to abortion rather than accept a theory which leads to such conclusions about infanticide. But all that these facts show is that abortion is not an isolated moral issue; to fully understand the moral status of abortion we may have to reconsider

other moral issues as well, issues not just about infanticide and euthanasia, but also about the moral rights of women and of non-human animals. It is a philosopher's task to criticize mistaken beliefs which stand in the way of moral understanding, even when—perhaps especially when—those beliefs are popular and widespread. The belief that moral strictures against killing should apply equally to all genetically human entities, and only to genetically human entities, is such an error. The overcoming of this error will undoubtedly require long and often painful struggle; but it must be done.

Notes 1.

2. 3.

0 f course, the principle that it is (always) wrong to kill innocent human beings is in need of many other modifications, e.g., that it may be permissible to do so to save a greater number of other innocent human beings, but we may safely ignore these complications here. John Noonan, (1968), "Deciding Who Is Human," in Natural Law Forum, 13: 135. From here on, we will use "human" to mean genetically human, since the moral sense seems closely connected to, and perhaps derived from, the assumption that genetic humanity is sufficient for membership in the moral community.

A Defence of Abortion' Judith Jarvis Thomson

Most opposition to abortion relies on the premise that the fetus is a human being, a person, from the moment of conception. The premise is argued for, but, as I think, not well. Take, for example, the most common argument. We are asked to notice that the development of a human being from conception through birth into childhood is continuous; then it is said that to draw a line, to choose a point in this development and say "before this point the thing is not a person, after this point it is a person" is to make an arbitrary choice, a choice for which in the nature of things no good reason can be given. It is concluded that the fetus is, or anyway that we had better say it is, a person from the moment of conception. But this conclusion does not follow. Similar things might be said about the development of an acorn into an oak tree, and it does not follow that acorns are oak trees, or that we had better say

4.

5. 6.

Thomas L. Hayes, (1967), "A Biological View," in Commonweal 85 (17 March): 677-78; quoted by Daniel Callahan, in Abortion: Law, Choice and Morality (1970; Macmillan & Co.: London). That is, insofar as the death rate, for the woman, is higher for childbirth than for early abortion. My thanks to the following people, who were kind enough to read and criticize an earlier version of this paper: Herbert Gold, Gene Glass, Anne Lauterbach, Judith Thomson, Mary Mothersill, and Timothy Binkley.

they are. Arguments of this form are sometimes called "slippery slope arguments"—the phrase is perhaps self-explanatory—and it is dismaying that opponents of abortion rely on them so heavily and uncritically. I am inclined to agree, however, that the prospects for "drawing a line" in the development of the fetus look dim. I am inclined to think also that we shall probably have to agree that the fetus has already become a human person well before birth. Indeed, it comes as a surprise when one first learns how early in its life it begins to acquire human characteristics. By the tenth week, for example, it already has a face, arms and legs, fingers and toes; it has internal organs, and brain activity is detectable.2 On the other hand, I think that the premise is false, that the fetus is not a person from the moment of conception. A newly fertilized ovum, a newly implanted clump of cells, is no more a person than an acorn is an oak tree. But I shall not discuss any of this. For it seems to me to be of great interest to ask what happens if, for the sake of argument, we allow the premise. How, precisely, are we supposed to

Source: From Philosophy and Public Affairs 1 (1971, (1), 47-66. Reproduced with permission of Blackwell Publishing Ltd.

I

II ,21

II

111111

5 Reproductive Ethics

get from there to the conclusion that abortion is morally impermissible? Opponents of abortion commonly spend most of their time establishing that the fetus is a person, and hardly any time explaining the step from there to the impermissibility of abortion. Perhaps they think the step too simple and obvious to require much comment. Or perhaps instead they are simply being economical in argument. Many of those who defend abortion rely on the premise that the fetus is not a person, but only a bit of tissue that will become a person at birth; and why pay out more arguments than you have to? Whatever the explanation, I suggest that the step they take is neither easy nor obvious, that it calls for closer examination than it is commonly given, and that when we do give it this closer examination we shall feel inclined to reject it. I propose, then, that we grant that the fetus is a person from the moment of conception. How does the argument go from here? Something like this, I take it. Every person has a right to life. So the fetus has a right to life. No doubt the mother has a right to decide what shall happen in and to her body; everyone would grant that. But surely a person's right to life is stronger and more stringent than the mother's right to decide what happens in and to her body, and so outweighs it. So the fetus may not be killed; an abortion may not be performed. It sounds plausible. But now let me ask you to imagine this. You wake up in the morning and find yourself back to back in bed with an unconscious violinist. A famous unconscious violinist. He has been found to have a fatal kidney ailment, and the Society of Music Lovers has canvassed all the available medical records and found that you alone have the right blood type to help. They have therefore kidnapped you, and last night the violinist's circulatory system was plugged into yours, so that your kidneys can be used to extract poisons from his blood as well as your own. The director of the hospital now tells you, "Look, we're sorry the Society of Music Lovers did this to you—we would never have permitted it if we had known. But still, they did it, and the violinist now is plugged into you. To unplug you would be to kill him. But never mind, it's only for nine months. By then he will have recovered from his ailment, and can safely be unplugged from you." Is it morally incumbent on you to accede to this situation? No doubt it would be very nice of you if you did, a great kindness. But do you have to accede to it? What if it were not nine months, but nine years? Or longer still? What if the director of the hospital says, 'Tough luck, I agree, but you've now got to stay in bed, with the violinist plugged into you, for the rest of your life. Because remember this. All persons have a right

to life, and violinists are persons. Granted you have a right to decide what happens in and to your body, but a persons right to life outweighs your right to decide what happens in and to your body. So you cannot ever be unplugged from him." I imagine you would regard this as outrageous, which suggests that something really is wrong with that plausible-sounding argument I mentioned a moment ago. In this case, of course, you were kidnapped; you didn't volunteer for the operation that plugged the violinist into your kidneys. Can those who oppose abortion on the ground I mentioned make an exception for a pregnancy due to rape? Certainly. They can say that persons have a right to life only if they didn't come into existence because of rape; or they can say that all persons have a right to life, but that some have less of a right to life than others, in particular, that those who came into existence because of rape have less. But these statements have a rather unpleasant sound. Surely the question of whether you have a right to life at all, or how much of it you have, shouldn't turn on the question of whether or not you are the product of a rape. And in fact the people who oppose abortion on the ground I mentioned do not make this distinction, and hence do not make an exception in case of rape. Nor do they make an exception for a case in which the mother has to spend the nine months of her pregnancy in bed. They would agree that would be a great pity, and hard on the mother, but all the same, all persons have a right to life, the fetus is a person, and so on. I suspect, in fact, that they would not make an exception for a case in which, miraculously enough, the pregnancy went on for nine years, or even the rest of the mother's life. Some won't even make an exception for a case in which continuation of the pregnancy is likely to shorten the mother's life; they regard abortion as impermissible even to save the mother's life. Such cases are nowadays very rare, and many opponents of abortion do not accept this extreme view. All the same, it is a good place to begin: a number of points of interest come out in respect to it. 1 Let us call the view that abortion is impermissible even to save the mother's life "the extreme view" I want to suggest first that it does not issue from the argument I mentioned earlier without the addition of some fairly powerful premises. Suppose a woman has become pregnant, and now learns that she has a cardiac condition such that she will die if she carries the baby to term.

Thomson: A Defence of Abortion

What may be done for her? The fetus, being a person, has a right to life, but as the mother is a person too, so has she a right to life. Presumably they have an equal right to life. How is it supposed to come out that an abortion may not be performed? If mother and child have an equal right to life, shouldn't we perhaps flip a coin? Or should we add to the motherb right to life her right to decide what happens in and to her body, which everybody seems to be ready to grant—the sum of her rights now outweighing the fetusb right to life? The most familiar argument here is the following. We are told that performing the abortion would be directly killing' the child, whereas doing nothing would not be killing the mother, but only letting her die. Moreover, in killing the child, one would be killing an innocent person, for the child has committed no crime, and is not aiming at his motherb death. And then there are a variety of ways in which this might be continued. (1) But as directly killing an innocent person is always and absolutely impermissible, an abortion may not be performed. Or (2) as directly killing an innocent person is murder, and murder is always and absolutely impermissible, an abortion may not be performed.' Or (3) as ones duty to refrain from directly killing an innocent person is more stringent than ones duty to keep a person from dying, an abortion may not be performed. Or (4) if ones only options are directly killing an innocent person or letting a person die, one must prefer letting the person die, and thus an abortion may not be performed.' Some people seem to have thought that these are not further premises which must be added if the conclusion is to be reached, but that they follow from the very fact that an innocent person has a right to life.' But this seems to me to be a mistake, and perhaps the simplest way to show this is to bring out that while we must certainly grant that innocent persons have a right to life, the theses in (1) through (4) are all false. Take (2), for example. If directly killing an innocent person is murder, and thus is impermissible, then the motherb directly killing the innocent person inside her is murder, and thus is impermissible. But it cannot seriously be thought to be murder if the mother performs an abortion on herself to save her life. It cannot seriously be said that she must refrain, that she must sit passively by and wait for her death. Let us look again at the case of you and the violinist. There you are, in bed with the violinist, and the director of the hospital says to you, "It's all most distressing, and I deeply sympathize, but you see this is putting an additional strain on your kidneys, and you'll be dead within the month. But you have to stay where you are all the same. Because unplugging you would be directly killing an innocent violinist, and that's murder, and that's impermissible."

If anything in the world is true, it is that you do not commit murder, you do not do what is impermissible, if you reach around to your back and unplug yourself from that violinist to save your life. The main focus of attention in writings on abortion has been on what a third party may or may not do in answer to a request from a woman for an abortion. This is in a way understandable. Things being as they are, there isn't much a woman can safely do to abort herself. So the question asked is what a third party may do, and what the mother may do, if it is mentioned at all, is deduced, almost as an afterthought, from what it is concluded that third parties may do. But it seems to me that to treat the matter in this way is to refuse to grant to the mother that very status of person which is so firmly insisted on for the fetus. For we cannot simply read off what a person may do from what a third party may do. Suppose you find yourself trapped in a tiny house with a growing child. I mean a very tiny house, and a rapidly growing child—you are already up against the wall of the house and in a few minutes you'll be crushed to death. The child on the other hand won't be crushed to death; if nothing is done to stop him from growing he'll be hurt, but in the end he'll simply burst open the house and walk out a free man. Now I could well understand it if a bystander were to say, "There's nothing we can do for you. We cannot choose between your life and his, we cannot be the ones to decide who is to live, we cannot intervene." But it cannot be concluded that you too can do nothing, that you cannot attack it to save your life. However innocent the child may be, you do not have to wait passively while it crushes you to death. Perhaps a pregnant woman is vaguely felt to have the status of house, to which we don't allow the right of self-defence. But if the woman houses the child, it should be remembered that she is a person who houses it. I should perhaps stop to say explicitly that I am not claiming that people have a right to do anything whatever to save their lives. I think, rather, that there are drastic limits to the right of self-defence. If someone threatens you with death unless you torture someone else to death, I think you have not the right, even to save your life, to do so. But the case under consideration here is very different. In our case there are only two people involved, one whose life is threatened, and one who threatens it. Both are innocent: the one who is threatened is not threatened because of any fault, the one who threatens does not threaten because of any fault. For this reason we may feel that we bystanders cannot intervene. But the person threatened can. In sum, a woman surely can defend her life against the threat to it posed by the unborn child, even if doing so

II iY

256

1111111

5 Reproductive Ethics

involves its death. And this shows not merely that the theses in (1) through (4) are false; it shows also that the extreme view of abortion is false, and so we need not canvass any other possible ways of arriving at it from the argument I mentioned at the outset. 2 The extreme view could of course be weakened to say that while abortion is permissible to save the mother's life, it may not be performed by a third party, but only by the mother herself. But this cannot be right either. For what we have to keep in mind is that the mother and the unborn child are not like two tenants in a small house which has, by an unfortunate mistake, been rented to both: the mother owns the house. The fact that she does adds to the offensiveness of deducing that the mother can do nothing from the supposition that third parties can do nothing. But it does more than this: it casts a bright light on the supposition that third parties can do nothing. Certainly it lets us see that a third party who says "I cannot choose between you" is fooling himself if he thinks this is impartiality. If Jones has found and fastened on a certain coat, which he needs to keep him from freezing, but which Smith also needs to keep him from freezing, then it is not impartiality that says "I cannot choose between you" when Smith owns the coat. Women have said again and again "This body is my body!" and they have reason to feel angry, reason to feel that it has been like shouting into the wind. Smith, after all, is hardly likely to bless us if we say to him, "Of course it's your coat, anybody would grant that it is. But no one may choose between you and Jones who is to have it." We should really ask what it is that says "no one may choose" in the face of the fact that the body that houses the child is the mother's body. It may be simply a failure to appreciate this fact. But it may be something more interesting, namely the sense that one has a right to ref use to lay hands on people, even where it would be just and fair to do so, even where justice seems to require that somebody do so. Thus justice might call for somebody to get Smith's coat back from Jones, and yet you have a right to ref use to be the one to lay hands on Jones, a right to refuse to do physical violence to him. This, I think, must be granted. But then what should be said is not "no one may choose," but only "I cannot choose," and indeed not even this, but "I will not act," leaving it open that somebody else can or should, and in particular that anyone in a position of authority, with the job of securing people's rights, both can and should. So this is no difficulty. I have not been arguing that any given third party must accede to the mother's

request that he perform an abortion to save her life, but only that he may. I suppose that in some views of human life the mother's body is only on loan to her, the loan not being one which gives her any prior claim to it. One who held this view might well think it impartiality to say "I cannot choose." But I shall simply ignore this possibility. My own view is that if a human being has any just, prior claim to anything at all, he has a just, prior claim to his own body. And perhaps this needn't be argued for here anyway, since, as I mentioned, the arguments against abortion we are looking at do grant that the woman has a right to decide what happens in and to her body. But although they do grant it, I have tried to show that they do not take seriously what is done in granting it. I suggest the same thing will reappear even more clearly when we turn away from cases in which the mother's life is at stake, and attend, as I propose we now do, to the vastly more common cases in which a woman wants an abortion for some less weighty reason than preserving her own life. 3 Where the mother's life is not at stake, the argument I mentioned at the outset seems to have a much stronger pull. "Everyone has a right to life, so the unborn person has a right to life." And isn't the child right to life weightier than anything other than the mother's own right to life, which she might put forward as ground for an abortion? This argument treats the right to life as if it were unproblematic. It is not, and this seems to me to be precisely the source of the mistake. For we should now, at long last, ask what it comes to, to have a right to life. In some views having a right to life includes having a right to be given at least the bare minimum one needs for continued life. But suppose that what in fact is the bare minimum a man needs for continued life is something he has no right at all to be given? If I am sick unto death, and the only thing that will save my life is the touch of Henry Fonda's cool hand on my fevered brow, then all the same, I have no right to be given the touch of Henry Fonda's cool hand on my fevered brow. It would be frightfully nice of him to fly in from the West Coast to provide it. It would be less nice, though no doubt well meant, if my friends flew out to the West Coast and carried Henry Fonda back with them. But I have no right at all against anybody that he should do this for me. Or again, to return to the story I told earlier, the fact that for continued life that violinist needs the continued

Thomson: A Defence of Abortion

use of your kidneys does not establish that he has a right to be given the continued use of your kidneys. He certainly has no right against you that you should give him continued use of your kidneys. For nobody has any right to use your kidneys unless you give him such a right; and nobody has the right against you that you shall give him this right—if you do allow him to go on using your kidneys, this is a kindness on your part, and not something he can claim from you as his due. Nor has he any right against anybody else that they should give him continued use of your kidneys. Certainly he had no right against the Society of Music Lovers that they should plug him into you in the first place. And if you now start to unplug yourself, having learned that you will otherwise have to spend nine years in bed with him, there is nobody in the world who must try to prevent you, in order to see to it that he is given something he has a right to be given. Some people are rather stricter about the right to life. In their view, it does not include the right to be given anything, but amounts to, and only to, the right not to be killed by anybody. But here a related difficulty arises. If everybody is to refrain from lulling that violinist, then everybody must refrain from doing a great many different sorts of things. Everybody must refrain from slitting his throat, everybody must refrain from shooting him—and everybody must refrain from unplugging you from him. But does he have a right against everybody that they shall refrain from unplugging you from him? To refrain from doing this is to allow him to continue to use your kidneys. It could be argued that he has a right against us that we should allow him to continue to use your kidneys. That is, while he had no right against us that we should give him the use of your kidneys, it might be argued that he anyway has a right against us that we shall not now intervene and deprive him of the use of your kidneys. I shall come back to third-party interventions later. But certainly the violinist has no right against you that you shall allow him to continue to use your kidneys. As I said, if you do allow him to use them, it is a kindness on your part, and not something you owe him. The difficulty I point to here is not peculiar to the right to life. It reappears in connection with all the other natural rights, and it is something which an adequate account of rights must deal with. For present purposes it is enough just to draw attention to it. But I would stress that I am not arguing that people do not have a right to life—quite to the contrary, it seems to me that the primary control we must place on the acceptability of an account of rights is that it should turn out in that account to be a truth that all persons have a right to life. I am arguing only that having a right to life

does not guarantee having either a right to be given the use of or a right to be allowed continued use of another person's body—even if one needs it for life itself. So the right to life will not serve the opponents of abortion in the very simple and clear way in which they seem to have thought it would. 4 There is another way to bring out the difficulty. In the most ordinary sort of case, to deprive someone of what he has a right to is to treat him unjustly. Suppose a boy and his small brother are jointly given a box of chocolates for Christmas. If the older boy takes the box and refuses to give his brother any of the chocolates, he is unjust to him, for the brother has been given a right to half of them. But suppose that, having learned that otherwise it means nine years in bed with that violinist, you unplug yourself from him. You surely are not being unjust to him, for you gave him no right to use your kidneys, and no one else can have given him any such right. But we have to notice that in unplugging yourself, you are killing him; and violinists, like everybody else, have a right to life, and thus in the view we were considering just now, the right not to be killed. So here you do what he supposedly has a right you shall not do, but you do not act unjustly to him in doing it. The emendation which may be made at this point is this the right to life consists not in the right not to be killed, but rather in the right not to be killed unjustly. This runs a risk of circularity, but never mind: it would enable us to square the fact that the violinist has a right to life with the fact that you do not act unjustly toward him in unplugging yourself, thereby killing him. For if you do not kill him unjustly, you do not violate his right to life, and so it is no wonder you do him no injustice. But if this emendation is accepted, the gap in the argument against abortion stares us plainly in the face: it is by no means enough to show that the fetus is a person, and to remind us that all persons have a right to life—we need to be shown also that killing the fetus violates its right to life, i.e., that abortion is unjust killing. And is it? I suppose we may take it as a datum that in a case of pregnancy due to rape the mother has not given the unborn person a right to the use of her body for food and shelter. Indeed, in what pregnancy could it be supposed that the mother has given the unborn person such a right? It is not as if there were unborn persons drifting about the world, to whom a woman who wants a child says "I invite you in." But it might be argued that there are other ways one can have acquired a right to the use of another

E

258

II

II

.- 311

IIIIJl

5 Reproductive Ethics

person's body than by having been invited to use it by that person. Suppose a woman voluntarily indulges in intercourse, knowing of the chance it will issue in pregnancy, and then she does become pregnant; is she not in part responsible for the presence, in fact the very existence, of the unborn person inside her? No doubt she did not invite it in. But doesn't her partial responsibility for its being there itself give it a right to the use of her body?' If so, then her aborting it would be more like the boy's taking away the chocolates, and less like your unplugging yourself from the violinist—doing so would be depriving it of what it does have a right to, and thus would be doing it an injustice. And then, too, it might be asked whether or not she can kill it even to save her own life: if she voluntarily called it into existence, how can she now kill it, even in self-defence? The first thing to be said about this is that it is something new. Opponents of abortion have been so concerned to make out the independence of the fetus, in order to establish that it has a right to life, just as its mother does, that they have tended to overlook the possible support they might gain from making out that the fetus is dependent on the mother, in order to establish that she has a special kind of responsibility for it, a responsibility that gives it rights against her which are not possessed by any independent person—such as an ailing violinist who is a stranger to her. On the other hand, this argument would give the unborn person a right to its mother's body only if her pregnancy resulted from a voluntary act, undertaken in full knowledge of the chance a pregnancy might result from it. It would leave out entirely the unborn person whose existence is due to rape. Pending the availability of some further argument, then, we would be left with the conclusion that unborn persons whose existence is due to rape have no right to the use of their mothers' bodies, and thus that aborting them is not depriving them of anything they have a right to and hence is not unjust killing. And we should also notice that it is not at all plain that this argument really does go even as far as it purports to. For there are cases and cases, and the details make a difference. If the room is stuffy, and I therefore open a window to air it, and a burglar climbs in, it would be absurd to say, "Ah, now he can stay, she's given him a right to the use of her house—for she is partially responsible for his presence there, having voluntarily done what enabled him to get in, in full knowledge that there are such things as burglars, and that burglars burgle." It would be still more absurd to say this if I had had bars installed outside my windows, precisely to prevent burglars from getting in, and a

burglar got in only because of a defect in the bars. It remains equally absurd if we imagine it is not a burglar who climbs in, but an innocent person who blunders or falls in. Again, suppose it were like this: peopleseeds drift about in the air like pollen, and if you open your windows, one may drift in and take root in your carpets or upholstery. You don't want children, so you fix up your windows with fine mesh screens, the very best you can buy. As can happen, however, and on very, very rare occasions does happen, one of the screens is defective; and a seed drifts in and takes root. Does the person-plant who now develops have a right to the use of your house? Surely not—despite the fact that you voluntarily opened your windows, you knowingly kept carpets and upholstered furniture, and you knew that screens were sometimes defective. Someone may argue that you are responsible for its rooting, that it does have a right to your house, because after all you could have lived out your life with bare floors and furniture, or with sealed windows and doors. But this wont do—for by the same token anyone can avoid a pregnancy due to rape by having a hysterectomy, or anyway by never leaving home without a (reliable!) army. It seems to me that the argument we are looking at can establish at most that there are some cases in which the unborn person has a right to the use of its mother's body, and therefore some cases in which abortion is unjust killing. There is room for much discussion and argument as to precisely which, if any. But I think we should side-step this issue and leave it open, for at any rate the argument certainly does not establish that all abortion is unjust killing.

5 There is room for yet another argument here, however. We surely must all grant that there may be cases in which it would be morally indecent to detach a person from your body at the cost of his life. Suppose you learn that what the violinist needs is not nine years of your life, but only one hour all you need do to save his life is to spend one hour in that bed with him. Suppose also that letting him use your kidneys for that one hour would not affect your health in the slightest. Admittedly you were kidnapped. Admittedly you did not give anyone permission to plug him into you. Nevertheless it seems to me plain you ought to allow him to use your kidneys for that hour—it would be indecent to refuse. Again, suppose pregnancy lasted only an hour, and constituted no threat to life or health. And suppose that a woman becomes pregnant as a result of rape. Admittedly she did not voluntarily do anything to bring about the existence of a child. Admittedly she

Thomson: A Defence of Abortion

did nothing at all which would give the unborn person a right to the use of her body. All the same it might well be said, as in the newly emended violinist story, that she ought to allow it to remain for that hour—that it would be indecent in her to refuse. Now some people are inclined to use the term "right" in such a way that it follows from the fact that you ought to allow a person to use your body for the hour he needs, that he has a right to use your body for the hour he needs, even though he has not been given that right by any person or act. They may say that it follows also that if you refuse, you act unjustly toward him. This use of the term is perhaps so common that it cannot be called wrong; nevertheless it seems to me to be an unfortunate loosening of what we would do better to keep a tight rein on. Suppose that box of chocolates I mentioned earlier had not been given to both boys jointly, but was given only to the older boy. There he sits, stolidly eating his way through the box, his small brother watching enviously. Here we are likely to say, "You ought not to be so mean. You ought to give your brother some of those chocolates." My own view is that it just does not follow from the truth of this that the brother has any right to any of the chocolates. If the boy refuses to give his brother any, he is greedy, stingy, callous—but not unjust. I suppose that the people I have in mind will say it does follow that the brother has a right to some of the chocolates, and thus that the boy does act unjustly if he refuses to give his brother any. But the effect of saying this is to obscure what we should keep distinct, namely the difference between the boy's refusal in this case and the boy's refusal in the earlier case, in which the box was given to both boys jointly, and in which the small brother thus had what was from any point of view clear title to half. A further objection to so using the term "right" that from the fact that A ought to do a thing for B, it follows that B has a right against A that A do it for him, is that it is going to make the question of whether or not a man has a right to a thing turn on how easy it is to provide him with it; and this seems not merely unfortunate, but morally unacceptable. Take the case of Henry Fonda again. I said earlier that I had no right to the touch of his cool hand on my fevered brow, even though I needed it to save my life. I said it would be frightfully nice of him to fly in from the West Coast to provide me with it, but that I had no right against him that he should do so. But suppose he isn't on the West Coast. Suppose he has only to walk across the room, place a hand briefly on my brow—and lo, my life is saved. Then surely he ought to do it, it would be indecent to refuse. Is it to be said, "Ah, well, it follows

that in this case she has a right to the touch of his hand on her brow, and so it would be an injustice in him to refuse"? So that I have a right to it when it is easy for him to provide it, though no right when it's hard? It's rather a shocking idea that anyone's rights should fade away and disappear as it gets harder and harder to accord them to him. So my own view is that even though you ought to let the violinist use your kidneys for the one hour he needs, we should not conclude that he has a right to do so—we should say that if you refuse, you are, like the boy who owns all the chocolates and will give none away, self-centred and callous, indecent in fact, but not unjust. And similarly, that even supposing a case in which a woman pregnant due to rape ought to allow the unborn person to use her body for the hour he needs, we should not conclude that he has a right to do so; we should conclude that she is self-centred, callous, indecent, but not unjust, if she refuses. The complaints are no less grave; they are just different. However, there is no need to insist on this point. If anyone does wish to deduce "he has a right" from "you ought," then all the same he must surely grant that there are cases in which it is not morally required of you that you allow that violinist to use your kidneys, and in which he does not have a right to use them, and in which you do not do him an injustice if you refuse. And so also for mother and unborn child. Except in such cases as the unborn person has a right to demand it—and we were leaving open the possibility that there may be such cases—nobody is morally required to make large sacrifices, of health, of all other interests and concerns, of all other duties and commitments, for nine years, or even for nine months, in order to keep another person alive.

6 We have in fact to distinguish between two kinds of Samaritan: the Good Samaritan and what we might call the Minimally Decent Samaritan. The story of the Good Samaritan, you will remember, goes like this: A certain man went down from Jerusalem to Jericho, and fell among thieves, which stripped him of his raiment, and wounded him, and departed, leaving him half dead. And by chance there came down a certain priest that way; and when he saw him, he passed by on the other side. And likewise a Levite, when he was at the place, came and looked on him, and passed by

M III

II

11

II 311

111111

5 Reproductive Ethics

on the other side. But a certain Samaritan, as he journeyed, came where he was; and when he saw him he had compassion on him. And went to him, and bound up his wounds, pouring in oil and wine, and set him on his own beast, and brought him to an inn, and took care of him. And on the morrow, when he departed, he took out two pence, and gave them to the host, and said unto him, "Take care of him; and whatsoever thou spendest more, when I come again, I will repay thee." (Luke 10: 30-35) The Good Samaritan went out of his way, at some cost to himself, to help one in need of it. We are not told what the options were, that is, whether or not the priest and the Levite could have helped by doing less than the Good Samaritan did, but assuming they could have, then the fact they did nothing at all shows they were not even Minimally Decent Samaritans, not because they were not Samaritans, but because they were not even minimally decent. These things are a matter of degree, of course, but there is a difference, and it comes out perhaps most clearly in the story of Kitty Genovese, who, as you will remember, was murdered while 38 people watched or listened, and did nothing at all to help her. A Good Samaritan would have rushed out to give direct assistance against the murderer. Or perhaps we had better allow that it would have been a Splendid Samaritan who did this, on the ground that it would have involved a risk of death for himself. But the 38 not only did not do this, they did not even trouble to pick up a phone to call the police. Minimally Decent Samaritanism would call for doing at least that, and their not having done it was monstrous. After telling the story of the Good Samaritan, Jesus said, "Go, and do thou likewise." Perhaps he meant that we are morally required to act as the Good Samaritan did. Perhaps he was urging people to do more than is morally required of them. At all events it seems plain that it was not morally required of any of the 38 that he rush out to give direct assistance at the risk of his own life, and that it is not morally required of anyone that he give long stretches of his life—nine years or nine months—to sustaining the life of a person who has no special right (we were leaving open the possibility of this) to demand it. Indeed, with one rather striking class of exceptions, no one in any country in the world is legally required to do anywhere near as much as this for anyone else. The class of exceptions is obvious. My main concern here is not the state of the law in

respect to abortion, but it is worth drawing attention to the fact that in no state in this country is any man compelled by law to be even a Minimally Decent Samaritan to any person; there is no law under which charges could be brought against the 38 who stood by while Kitty Genovese died. By contrast, in most states in this country women are compelled by law to be not merely Minimally Decent Samaritans but Good Samaritans to unborn persons inside them. This doesn't by itself settle anything one way or the other, because it may well be argued that there should be laws in this country—as there are in many European countries—compelling at least Minimally Decent Samaritanism.8 But it does show that there is a gross injustice in the existing state of the law. And it shows also that the groups currently working against liberalization of abortion laws, in fact working toward having it declared unconstitutional for a state to permit abortion, had better start working for the adoption of Good Samaritan laws generally, or earn the charge that they are acting in bad faith. I should think, myself, that Minimally Decent Samaritan laws would be one thing, Good Samaritan laws quite another, and in fact highly improper. But we are not here concerned with the law. What we should ask is not whether anybody should be compelled by law to be a Good Samaritan, but whether we must accede to a situation in which somebody is being compelled—by nature, perhaps—to be a Good Samaritan. We have, in other words, to look now at third-party interventions. I have been arguing that no person is morally required to make large sacrifices to sustain the life of another who has no right to demand them, and this even where the sacrifices do not include life itself; we are not morally required to be Good Samaritans or anyway Very Good Samaritans to one another. But what if a man cannot extricate himself from such a situation? What if he appeals to us to extricate him? It seems to me plain that there are cases in which we can, cases in which a Good Samaritan would extricate him. There you are, you were kidnapped, and nine years in bed with that violinist lie ahead of you. You have your own life to lead. You are sorry, but you simply cannot see giving up so much of your life to the sustaining of his. You cannot extricate yourself, and ask us to do so. I should have thought that—in light of his having no right to the use of your body—it was obvious that we do not have to accede to your being forced to give up so much. We can do what you ask. There is no injustice to the violinist in our doing so.

Thomson: A Defence of Abortion 261

7 Following the lead of the opponents of abortion, I have throughout been speaking of the fetus merely as a person, and what I have been asking is whether or not the argument we began with, which proceeds only from the fetus's being a person, really does establish its conclusion. I have argued that it does not. But of course there are arguments and arguments, and it may be said that I have simply fastened on the wrong one. It may be said that what is important is not merely the fact that the fetus is a person, but that it is a person for whom the woman has a special kind of responsibility issuing from the fact that she is its mother. And it might be argued that all my analogies are therefore irrelevant—for you do not have that special kind of responsibility for that violinist, Henry Fonda does not have that special kind of responsibility for me. And our attention might be drawn to the fact that men and women both are compelled by law to provide support for their children. I have in effect dealt (briefly) with this argument in section 4 above; but a (still briefer) recapitulation now may be in order. Surely we do not have any such "special responsibility" for a person unless we have assumed it, explicitly or implicitly. If a set of parents do not try to prevent pregnancy, do not obtain an abortion, and then at the time of birth of the child do not put it out for adoption, but rather take it home with them, then they have assumed responsibility for it, they have given it rights, and they cannot now withdraw support from it at the cost of its life because they now find it difficult to go on providing for it. But if they have taken all reasonable precautions against having a child, they do not simply by virtue of their biological relationship to the child who comes into existence have a special responsibility for it. They may wish to assume responsibility for it, or they may not wish to. And I am suggesting that if assuming responsibility for it would require large sacrifices, then they may refuse. A Good Samaritan would not refuse—or anyway a Splendid Samaritan, if the sacrifices that had to be made were enormous. But then so would a Good Samaritan assume responsibility for that violinist; so would Henry Fonda, if he is a Good Samaritan, fly in from the West Coast and assume responsibility for me. 8 My argument will be found unsatisfactory on two counts by many of those who want to regard abortion as morally permissible. First, while I do argue that abortion is not impermissible, I do not argue that it is always permissible. There may well be cases in which carrying

the child to term requires only Minimally Decent Samaritanism of the mother, and this is a standard we must not fall below. I am inclined to think it a merit of my account precisely that it does not give a general yes or a general no. It allows for and supports our sense that, for example, a sick and desperately frightened 14-yearold schoolgirl, pregnant due to rape, may of course choose abortion, and that any law which rules this out is an insane law: And it also allows for and supports our sense that in other cases resort to abortion is even positively indecent. It would be indecent in the woman to request an abortion, and indecent in a doctor to perform it, if she is in her seventh month, and wants the abortion just to avoid the nuisance of postponing a trip abroad. The very fact that the arguments I have been drawing attention to treat all cases of abortion, or even all cases of abortion in which the mother's life is not at stake, as morally on a par ought to have made them suspect at the outset. Secondly, while I am arguing for the permissibility of abortion in some cases, I am not arguing for the right to secure the death of the unborn child. It is easy to confuse these two things in that up to a certain point in the life of the fetus it is not able to survive outside the mother's body; hence removing it from her body guarantees its death. But they are importantly different. I have argued that you are not morally required to spend nine months in bed, sustaining the life of that violinist; but to say this is by no means to say that if, when you unplug yourself, there is a miracle and he survives, you then have a right to turn round and slit his throat. You may detach yourself even if this costs him his life; you have no right to be guaranteed his death, by some other means, if unplugging yourself does not kill him. There are some people who will feel dissatisfied by this feature of my argument. A woman may be utterly devastated by the thought of a child, a bit of herself, put out for adoption and never seen or heard of again. She may therefore want not merely that the child be detached from her, but more, that it die. Some opponents of abortion are inclined to regard this as beneath contempt—thereby showing insensitivity to what is surely a powerful source of despair. All the same, I agree that the desire for the child's death is not one which anybody may gratify, should it turn out to be possible to detach the child alive. At this place, however, it should be remembered that we have only been pretending throughout that the fetus is a human being from the moment of conception. A very early abortion is surely not the killing of a person, and so is not dealt with by anything I have said here.

■■ 1

11

5 Reproductive Ethics

Notes 1. 2.

3.

4.

I am very much indebted to James Thomson for discussion, criticism, and many helpful suggestions. Daniel Callahan, (1970), Abortion: Law, Choice and Morality (New York), 373. This book gives a fascinating survey of the available information on abortion. The Jewish tradition is surveyed in David M. Feldman, (1968), Birth Control in Jewish Law (New York), Part 5; the Catholic tradition, in John T. Noonan, Jr, (1970), "An Almost Absolute Value in History," in The Morality of Abortion, ed. John T. Noonan, Jr (Cambridge, MA). The term "direct" in the arguments I refer to is a technical one. Roughly, what is meant by "direct killing" is either killing as an end in itself, or killing as a means to some end, for example, the end of saving someone else's life. See note 6, below, for an example of its use. Cf. Encyclical Letter of Pope Pius XI on Christian Marriage, St. Paul Editions (Boston, n.d.), 32: "however much we may pity the mother whose health and even life is gravely imperiled in the performance of the duty allotted to her by nature, nevertheless what could ever be a sufficient reason for excusing in any way the direct murder of the innocent? This is precisely what we are dealing with here." Noonan (The Morality of Abortion, p. 43) reads this as follows: "What cause can ever avail to excuse in any way the direct killing of the innocent? For it is a question of that."

A Third Way L.W. Sumner [. . .] In most countries in the West, public discussion of abortion has been distorted by the dominance of two . . . views. The liberal view, espoused by "pro-choice" groups, holds that (voluntary) abortion is always morally innocuous and (therefore) that the only acceptable abortion policy is one which treats abortion as another variety of minor elective surgery. The conservative view, espoused by "pro-life" groups, holds that abortion is always morally serious and (therefore) that the only acceptable abortion policy is one which treats abortion as another variety of homicide. Because they define the extremities of the continuum of possible positions, and because each is sufficiently simple and forceful to be advocated by a powerful movement, these established views constitute the familiar reference points in our abortion landscape. Yet neither has managed to command the

5.

6.

7.

8.

The thesis in (4) is in an interesting way weaker than those in (1), (2), and (3): they rule out abortion even in cases in which both mother and child will die if the abortion is not performed. By contrast, one who held the view expressed in (4) could consistently say that one needn't prefer letting two persons die to killing one. Cf. the following passage from Pius xii, Address to the Italian Catholic Society of Midwives: "The baby in the maternal breast has the right to life immediately from God. Hence there is no man, no human authority, no science, no medical, eugenic, social, economic or moral 'indication' which can establish or grant a valid juridical ground for a direct deliberate disposition of an innocent human life, that is a disposition which looks to its destruction either as an end or as a means to another end perhaps in itself not illicit. The baby, still not born, is a man in the same degree and for the same reason as the mother" (quoted in Noonan, The Morality of Abortion, 45). The need for a discussion of this argument was brought home to me by members of the Society for Ethical and Legal Philosophy, to whom this paper was originally presented. The need for a discussion of this argument was brought home to me by members of the Society for Ethical and Legal Philosophy, to whom this paper was originally presented.

allegiance of more than a small minority of the public. For the rest of us who are unwilling to embrace either of the extreme options, the problem has been the lack of a well-defined middle ground between them. In contrast to the power of the established views more moderate alternatives may appear both indistinct and indecisive. Public distrust of the established views is well grounded: neither stands up under critical scrutiny' If their demise is not to leave us without any credible view of abortion, three tasks must be successfully completed. The first is to define a third way with abortion and to distinguish it from both of the views which it will supersede. The second is to give it an intuitive defense by showing that it coheres better than either of its predecessors with our considered moral judgments both on abortion itself and on closely related issues. Then, finally, the third way must be grounded in a moral theory. The first two of these tasks will be undertaken here; the more daunting theoretical challenge is confronted elsewhere.'

Source: From Abortion and Moral Theory (Princeton NJ: Princeton University Press, 1981), 124-60. Reprinted with permission.

Sumner: A Third Way 263

1. Specifications

Despite their opposition, the two established views suffer from similar defects. Collating their failures will provide us with some positive guidelines to follow in building a more satisfactory alternative. The central issue in the morality of abortion is the moral status of the fetus. Let us say that a creature has moral standing if, for the purpose of moral decision-making, it must be counted for something in its own right. To count for nothing is to have no moral standing to count for as much as possible (as much, that is, as any creature does) is to have full moral standing. We may, for the purpose of the present discussion, make this rather vague notion more precise by adopting the rights vocabulary favored by both of the established views. We will suppose that having (some) moral standing is equivalent to having (some) right to life. The central issue in the morality of abortion is then whether fetuses have moral standing in this sense.3 The conservative view, and also the more naive versions of the liberal view, select a precise point (conception, birth, etc.) as the threshold of moral standing, implying that the transition from no standing to full standing occurs abruptly. In doing so they rest more weight on these sudden events than they are capable of bearing. A view that avoids this defect will allow full moral standing to be acquired gradually. It will therefore attempt to locate not a threshold point but a threshold period or stage. Both of the established views attribute a uniform moral status to all fetuses, regardless of their dissimilarities. Each, for example, counts a newly conceived zygote for precisely as much (or as little) as a full-term fetus, despite the enormous differences between them. A view that avoids this defect will assign moral status differentially so that the threshold stage occurs sometime during pregnancy. A consequence of the uniform approach adopted by both of the established views is that neither can attach any significance to the development of the fetus during gestation. Yet this development is the most obvious feature of gestation. A view that avoids this defect will base the (differential) moral standing of the fetus at least in part on its level of development. It will thus assign undeveloped fetuses a moral status akin to that of ova and spermatozoa, whereas it will assign developed fetuses a moral status akin to that of infants. So far, then, an adequate view of the fetus must be gradual, differential, and developmental. It must also be derived from a satisfactory criterion of moral standing. Such a criterion must be general (applicable to beings other than fetuses), it must connect moral

standing with the empirical properties of such beings, and it must be morally relevant. Its moral relevance is partly testable by appeal to intuition, for arbitrary or shallow criteria will be vulnerable to counterexamples. But the final test of moral relevance is grounding in a moral theory. An adequate view of the fetus promises a morally significant division between early abortions (before the threshold stage) and late abortions (after the threshold stage). It also promises borderline cases (during the threshold stage). Wherever that stage is located, abortions that precede it will be private matters, since the fetus will at that stage lack moral standing. Thus the provisions of the liberal view will apply to early abortions: they will be morally innocent (as long as the usual conditions of maternal consent, etc., are satisfied) and ought to be legally unregulated (except for rules equally applicable to all other medical procedures). Early abortion will have the same moral status as contraception. Abortions that follow the threshold stage will be interpersonal matters, since the fetus will at that stage possess moral standing. Thus the provisions of the conservative view will apply to late abortions: they must be assessed on a case-by-case basis, and they ought to be legally permitted only on appropriate grounds. Late abortions will have the same moral status as infanticide, except for the difference made by the physical connection between fetus and mother. A third way with abortion is thus a moderate and differential view, combining elements of the liberal view for early abortions with elements of (a weakened version of) the conservative view for late abortions. The policy that a moderate view will support is a moderate policy permissive in the early stages of pregnancy and more restrictive (though not as restrictive as conservatives think appropriate) in the later stages. So far as the personal question of the moral evaluation of particular abortions is concerned, there is no pressing need to resolve the borderline cases around the threshold stage. But a workable abortion policy cannot tolerate this vagueness and will need to establish a definite time limit beyond which the stipulated grounds will come into play. Although the precise location of the time limit will unavoidably be somewhat arbitrary it will be defensible as long as it falls somewhere within the threshold stage. Abortion on request up to the time limit and only for cause thereafter: these are the elements of a satisfactory abortion policy. E. . .1 2. A Criterion of Moral Standing

We are assuming that for a creature to have moral standing is for it to have a right to life. Any such right

5 Reproductive Ethics

imposes duties on moral agents; these duties may be either negative (not to deprive the creature of life) or positive (to support the creature's life). Possession of a right to life implies at least some immunity against attack by others and possibly also some entitlement to the aid of others. As the duties may vary in strength, so may the corresponding rights. To have some moral standing is to have some right to life, whether or not it may be overridden by the rights of others. To have full moral standing is to have the strongest right to life possessed by anyone, the right to life of the paradigm person. Depending on one's moral theory, this right may or may not be inviolable and indefeasible and thus may or may not impose absolute duties on others. To which creatures should we distribute (some degree of) moral standing? On which criterion should we base this distribution? It may be easier to answer these questions if we begin with the clear case and work outward to the unclear ones. If we can determine why we ascribe full standing to the paradigm case, we may learn what to look for in other creatures when deciding whether or not to include them in the moral sphere. The paradigm bearer of moral standing is an adult human being with normal capacities of intellect, emotion, perception, sensation, decision, action, and the like. If we think of such a person as a complex bundle of natural properties, then in principle we could employ as a criterion any of the properties common to all normal and mature members of our species. Selecting a particular property or set of properties will define a class of creatures with moral standing, namely, all (and only) those who share that property. The extension of that class will depend on how widely the property in question is distributed. Some putative criteria will be obviously frivolous and will immediately fail the tests of generality or moral relevance. But even after excluding the silly candidates, we are left with a number of serious ones. There are four that appear to be the most serious: we might attribute full moral standing to the paradigm person on the ground that he/she is (a) intrinsically valuable, (b) alive, (c) sentient, or (d) rational. An intuitive test of the adequacy of any of these candidates will involve first enumerating the class of beings to whom it will distribute moral standing and then determining whether that class either excludes creatures that upon careful reflection we believe ought to be included or includes creatures that we believe ought to be excluded. In the former case the criterion draws the boundary of the moral sphere too narrowly and fails as a necessary condition of moral standing. In the latter case the criterion draws the boundary too broadly and fails as a sufficient condition. (A given criterion may, of course, be defective in both respects.) E. . .1

The first candidate for a criterion takes a direction rather different from that of the remaining three. It is a commonplace in moral philosophy to attribute to (normal adult) human beings a special worth or value or dignity in virtue of which they possess (among other rights) a full right to life. This position implies that (some degree of) moral standing extends just as far as (some degree of) this intrinsic value, a higher degree of the latter entailing a higher degree of the former. E. . .1 Such a criterion is undeniably attractive in certain respects[.] [. . .1 But it is clear that it cannot serve our present purpose. A criterion must connect moral standing with some property of things whose presence or absence can be confirmed by a settled, objective, and public method of investigation. The property of being intrinsically valuable is not subject to such verification. A criterion based on intrinsic value cannot be applied without a theory of intrinsic value. Such a theory will supply a criterion of intrinsic value by specifying the natural properties of things in virtue of which they possess such value. But if things have moral standing in virtue of having intrinsic value, and if they have intrinsic value in virtue of having some natural property, then it is that natural property which is serving as the real criterion of moral standing, and the middle term of intrinsic value is eliminable without loss. A theory of intrinsic value may thus entail a criterion of moral standing, but intrinsic value cannot itself serve as that criterion. E. . .1 A criterion of life (or teleology) is too weak, admitting classes of beings (animate and inanimate) who are not suitable loci for moral rights; being alive is necessary for having standing, but it is not sufficient. A criterion of rationality (or moral agency) is too strong, excluding classes of beings (human and nonhuman) who are suitable loci for rights; being rational is sufficient for having standing, but it is not necessary A criterion of sentience (or consciousness) is a promising middle path between these extremes. Sentience is the capacity for feeling or affect. In its most primitive form it is the ability to experience sensations of pleasure and pain and thus the ability to enjoy and suffer. Its more developed forms include wants, aims, and desires (and thus the ability to be satisfied and frustrated); attitudes, tastes, and values; and moods, emotions, sentiments, and passions. Consciousness is a necessary condition of sentience, for feelings are states of mind of which their owner is aware. But it is not sufficient; it is at least possible in principle for beings to be conscious (percipient, for instance, or even rational) while utterly lacking feelings. If rationality embraces a set of cognitive capacities, then sentience is rooted in a beings affective and conative life. It is in virtue of being sentient

Sumner: A Third Way

that creatures have interests, which are compounded either out of their desires or out of the experiences they find agreeable (or both). If morality has to do with the protection and promotion of interests, it is a plausible conjecture that we owe moral duties to all those beings capable of having interests. But this will include all sentient creatures. Like rationality, and unlike life, it makes sense to think of sentience as admitting of degrees. Within any given mode, such as the perception of pain, one creature may be more or less sensitive than another. But there is a further sense in which more developed (more rational) creatures possess a higher degree of sentience. The expansion of consciousness and of intelligence opens up new ways of experiencing the world and therefore new ways of being affected by the world. More rational beings are capable of finding either fulfillment or frustration in activities and states of affairs to which less developed creatures are, both cognitively and affectively, blind. It is in this sense of a broader and deeper sensibility that a higher being is capable of a richer, fuller, and more varied existence. The fact that sentience admits of degrees (whether of sensitivity or sensibility) enables us to employ it both as an inclusion criterion and as a comparison criterion of moral standing. The animal kingdom presents us with a hierarchy of sentience. Nonsentient beings have no moral standing; among sentient beings the more developed have greater standing than the less developed, the upper limit being occupied by the paradigm of a normal adult human being. Although sentience is the criterion of moral standing, it is also possible to explain the relevance of rationality. The evolutionary order is one of ascending intelligence. Since rationality expands a creature's interests, it is a reliable indicator of the degree of moral standing which that creature possesses. Creatures less rational than human beings do not altogether lack standing, but they do lack full standing [. . .] A criterion of sentience can thus allow for the gradual emergence of moral standing in the order of nature. It can explain why no moral issues arise (directly) in our dealings with inanimate objects, plants, and the simpler forms of animal life. It can also function as a moral guideline in our encounters with novel life forms on other planets. If the creatures we meet have interests and are capable of enjoyment and suffering, we must grant them some moral standing. We thereby constrain ourselves not to exploit them ruthlessly for our own advantage. The kind of standing that they deserve may be determined by the range and depth of their sensibility, and in ordinary circumstances this will vary with their intelligence. We should therefore recognize as equals beings who are as rational and sensitive as

ourselves. The criterion also implies that if we encounter creatures who are rational but nonsentient—who utterly lack affect and desire—nothing we can do will adversely affect such creatures (in morally relevant ways). We would be entitled, for instance, to treat them as a species of organic computer. The same obviously holds for forms of artificial intelligence; in deciding whether to extend moral standing to sophisticated machines, the question (as Bentham put it) is not whether they can reason but whether they can suffer. A criterion of sentience also requires gentle usage of the severely abnormal. Cognitive disabilities and disorders may impair a person's range of sensibility, but they do not generally reduce that person to the level of a nonsentient being. Even the grossly retarded or deranged will still be capable of some forms of enjoyment and suffering and thus will still possess (some) moral standing in their own right. This standing diminishes to the vanishing point only when sentience is entirely lost or never gained in the first place. If all affect and responsivity are absent, and if they cannot be engendered, then (but only then) are we no longer dealing with a sentient creature. This verdict accords well with the contemporary trend toward defining death in terms of the permanent loss of cerebral functioning. Although such patients are in one obvious sense still alive (their blood circulates and is oxygenated), in the morally relevant sense they are now beyond our reach, for we can cause them neither good nor ill. A criterion of life would require us to continue treating them as beings with (full?) moral standing, whereas a criterion of rationality would withdraw that standing when reason was lost even though sensibility should remain. Again a criterion of sentience enables us to find a middle way. Fastening upon sentience as the criterion for possession of a right to life thus opens up the possibility of a reasonable and moderate treatment of moral problems other than abortion, problems pertaining to the treatment of nonhuman animals, extraterrestrial life, artificial intelligence, "defective" human beings, and persons at the end of life. We need now to trace out its implications for the fetus. 3. The Morality of Abortion The adoption of sentience as a criterion determines the location of a threshold of moral standing. Since sentience admits of degrees, we can in principle construct a continuum ranging from fully sentient creatures at one extreme to completely nonsentient creatures at the other. The threshold of moral standing is that area of the continuum through which sentience fades into nonsentience. In phylogenesis the continuum extends

II

266

'..111

1111:13

5 Reproductive Ethics

from homo sapiens to the simple animals and plants, and the threshold area is the boundary between vertebrates and invertebrates. In pathology the continuum extends from the fully normal to the totally incapacitated, and the threshold area is the transition from consciousness to unconsciousness. Human ontogenesis also presents us with a continuum from adult to zygote. The threshold area will be the stage at which sentience first emerges, but where is that to be located? [. . .1 The information we now possess does not enable us to date with accuracy the emergence of fetal sentience. Of some judgments, however, we can be reasonably confident. First-trimester fetuses are clearly not yet sentient. Third-trimester fetuses probably possess some degree of sentience, however minimal. The threshold of sentience thus appears to fall in the second trimester. More ancient and primitive than cognition, the ability to discriminate simple sensations of pleasure and pain is probably the first form of consciousness to appear in the ontogenetic order. Further, when sentience emerges it does not do so suddenly. The best we can hope for is to locate a threshold stage or period in the second trimester. It is at present unclear just how far into that trimester this stage occurs. The phylogenetic and pathological continua yield us clear cases at the extremes and unclear cases in the middle. The ontogenetic continuum does the same. Because there is no quantum leap into consciousness during fetal development, there is no dean and sharp boundary between sentient and nonsentient fetuses. There is therefore no precise point at which a fetus acquires moral standing. More and better information may enable us to locate the threshold stage ever more accurately, but it will never collapse that stage into a point. We are therefore inevitably confronted with a class of fetuses around the threshold stage whose sentience, and therefore whose moral status, is indeterminate. [. . .1 Our common moral consciousness locates contraception and infanticide in quite different moral categories. This fact suggests implicit recognition of a basic asymmetry between choosing not to create a new life in the first place and choosing to destroy a new life once it has been created. The boundary between the two kinds of act is the threshold at which that life gains moral protection. Since gametes lack moral standing, contraception (however it is carried out) merely prevents the creation of a new person. Since an infant has moral standing, infanticide (however it is carried out) destroys a new person. A second-trimester threshold of moral standing introduces this asymmetry into the moral assessment of abortion. We may define an early abortion as one performed sometime during the first trimester or early in the second and a late abortion as

one performed sometime late in the second trimester or during the third. An early abortion belongs in the same moral category as contraception: it prevents the emergence of a new being with moral standing. A late abortion belongs in the same moral category as infanticide: it terminates the life of a new being with moral standing. The threshold of sentience thus extends the morality of contraception forward to cover early abortion and extends the morality of infanticide backward to cover late abortion. [. . .1 Settling on sentience as a criterion of moral standing thus leads us to a view of the moral status of the fetus, and of the morality of abortion, which satisfies the constraints set out in Section 1. It is gradual, since it locates a threshold stage rather than a point and allows moral standing to be acquired incrementally. It is differential, since it locates the threshold stage during gestation and thus distinguishes the moral status of newly conceived and full-term fetuses. It is developmental, since it grounds the acquisition of moral standing in one aspect of the normal development of the fetus. And it is moderate, since it distinguishes the moral status of early and late abortions and applies each of the established views to that range of cases for which it is appropriate.

An Abortion Policy A differential view of the morality of abortion leads to a differential abortion policy—one that draws a legal distinction between early and late abortions. If we work within the framework of a liberal social theory, then it is understood that the state has no right to interfere in the private activities of individuals. An early abortion is a private act—or, rather, a private transaction between a woman and her physician. [. . .1 Grounds for abortion become pertinent only when we turn to late abortions. Since virtually all such abortions will result in the death of a being that has a right to life (though not all will violate that right), the state has a legitimate role to play in governing trade in abortion at this stage. Legal grounds for late abortion are a special case of conditions for justifiable homicide. As much as possible (allowing for the unique relation between mother and fetus) these grounds should authorize abortion when killing would also be justified in relevantly similar cases not involving fetuses. Two general conditions for justifiable homicide will be applicable to abortions: self-defense and euthanasia. The usual legal grounds for abortion provided by moderate policies may be divided into four categories: (a) therapeutic (threat to maternal life or health); (b) eugenic (risk of fetal abnormality); (c) humanitarian

Sumner: AThird Way

(pregnancy due to the commission of a crime, such as rape or incest); (d) socioeconomic (poverty, family size, etc.). If a moderate treatment of late abortion is coupled (as it should be) with a permissive treatment of early ones, only the first two categories are necessary. Therapeutic grounds for abortion follow from a woman's right of self-defense. The threat, however, must be serious in two different respects: the injury in prospect must be more than trivial, and the probability of its occurrence must be greater than normal. The risks generally associated with pregnancy will not here suffice. Further, there must be good medical reason not to delay until the fetus has a better chance of survival, and every effort must be made to save the fetus' life if this is possible. Thus late abortion for therapeutic reasons ought to be reserved for genuine medical emergencies in which no other course of action would qualify as proper care of the mother. In many putatively moderate policies, therapeutic grounds for abortion (especially mental health clauses) are interpreted so liberally as to cover large numbers of cases that are not by any stretch of the imagination medical emergencies. This is the standard device whereby a policy moderate in principle becomes permissive in practice. Since the policy here advanced is permissive in principle (for early abortions), a strict interpretation of the therapeutic grounds for late abortions will be mandatory. The same strictures will apply to eugenic grounds. Where there is a substantial risk of some severe anomaly (rubella, spina bifida, Tay-Sachs disease, etc.), abortion may be the best course of action for the fetus. This is not obviously the case for less severe defects (Down's syndrome, dwarfism, etc.). Again, there will be no justification for an interpretation of eugenic grounds so elastic that it permits abortion whenever the child is unwanted (because, say, it is the "wrong" sex). A rough rule of thumb is that late abortion for reasons of fetal abnormality is permissible only in those cases in which euthanasia for defective newborns would also be permissible. Probability will play a different role in the two kinds of case, since prenatal diagnosis of these conditions is often less certain than postnatal. But against this reason for delay we must balance the anguish of a woman carrying a fetus who may turn out at birth to be grossly deformed. Since diagnostic techniques such as ultrasound and amniocentesis cannot be employed until the second trimester, a permissive treatment of early abortions will not eliminate the need for late abortions on eugenic grounds. Both therapeutic and eugenic grounds can be alleged for a wide range of abortions. Some of these cases will be clearly justified, others will be just as clearly unjustified, and the remainder will just be hard cases.

There is no formula that can be applied mechanically to decide the hard cases. We should look to a statute for only the most perfunctory statement of justifying grounds for abortion. Particular decisions (the analogue of case law) are best undertaken by persons with the relevant medical expertise. This might be a hospital or clinic committee established especially to monitor late abortions or an "ethics committee" with broader responsibilities. In either case, establishing the right sort of screening mechanism is the best means of ensuring that the justifying grounds are given a reasonable application. There is no need for any special notice of humanitarian grounds. It is doubtful indeed whether incest ought to be a crime, except in those cases in which someone is being exploited. In any case, any woman who has become pregnant due to incestuous intercourse will have ready access to an early abortion. If she declines this opportunity and if there is no evidence of genetic abnormality, she may not simply change her mind later. The same obviously applies to pregnancy due to rape, including statutory rape. The practical problems should be approached by providing suitable counseling. A permissive policy for early abortions will also render socioeconomic grounds redundant. Since social constraints do not normally create an emergency for which abortion is the only solution, and since women will be able to terminate pregnancies at will in the early stages, there is no need for separate recognition of social or economic justifications for abortion. An adequate abortion policy is thus a conjunction of a permissive policy for early abortions and a moderate policy for late abortions. The obvious remaining question is where to draw the boundary between the two classes of cases. When we are dealing with the morality of abortion, borderline fuzziness is both inevitable and tolerable. Many moral problems turn on factors that are matters of degree. Where such factors are present, we cannot avoid borderline cases whose status is unclear or indeterminate. It is a defect in a moral theory to draw sharp lines where there are none or to treat hard cases as though they were easy. But what makes for good morals may also make for bad law. An abortion policy must be enforceable and so must divide cases as clearly as possible. A threshold stage separating early from late abortions must here give way to a cutoff point. Since there is no threshold point in fetal development, any precise upper limit on the application of a permissive policy will be to some extent arbitrary. Clearly it must be located within the threshold period, thus sometime in the second trimester. Beyond this constraint the choice of a time limit may be made on pragmatic grounds.

I

II

II

Al

111111.1

5 Reproductive Ethics

If a permissive policy for early abortions is to promote their autonomy, women must have time to discover that they are pregnant and to decide on a course of action. This factor will tend to push the cutoff point toward the end of the second trimester. On the other hand, earlier abortions are substantially safer and more economical of scarce medical resources than later ones. This factor will tend to pull the cutoff point toward the beginning of the second trimester. Balancing these considerations would indine one toward a time limit located sometime around the midpoint of pregnancy. But it should not be pretended that there is a unique solution to this policy problem. Differential policies may legitimately vary (within constraints) in their choice of a boundary between permissiveness and moderation. E. . .1 The abortion policy here proposed is not novel: a differential policy with a time limit in the second trimester is already in operation in a number of countries.4 But these policies seem usually to have been settled on as compromises between the opposed demands of liberals and conservatives rather than as matters of principle. Such compromises are attractive to politicians, who do not seek any deeper justification for the policies they devise. But there is a deeper justification for this policy. Although it does define a middle ground between the established views, it has not been

defended here as the outcome of a bargaining procedure. Instead it has been advanced as the only policy congruent with an adequate criterion of moral standing and proper recognition of both a woman's right to autonomy and a fetus's right to life. A differential policy does not mediate between alternatives both of which are rationally defensible; instead it supersedes alternatives both of which have been discredited. There is, therefore, a third way with the abortion issue. Its superiority over the established views lies largely in its sensitivity to a factor which both of them are committed to ignoring the manifest differences between a fetus at the beginning and at the end of its prenatal existence. Views which deny the relevance of this factor deserve to command no more than minority support. Those who, for this reason, can embrace neither of the established views need feel no diffidence about seeking a middle ground between them. A moderate and differential view of abortion is capable of drawing the common-sense distinction between early and late abortions and of showing that such a distinction is neither shallow nor arbitrary. The view from the middle lacks of course the simplicity which has made it so easy to market its more extreme counterparts. But then why should we think that the moral problems raised by abortion are simple?

Notes 1.

2. 3.

I will not b e defending this assessment i n the present paper. For the arguments, see Abortion and Moral Theory, chs. 2 and 3. Abortion and Moral Theory, chs. 5 and 6. The adoption of this working definition of moral standing should not be construed as a concession that rights are the appropriate category for dealing with the moral issues posed by abortion. But since both of the established views employ the rhetoric of rights, there is some point

Abortion through a Feminist Ethics Lens Susan Sherwin

Abortion has long been a central issue in the arena of applied ethics, but the distinctive analysis of feminist ethics is generally overlooked in most philosophic

4.

to showing how that rhetoric is equally available to a moderate view. For a generalized notion of moral standing freed from all connection with rights, see Abortion and Moral Theory, Section 23. Notably the United States, Great Britain, France, Italy, Sweden, the Soviet Union, China, India, Japan, and most of the countries of Eastern Europe. The cutoff points in these jurisdictions vary from the beginning to the end of the second trimester.

discussions. Authors and readers commonly presume a familiarity with the feminist position and equate it with liberal defences of women's right to choose abortion, but, in fact, feminist ethics yields a different analysis of the moral questions surrounding abortion than that usually offered by the more familiar liberal defenders of abortion rights. Most feminists can agree with some of the conclusions that arise from certain non-feminist

Source: From Dialogue: Canadian Philosophical Review 30 (1987) (3), 265-84. Reprinted by permission of the author.

Sherwin: Abortion through a Feminist Ethics Lens 269

arguments on abortion, but they often disagree about the way the issues are formulated and the sorts of reasons that are invoked in the mainstream literature. Among the many differences found between feminist and non-feminist arguments about abortion, is the fact that most non-feminist discussions of abortion consider the questions of the moral or legal permissibility of abortion in isolation from other questions, ignoring (and thereby obscuring) relevant connections to other social practices that oppress women. They are generally grounded in masculinist conceptions of freedom (e.g., privacy, individual choice, individuals' property rights in their own bodies) that do not meet the needs, interests, and intuitions of many of the women concerned. In contrast, feminists seek to couch their arguments in moral concepts that support their general campaign of overcoming injustice in all its dimensions, including those inherent in moral theory itself.' There is even disagreement about how best to understand the moral question at issue: non-feminist arguments focus exclusively on the morality and/or legality of performing abortions, whereas feminists insist that other questions, including ones about accessibility and delivery of abortion services must also be addressed. Although feminists welcome the support of nonfeminists in pursuing policies that will grant women control over abortion decisions, they generally envision very different sorts of policies for this purpose than those considered by non-feminist sympathizers. For example, Kathleen McDonnell (1984) urges feminists to develop an explicitly "'feminist morality' of abortion. . . . At its root it would be characterized by the deep appreciations of the complexities of life, the refusal to polarize and adopt simplistic formulas" (52). Here, I propose one conception of the shape such an analysis should take.

Women and Abortion The most obvious difference between feminist and non-feminist approaches to abortion can be seen in the relative attention each gives to the interests and experiences of women in its analysis. Feminists consider it self-evident that the pregnant woman is a subject of principal concern in abortion decisions. In most non-feminist accounts, however, not only is she not perceived as central, she is rendered virtually invisible. Non-feminist theorists, whether they support or oppose women's right to choose abortion, focus almost all their attention on the moral status of the developing embryo or the fetus. In pursuing a distinctively feminist ethics, it is appropriate to begin with a look at the role of abortion

in women's lives. Clearly, the need for abortion can be very intense; women have pursued abortions under appalling and dangerous conditions, across widely diverse cultures and historical periods. No one denies that if abortion is not made legal, safe, and accessible, women will seek out illegal and life-threatening abortions to terminate pregnancies they cannot accept. Anti-abortion activists seem willing to accept this price, but feminists judge the inevitable loss of women's lives associated with restrictive abortion policies to be a matter of fundamental concern. Although anti-abortion campaigners imagine that women often make frivolous and irresponsible decisions about abortion, feminists recognize that women have abortions for a wide variety of reasons. Some women, for instance, find themselves seriously ill and incapacitated throughout pregnancy; they cannot continue in their jobs and may face enormous difficulties in fulfilling their responsibilities at home. Many employers and schools will not tolerate pregnancy in their employees or students, and not every woman is able to put her job, career, or studies on hold. Women of limited means may be unable to take adequate care of children they have already borne and they may know that another mouth to feed will reduce their ability to provide for their existing children. Women who suffer from chronic disease, or who feel too young, or too old, or who are unable to maintain lasting relationships may recognize that they will not be able to care properly for a child at this time. Some who are homeless, or addicted to drugs, or who are diagnosed as carrying the AIDS virus may be unwilling to allow a child to enter the world under such circumstances. If the pregnancy is a result of rape or incest, the psychological pain of carrying it to term may be unbearable, and the woman may recognize that her attitude to the child after birth will always be tinged with bitterness. Some women have learned that the fetuses they carry have serious chromosomal anomalies and consider it best to prevent them from being born with a condition bound to cause suffering. Others, knowing the fathers to be brutal and violent, may be unwilling to subject a child to the beatings or incestuous attacks they anticipate; some may have no other realistic way to remove the child (or themselves) from the relationship. Or a woman may simply believe that bearing a child is incompatible with her life plans at this time, since continuing a pregnancy is likely to have profound repercussions throughout a woman's entire life. If the woman is young, a pregnancy will very likely reduce her chances of education and hence limit her career and life opportunities "The earlier a woman has a baby, it seems, the more likely she is to drop out of school;

uu

I

II ,.M

1111141

5 Reproductive Ethics

the less education she gets, the more likely she is to remain poorly paid, peripheral to the labour market, or unemployed, and the more children she will have— between one and three more than her working childless counterpart" (Petchesky 1984, 150). In many circumstances, having a child will exacerbate the social and economic forces already stacked against her by virtue of her sex (and her race, class, age, sexual orientation, or the effects of some disability, etc.). Access to abortion is a necessary option for many women if they are to escape the oppressive conditions of poverty. Whatever the reason, most feminists believe that a pregnant woman is in the best position to judge whether abortion is the appropriate response to her circumstances. Since she is usually the only one able to weigh all the relevant factors, most feminists reject attempts to offer any general abstract rules for determining when abortion is morally justified. Women's personal deliberations about abortion include contextually defined considerations reflecting her commitment to the needs and interests of everyone concerned—including herself, the fetus she carries, other members of her household, etc. Because there is no single formula available for balancing these complex factors through all possible cases, it is vital that feminists insist on protecting each woman's right to come to her own conclusions. Abortion decisions are, by their very nature, dependent on specific features of each woman' experience; theoretically dispassionate philosophers and other moralists should not expect to set the agenda for these considerations in any universal way. Women must be acknowledged as full moral agents with the responsibility for making moral decisions about their own pregnancies.2 Although I think that it is possible for a woman to make a mistake in her moral judgment on this matter (i.e., it is possible that a woman may come to believe that she was wrong about her decision to continue or terminate a pregnancy), the intimate nature of this sort of decision makes it unlikely that anyone else is in a position to arrive at a more reliable conclusion; it is, therefore, improper to grant others the authority to interfere in women's decisions to seek abortions. Feminist analysis regards the effects of unwanted pregnancies on the lives of women individually and collectively as a central element in the moral evaluation of abortion. Even without patriarchy bearing a child would be a very important event in a woman's life. It involves significant physical, emotional, social, and (usually) economic changes for her. The ability to exert control over the incidence, timing, and frequency of childbearing is often tied to her ability to control most other things she values. Since we live in a patriarchal

society it is especially important to ensure that women have the authority to control their own reproduction.3 Despite the diversity of opinion among feminists on most other matters, virtually all feminists seem to agree that women must gain full control over their own reproductive lives if they are to free themselves from male dominance.4 Many perceive the commitment of the political right wing to opposing abortion as part of a general strategy to reassert patriarchal control over women in the face of significant feminist influence (Petchesky 1980, 112). Women's freedom to choose abortion is also linked with their ability to control their own sexuality. Women's subordinate status often prevents them from refusing men sexual access to their bodies. If women cannot end the unwanted pregnancies that result from male sexual dominance, their sexual vulnerability to particular men can increase, because caring for an(other) infant involves greater financial needs and reduced economic opportunities for women.' As a result, pregnancy often forces women to become dependent on men. Since a woman's dependence on a man is assumed to entail that she will remain sexually loyal to him, restriction of abortion serves to channel women's sexuality and further perpetuates the cycle of oppression. In contrast to most non-feminist accounts, feminist analyses of abortion direct attention to the question of how women get pregnant. Those who reject abortion seem to believe that women can avoid unwanted pregnancies by avoiding sexual intercourse. Such views show little appreciation for the power of sexual politics in a culture that oppresses women. Existing patterns of sexual dominance mean that women often have little control over their sexual lives. They may be subject to rape by strangers, or by their husbands, boyfriends, colleagues, employers, customers, fathers, brothers, uncles, and dates. Often, the sexual coercion is not even recognized as such by the participants, but is the price of continued "good we—popularity, economic survival, peace, or simple acceptance. Few women have not found themselves in circumstances where they do not feel free to refuse a man's demands for intercourse, either because he is holding a gun to her head or because he threatens to be emotionally hurt if she refuses (or both). Women are socialized to be compliant and accommodating, sensitive to the feelings of others, and frightened of physical power; men are socialized to take advantage of every opportunity to engage in sexual intercourse and to use sex to express dominance and power. Under such circumstances, it is difficult to argue that women could simply "choose" to avoid heterosexual activity if they wish to avoid pregnancy. Catherine MacKinnon neatly sums it up: "the

Sherwin: Abortion through a Feminist Ethics Lens

logic by which women are supposed to consent to sex [is]: preclude the alternatives, then call the remaining option 'her choice- (MacKinnon 1989, 192). Nor can women rely on birth control alone to avoid pregnancy. There simply is no form of reversible contraception available that is fully safe and reliable. The pill and the IUD are the most effective means offered, but both involve significant health hazards to women and are quite dangerous for some. No woman should spend the 30 to 40 years of her reproductive life on either form of birth control. Further, both have been associated with subsequent problems of involuntary infertility, so they are far from optimum for women who seek to control the timing of their pregnancies. The safest form of birth control involves the use of barrier methods (condoms or diaphragms) in combination with spermicidal foams or jelly. But these methods also pose difficulties for women. They may be socially awkward to use: young women are discouraged from preparing for sexual activity that might never happen and are offered instead romantic models of spontaneous passion. (Few films or novels interrupt scenes of seduction for the fetching of contraceptives.) Many women find their male partners unwilling to use barrier methods of contraception and they do not have the power to insist. Further, cost is a limiting factor for many women. Condoms and spermicides are expensive and are not covered under most health care plans. There is only one contraceptive option which offers women safe and fully effective birth control: barrier methods with the back-up option of abortion.6 From a feminist perspective, a central moral feature of pregnancy is that it takes place in women's bodies and has profound effects on women's lives. Gender-neutral accounts of pregnancy are not available; pregnancy is explicitly a condition associated with the female body.' Because the need for abortion is experienced only by women, policies about abortion affect women uniquely. Thus, it is important to consider how proposed policies on abortion fit into general patterns of oppression for women. Unlike non-feminist accounts, feminist ethics demands that the effects on the oppression of women be a principal consideration when evaluating abortion policies. The Fetus

In contrast, most non-feminist analysts believe that the moral acceptability of abortion turns on the question of the moral status of the fetus. Even those who support women's right to choose abortion tend to accept the central premise of the anti-abortion proponents that abortion can only be tolerated if it can be proved that

the fetus is lacking some criterion of full personhood.6 Opponents of abortion have structured the debate so that it is necessary to define the status of the fetus as either valued the same as other humans (and hence entitled not to be killed) or as lacking in all value. Rather than challenging the logic of this formulation, many defenders of abortion have concentrated on showing that the fetus is indeed without significant value (Tooley 1972, Warren 1973); others, such as Wayne Sumner (1981), offer a more subtle account that reflects the gradual development of fetuses whereby there is some specific criterion that determines the degree of protection to be afforded them which is lacking in the early stages of pregnancy but present in the later stages. Thus, the debate often rages between abortion opponents who describe the fetus as an "innocent," vulnerable, morally important, separate being whose life is threatened and who must be protected at all costs, and abortion supporters who try to establish some sort of deficiency inherent to fetuses which removes them from the scope of the moral community. The woman on whom the fetus depends for survival is considered as secondary (if she is considered at all) in these debates. The actual experiences and responsibilities of real women are not perceived as morally relevant (unless they, too, can be proved innocent by establishing that their pregnancies are a result of rape or incest). It is a common assumption of both defenders and opponents of women's right to choose abortion that many women will be irresponsible in their choices. The important question, though, is whether fetuses have the sort of status that justifies interfering in women's choices at all. In some contexts, women's role in gestation is literally reduced to that of "fetal containers"; the individual women disappear or are perceived simply as mechanical life-support systems.9 The current rhetoric against abortion stresses the fact that the genetic make-up of the fetus is determined at conception and the genetic code is incontestably human. Lest there be any doubt about the humanity of the fetus, we are assailed with photographs of fetuses at various stages of development demonstrating the early appearance of recognizably human characteristics, e.g., eyes, fingers, and toes. The fact that the fetus in its early stages is microscopic, virtually indistinguishable from other primate fetuses to the untrained eye, and lacking in the capacities that make human life meaningful and valuable is not deemed relevant by the self-appointed defenders of fetuses. The anti-abortion campaign is directed at evoking sympathetic attitudes toward this tiny, helpless being whose life is threatened by its own mother; it urges us to see the fetus as entangled in an adversarial relationship with the (presumably

II

j11

MILLI

5 Reproductive Ethics

irresponsible) woman who carries it. We are encouraged to identify with the "unborn child" and not with the (selfish) woman whose life is also at issue. Within the non-feminist literature, both defenders and opponents of women's right to choose abortion agree that the difference between a late-term fetus and a newborn infant is "merely geographical" and cannot be considered morally significant. But a fetus inhabits a woman body and is wholly dependent on her unique contribution to its maintenance while a newborn is physically separate though still in need of a lot of care. One can only view the distinction between being in or out of a woman's womb as morally irrelevant if one discounts the perspective of the pregnant woman; feminists seem to be alone in recognizing her perspective as morally important.10 Within anti-abortion arguments, fetuses are identified as individuals; in our culture which views the (abstract) individual as sacred, fetuses qua individuals should be honoured and preserved. Extraordinary claims are made to try to establish the individuality and moral agency of fetuses. At the same time, the women who carry these fetal individuals are viewed as passive hosts whose only significant role is to refrain from aborting or harming their fetuses. Since it is widely believed that the woman does not actually have to do anything to protect the life of the fetus, pregnancy is often considered (abstractly) to be a tolerable burden to protect the life of an individual so like us.n Medicine has played its part in supporting these sorts of attitudes. Fetal medicine is a rapidly expanding specialty, and it is commonplace in professional medical journals to find references to pregnant women as "fetal environments." Fetal surgeons now have at their disposal a repertory of sophisticated technology that can save the lives of dangerously ill fetuses; in light of such heroic successes, it is perhaps understandable that women have disappeared from their view. These specialists see fetuses as their patients, not the women who nurture them. Doctors perceive themselves as the active agents in saving fetal lives and, hence, believe that they are the ones in direct relationship with the fetuses they treat. Perhaps even more distressing than the tendency to ignore the woman's agency altogether and view her as a purely passive participant in the medically controlled events of pregnancy and childbirth is the growing practice of viewing women as genuine threats to the well-being of the fetus. Increasingly women are viewed as irresponsible or hostile toward their fetuses, and the relationship between them is characterized as adversarial (Overall 1987, 60). Concern for the well-being of the fetus is taken as licence for doctors to intervene

to ensure that women comply with medical "advice." Courts are called upon to enforce the doctors' orders when moral pressure alone proves inadequate, and women are being coerced into undergoing unwanted Caesarean deliveries and technologically monitored hospital births. Some states have begun to imprison women for endangering their fetuses through drug abuse and other socially unacceptable behaviours. An Australian state recently introduced a bill that makes women liable to criminal prosecution "if they are found to have smoked during pregnancy, eaten unhealthful foods, or taken any other action which can be shown to have adversely affected the development of the fetus" (Warren 1989, 60). In other words, physicians have joined with antiabortionist activists in fostering a cultural acceptance of the view that fetuses are distinct individuals, who are physically, ontologically, and socially separate from the women whose bodies they inhabit, and who have their own distinct interests. In this picture, pregnant women are either ignored altogether or are viewed as deficient in some crucial respect and hence subject to coercion for the sake of their fetuses. In the former case, the interests of the women concerned are assumed to be identical with those of the fetus; in the latter, the women's interests are irrelevant because they are perceived as immoral, unimportant, or unnatural. Focus on the fetus as an independent entity has led to presumptions which deny pregnant women their roles as active, independent, moral agents with a primary interest in what becomes of the fetuses they carry. Emphasis on the fetus's status has led to an assumed licence to interfere with women's reproductive freedom. A Feminist View of the Fetus

Because the public debate has been set up as a competition between the rights of women and those of fetuses, feminists have often felt pushed to reject claims of fetal value in order to protect women's claims. Yet, as Addelson (1987) has argued, viewing abortion in this way "tears [it] out of the context of women's lives" (107). There are other accounts of fetal value that are more plausible and less oppressive to women. On a feminist account, fetal development is examined in the context in which it occurs, within women's bodies rather than in the imagined isolation implicit in many theoretical accounts. Fetuses develop in specific pregnancies which occur in the lives of particular women. They are not individuals housed in generic female wombs, nor are they full persons at risk only because they are small and subject to the whims of women. Their very existence is

Sherwin: Abortion through a Feminist Ethics Lens

relational, developing as they do within particular women's bodies, and their principal relationship is to the women who carry them. On this view, fetuses are morally significant, but their status is relational rather than absolute. Unlike other human beings, fetuses do not have any independent existence; their existence is uniquely tied to the support of a specific other. Most non-feminist commentators have ignored the relational dimension of fetal development and have presumed that the moral status of fetuses could be resolved solely in terms of abstract metaphysical criteria of personhood. They imagine that there is some set of properties (such as genetic heritage, moral agency, self-consciousness, language use, or self-determination) which will entitle all who possess them to be granted the moral status of persons (Warren 1973, Tooley 1972). They seek some particular feature by which we can neatly divide the world into the dichotomy of moral persons (who are to be valued and protected) and others (who are not entitled to the same group privileges); it follows that it is a merely empirical question whether or not fetuses possess the relevant properties. But this vision misinterprets what is involved in personhood and what it is that is especially valued about persons. Personhood is a social category, not an isolated state. Persons are members of a community, they develop as concrete, discrete, and specific individuals. To be a morally significant category, personhood must involve personality as well as biological integrity.'2 It is not sufficient to consider persons simply as Kantian atoms of rationality; persons are all embodied, conscious beings with particular social histories. Annette Baier (1985) has developed a concept of persons as "second persons" which helps explain the sort of social dimension that seems fundamental to any moral notion of personhood: A person, perhaps, is best seen as one who was long enough dependent upon other persons to acquire the essential arts of personhood. Persons essentially are second persons, who grow up with other persons. . . . The fact that a person has a life history, and that a people collectively have a history depends upon the humbler fact that each person has a childhood in which a cultural heritage is transmitted, ready for adolescent rejection and adult discriminating selection and contribution. Persons come after and before other persons. (84-5; her emphasis.) Persons, in other words, are members of a social community which shapes and values them, and personhood

is a relational concept that must be defined in terms of interactions and relationships with others. A fetus is a unique sort of being in that it cannot form relationships freely with others, nor can others readily form relationships with it. A fetus has a primary and particularly intimate relationship with the woman in whose womb it develops; any other relationship it may have is indirect, and must be mediated through the pregnant woman. The relationship that exists between a woman and her fetus is clearly asymmetrical, since she is the only party to the relationship who is capable of making a decision about whether the interaction should continue and since the fetus is wholly dependent on the woman who sustains it while she is quite capable of surviving without it. However much some might prefer it to be otherwise, no one else can do anything to support or harm a fetus without doing something to the woman who nurtures it. Because of this inexorable biological reality, she bears a unique responsibility and privilege in determining her fetus's place in the social scheme of things. Clearly, many pregnancies occur to women who place very high value on the lives of the particular fetuses they carry, and choose to see their pregnancies through to term despite the possible risks and costs involved; hence, it would be wrong of anyone to force such a woman to terminate her pregnancy under these circumstances. Other women, or some of these same women at other times, value other things more highly (e.g., their freedom, their health, or previous responsibilities which conflict with those generated by the pregnancies), and choose not to continue their pregnancies. The value that women ascribe to individual fetuses varies dramatically from case to case, and may well change over the course of any particular pregnancy. There is no absolute value that attaches to fetuses apart from their relational status determined in the context of their particular development. Since human beings are fundamentally relational beings, it is important to remember that fetuses are characteristically limited in the relationships in which they can participate; within those relationships, they can make only the most restricted "contributions."" After birth, human beings are capable of a much wider range of roles in relationships with an infinite variety of partners; it is that very diversity of possibility and experience that leads us to focus on the abstraction of the individual as a constant through all her/his relationships. But until birth, no such variety is possible, and the fetus is defined as an entity within a woman who will almost certainly be principally responsible for it for many years to come.

II II

II

AI

5 Reproductive Ethics

No human, and especially no fetus, can exist apart from relationships; feminist views of what is valuable about persons must reflect the social nature of their existence. Fetal lives can neither be sustained nor destroyed without affecting the women who support them. Because of a fetus's unique physical status—within and dependent on a particular woman—the responsibility and privilege of determining its specific social status and value must rest with the woman carrying it. Fetuses are not persons because they have not developed sufficiently in social relationships to be persons in any morally significant sense (i.e., they are not yet second persons). Newborns, although just beginning their development into persons, are immediately subject to social relationships, for they are capable of communication and response in interaction with a variety of other persons. Thus, feminist accounts of abortion stress the importance of protecting women's right to continue as well as to terminate pregnancies as each sees fit. Feminist Politics and Abortion

Feminist ethics directs us to look at abortion in the context of other issues of power and not to limit discussion to the standard questions about its moral and legal acceptability. Because coerced pregnancy has repercussions for women's oppressed status generally, it is important to ensure that abortion not only be made legal but that adequate services be made accessible to all women who seek them. This means that within Canada, where medically approved abortion is technically recognized as legal (at least for the moment), we must protest the fact that it is not made available to many of the women who have the greatest need for abortions: vast geographical areas offer no abortion services at all, but unless the women of those regions can afford to travel to urban clinics, they have no meaningful right to abortion. Because women depend on access to abortion in their pursuit of social equality, it is a matter of moral as well as political responsibility that provincial health plans should cover the cost of transport and service in the abortion facilities women choose. Ethical study of abortion involves understanding and critiquing the economic, age, and social barriers that currently restrict access to medically acceptable abortion services.14 Moreover, it is also important that abortion services be provided in an atmosphere that fosters women's health and well-being; hence, the care offered should be in a context that is supportive of the choices women make. Abortions should be seen as part of women's overall reproductive health and could be included within centres that deal with all matters of reproductive health in an open, patient-centred manner where

effective counselling is offered for a wide range of reproductive decisions.15 Providers need to recognize that abortion is a legitimate option so that services will be delivered with respect and concern for the physical, psychological, and emotional effects on a patient. All too frequently, hospital-based abortions are provided by practitioners who are uneasy about their role and treat the women involved with hostility and resentment. Increasingly, many anti-abortion activists have personalized their attacks and focused their attention on harassing the women who enter and leave abortion clinics. Surely requiring a woman to pass a gauntlet of hostile protesters on her way to and from an abortion is not conducive to effective health care. Ethical exploration of abortion raises questions about how women are treated when they seek abortions;16 achieving legal permission for women to dispose of their fetuses if they are determined enough to manage the struggle should not be accepted as the sole moral consideration. Nonetheless, feminists must formulate their distinctive response to legislative initiatives on abortion. The tendency of Canadian politicians confronted by vocal activists on both sides of the abortion issue has been to seek "compromises" that seem to give something to each (and, thereby, also deprives each of important features sought in policy formation). Thus, the House of Commons recently passed a law (Bill C-43) that allows a woman to have an abortion only if a doctor certifies that her physical, mental, or emotional health will be otherwise threatened. Many non-feminist supporters of women's right to choose consider this a victory and urge feminists to be satisfied with it, but feminists have good reason to object. Besides their obvious objection to having abortion returned to the Criminal Code, feminists also object that this policy considers doctors and not women the best judges of a woman's need for abortion; feminists have little reason to trust doctors to appreciate the political dimension of abortion or to respond adequately to women's needs. Abortion must be a woman's decision, and not one controlled by her doctor. Further, experience shows that doctors are already reluctant to provide abortions to women; the opportunity this law presents for criminal persecution of doctors by anti-abortion campaigners is a sufficient worry to inhibit their participation.'' Feminists want women's decisionmaking to be recognized as legitimate, and cannot be satisfied with a law that makes abortion a medical choice. Feminists support abortion on demand because they know that women must have control over their reproduction. For the same reason, they actively oppose forced abortion and coerced sterilization, practices that are sometimes inflicted on the most powerless women, especially those in the Third World. Feminist ethics

Sherwin: Abortion through a Feminist Ethics Lens

demands that access to voluntary, safe, effective birth control be part of any abortion discussion, so that women have access to other means of avoiding pregnancy.18 Feminist analysis addresses the context as well as the practice of abortion decisions. Thus, feminists also object to the conditions which lead women to abort wanted fetuses because there are not adequate financial and social supports available to care for a child. Because feminist accounts value fetuses that are wanted by the women who carry them, they oppose practices which force women to abort because of poverty or intimidation. Yet, the sorts of social changes necessary if we are to free women from having abortions out of economic necessity are vast; they include changes not only in legal and health care policy, but also in housing, child care, employment, etc. (Petchesky 1980, 112). Nonetheless, feminist ethics defines reproductive freedom as the condition under which women are able to make truly voluntary choices about their reproductive lives, and these many dimensions are implicit in the ideal. Clearly, feminists are not "pro-abortion," for they are concerned to ensure the safety of each pregnancy to the greatest degree possible; wanted fetuses should not be harmed or lost. Therefore, adequate pre- and postnatal care and nutrition are also important elements of any feminist position on reproductive freedom. Where anti-abortionists direct their energies to trying to prevent women from obtaining abortions, feminists seek to protect the health of wanted fetuses. They recognize that far more could be done to protect and care for fetuses if the state directed its resources at supporting women who continue their pregnancies, rather than draining away resources in order to police women who find that they must interrupt their pregnancies. Caring for the women

who carry fetuses is not only a more legitimate policy than is regulating them; it is probably also more effective at ensuring the health and well-being of more fetuses. Feminist ethics also explores how abortion policies fit within the politics of sexual domination. Most feminists are sensitive to the fact that many men support womens right to abortion out of the belief that women will be more willing sexual partners if they believe that they can readily terminate an unwanted pregnancy. Some men coerce their partners into obtaining abortions the women may not wanC9 Feminists understand that many women oppose abortion for this very reason, being unwilling to support a practice that increases women's sexual vulnerability (Luker 1984, 209-15). Thus, it is important that feminists develop a coherent analysis of reproductive freedom that includes sexual freedom (as women choose to define it). That requires an analysis of sexual freedom that includes women's right to refuse sex; such a right can only be assured if women have equal power to men and are not subject to domination by virtue of their sex. In sum, then, feminist ethics demands that moral discussions of abortion be more broadly defined than they have been in most philosophic discussions. Only by reflecting on the meaning of ethical pronouncements on actual women's lives and the connections between judgments on abortion and the conditions of domination and subordination can we come to an adequate understanding of the moral status of abortion in our society. As Rosalind Petchesky (1980) argues, feminist discussion of abortion "must be moved beyond the framework of a "woman's right to choose" and connected to a much broader revolutionary movement that addresses all of the conditions of women's liberation" (113).

Notes 1.

2.

For some idea of the ways in which traditional moral theory oppresses women, see Morgan (1987) and Hoagland (1988). Critics continue to want to structure the debate around the possibility of women making frivolous abortion decisions and hence want feminists to agree to setting boundaries on acceptable grounds for choosing abortion. Feminists ought to resist this injunction, though. There is no practical way of drawing a line fairly in the abstract; cases that may appear "frivolous" at a distance often turn out to be substantive when the details are revealed, i.e., frivolity is in the eyes of the beholder. There is no evidence to suggest that women actually make the sorts of choices worried critics hypothesize about: e.g., a woman eight months pregnant who chooses to abort because she wants to take a trip or gets in "a tiff' with her partner. These sorts of fantasies, on which demands to

3.

4.

distinguish between legitimate and illegitimate personal reasons for choosing abortion chiefly rest, reflect an offensive conception of women as irresponsible; they ought not to be perpetuated. Women, seeking moral guidance in their own deliberations about choosing abortion, do not find such hypothetical discussions of much use. In her monumental historical analysis of the early roots of Western patriarchy, Gerda Lerner (1986) determined that patriarchy began in the period from 3100 to 600 BC when men appropriated women's sexual and reproductive capacity; the earliest states entrenched patriarchy by institutionalizing the sexual and procreative subordination of women to men. There are some women who claim to be feminists against choice in abortion. See, for instance, Callahan (1987), though few spell out their full feminist program. For

II

Ai

111611

5 Reproductive Ethics

5.

6.

7.

8. 9.

10. 11.

12. 13.

reasons I develop in this paper, I do not think this is a consistent position. There is a lot the state could do to ameliorate this condition. If it provided women with adequate financial support, removed the inequities in the labour market, and provided affordable and reliable childcare, pregnancy need not so often lead to a woman's dependence on a particular man. The fact that it does not do so is evidence of the state§ complicity in maintaining women's subordinate position with respect to men. See Petchesky (1984), especially Chapter 5, "Considering the Alternatives: The Problems of Contraception," where she documents the risks and discomforts associated with pill use and nips and the increasing rate at which women are choosing the option of diaphragm or condom with the option of early legal abortions as backup. See Zillah Eisenstein (1988) for a comprehensive theory of the role of the pregnant body as the central element in the cultural subordination of women. Thomson (1971) is a notable exception to this trend. This seems reminiscent of Aristotle view of women as "flower pots" where men implant the seed with all the important genetic information and the movement necessary for development and women§ job is that of passive gestation, like the flower pot. For exploration of the flower pot picture of pregnancy, see Whitbeck (1973) and Lange (1983). Contrast Warren (1989) with Tooley (1972). The definition of pregnancy as a purely passive activity reaches its ghoulish conclusion in the increasing acceptability of sustaining brain-dead women on life support systems to continue their functions as incubators until the fetus can be safely delivered. For a discussion of this new trend, see Murphy (1989). This apt phrasing is taken from Petchesky (1986), 342. Fetuses are almost wholly individuated by the women who bear them. The fetal "contributions" to

14.

15. 16.

17.

18.

19.

the relationship are defined by the projections and interpretations of the pregnant woman in the latter stages of pregnancy if she chooses to perceive fetal movements in purposeful ways (e.g., "it likes classical music, wine, exercise"). Some feminists suggest we seek recognition of the legitimacy of non-medical abortion services. This would reduce costs and increase access dramatically, with no apparent increase in risk, provided that services were offered by trained, responsible practitioners concerned with the well-being of their clients. It would also allow the possibility of increasing women's control over abortion. See, for example McDonnell (1984) 8. For a useful model of such a centre, see Van Wagner and Lee (1989). See CARA1/Halifax (1990) for women's stories about their experiences with hospitals and freestanding abortion clinics. The Canadian Medical Association has confirmed those fears. In testimony before the House of Commons committee reviewing the bill, the CMA reported that over half the doctors surveyed who now perform abortions expect to stop offering them if the legislation goes through. Since the Commons passed the bill, the threats of withdrawal of service have increased. Many doctors plan to abandon their abortion service once the law is introduced, because they are unwilling to accept the harassment they anticipate from anti-abortion zealots. Even those who believe that they will eventually win any court case that arises fear the expense and anxiety involved as the case plays itself out. Therefore, the Soviet model, where women have access to multiple abortions but where there is no other birth control available, must also be opposed. See CARAL/Halifax (1990), 20-1, for examples of this sort of abuse.

References Callahan, Sidney. 1987. "A Pro-life Feminist Makes Her Case," in Utne Reader (March/April): 104-14. CARAL/Halifax. 1990. Telling Our Stories: Abortion Stories from Nova Scotia. CARAL/Halifax (Canadian Abortion Rights Action League): Halifax. Eisenstein, Zillah R. 1988. The Female Body and the Law. University of California Press: Berkeley. Hoagland, Sara Lucia. 1988. Lesbian Ethics: Toward New Value. Institute of Lesbian Studies: Palo Alto, CA. Lange, Lynda. 1983. "Woman is Not a Rational Animal: On Aristotle's Biology of Reproduction," in Discovering Reality: Feminist Perspectives on Epistemology, Metaphysics, Methodology, and Philosophy of Science, ed. Sandra Harding and Merrill B. Hintikka. D. Reidel: Dordrecht, Holland. Lerner, Gerda. 1986. The Creation of Patriarchy. Oxford: New York. Luker, Kristin. 1984. Abortion and the Politics of Motherhood. University of California Press: Berkeley.

McDonnell, Kathleen. 1984. Not an Easy Choice: A Feminist Re-examines Abortion. The Women's Press: Toronto. MacKinnon, Catherine. 1989. Toward a Feminism Theory of the State. Harvard University Press: Cambridge, MA. Morgan, Kathryn Pauly. 1987. "Women and Moral Madness," in Science, Morality and Feminist Theory, ed. Marsha Hanen and Kai Nielsen, eds. Canadian Journal of Philosophy, Supplementary Volume 13: 201-26. Petchesky, Rosalind Pollack. 1980. "Reproductive Freedom: Beyond 'A Woman's Right to Choose'," in Women: Sex and Sexuality, ed. Catharine R Stimpson and Ethel Spector Person. University of Chicago Press: Chicago. . 1984. Abortion and Woman's Choice: The State, Sexuality, and Reproductive Freedom. Northeastern University Press: Boston. Thomson, Judith Jarvis. 1971. "A Defense of Abortion," in Philosophy and Public Affairs 1: 47-66.

Sherwin: Abortion through a Feminist Ethics Lens

Tooley, Michael. 1972. "Abortion and Infanticide," in Philosophy and Public Affairs 2 (1; Fall): 37-65. Van Wagner, Vicki, and Lee, Bob. 1989. "Principles into Practice: An Activist Vision of Feminist Reproductive Health Care," in The Future of Human Reproduction, ed. Christine Overall. The Women's Press: Toronto.

Warren, Mary Amine. 1973. "On the Moral and Legal Status of Abortion," in The Monist 57: 43-61. . 1989. "The Moral Significance of Birth," in Hypatia 4 (2; Summer): 46-65. Whitbeck, Carolyn. 1973. "Theories of Sex Difference," in The Philosophic Forum, 5 (1-2; Fall/Winter 1973-4): 54-80.

5.4 Completing Reproduction

Judicial Intervention in Pregnancy and Birth Royal Commission on New Reproductive Technologies [...] Judicial Intervention Defined Our mandate directed us to examine "judicial interventions during gestation and birth." This examination involved considering how legislation and court decisions are or may be used to control a pregnant woman's behaviour. Such judicial intervention usually occurs when a woman is believed to be endangering the fetus she is carrying by refusing medical treatment believed necessary for fetal health, by abusing drugs or alcohol, or by engaging in behaviour such as prostitution. The matter is typically brought before the courts by a children's aid society, a health care facility, or, in some cases, a physician. Judicial interventions during gestation and birth can take several forms. Canadian courts, for example, have ordered pregnant women to refrain from specified behaviours and to undergo certain medical procedures considered necessary for the health of the fetus. [. . .1 US courts have issued similar directives, ordering women to engage or not to engage in certain behaviours during pregnancy, to undergo Caesarian section and other medical treatment, and to be incarcerated until they had given birth. Few cases have reached the courts in Canada, because the women most likely to encounter this situation are often in no position to resist and therefore they comply with the wishes of a physician or child welfare authority. An examination of the cases that have been reported shows that the women most likely to be subjected to judicial intervention are disproportionately poor, Aboriginal, or members of a racial or ethnic minority—all factors that influence their capacity to

resist intervention. Whether overt discrimination is at work or whether the life circumstances of these women are such that their behaviour during pregnancy is more likely to come under scrutiny is difficult to disentangle. Judicial intervention is an issue for all women in Canada, however, regardless of socioeconomic status, because its implications go beyond the consequences for an individual woman; it is an issue for women more generally if becoming pregnant means that they waive the constitutional protections afforded other citizens. [. . .1 The Ethical View The Commission's position on judicial intervention in pregnancy and birth [. . .1 relies primarily on our ethical stance and guiding principles, which we have applied throughout our report and in our recommendations. Although many legal and constitutional arguments can be made, our conclusions with respect to judicial intervention rest largely on our ethical reasoning. In our view, it is ethically (as well as legally) wrong to suggest that pregnant women's rights to make decisions about their medical care and treatment should be changed or lessened because they are pregnant. [. . .1 Consistent with the ethic of care—which is concerned with preventing conflicts instead of trying to resolve them after they arise—we begin by asking questions about how to ensure the best possible prenatal health and the maximum degree of well-being for both the pregnant woman and the fetus. Regardless of whether a fetus is a "person" with "rights," it is clear that the interests of the fetus are worthy of protection: what transpires before birth—the behaviour of the woman during pregnancy, the provision of medical treatment to her and to the fetus—can seriously affect the health and well-being of the child that is eventually born. Society

Source From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 951,953, 956-65. Reprinted with permission.

II AI

IIILLI

5 Reproductive Ethics

therefore has an interest in promoting the prenatal health and well-being of the fetus and of the woman carrying it. From the woman's perspective, however, considering the interests of her fetus separately from her own has the potential to create adversary situations with negative consequences for her autonomy and bodily integrity, for her relationship with her partner, and for her relationship with her physician. Judicial intervention is bound to precipitate crisis and conflict, instead of preventing them through support and care. It also ignores the basic components of women's fundamental human rights—the right to bodily integrity and the right to equality, privacy, and dignity. Importantly, as we will see, such measures are also unlikely to be effective and may not in fact protect the fetus. If we impose a legal obligation upon a woman to care for her fetus—even if it were possible to legislate a caring and nurturing relationship—the potential for curtailing women's choices and behaviour becomes staggering. The kinds of substances and activities that could pose a danger to the fetus are many, varied, and increasing: cigarettes, alcohol, drugs (both legal and illegal), environmental pollutants, strenuous exercise, saunas, and inadequate nutrition. As scientific knowledge develops, the list is becoming longer. Many women's management of pregnancy could be subject to challenge and scrutiny, and pregnancy could become the source of potential liability suits against women who failed to comply with certain standards of behaviour. In some cases, fearing a less-than-perfect outcome, a pregnant woman might feel compelled to seek abortion instead of care. Moreover, the threat of judicial intervention could have significant negative effects on fetal and maternal health. If women knew that they could be confined against their will, forced to submit to medical treatment, or charged with criminal offences, they might well avoid seeking medical care. Unfortunately, those who might avoid seeking care would likely be those who need it most—for example, women who are dependent on drugs or alcohol. As a result, health problems would escape detection and treatment—precisely the opposite effect sought by those who would use judicial means to intervene. E. . .1 The resort to judicial intervention also has serious implications for the relationship between a pregnant woman and her physician. If the physician is perceived to be potentially coercive instead of a caregiver, the woman might begin to withhold information or stop seeking prenatal care, with detrimental consequences for her health and that of the fetus. These dangers are recognized by many professional associations of physicians. Moreover, experience with judicial intervention has shown the uncertainties inherent in diagnosis and treatment; in several cases of judicial intervention, the medical treatment

deemed essential by the courts later turned out not to have been necessary. For example, a woman who went into hiding in defiance of a court order to undergo a Caesarian section later gave birth vaginally to a healthy child. Thus, medical and judicial judgements—even those made with the best of intentions can be mistaken. In addition, the very limited time frame within which most such decisions must be made and acted upon makes the process of judicial intervention unlikely to lead to fully considered, principled, or constitutional conclusions. Finally, judicial intervention both emerges from and reinforces a social perception of the role of women in reproduction that instrumentalizes them and devalues their humanity and individuality. At the core of the impulse toward judicial intervention in pregnancy and birth is the view that pregnant women are the means to an end—the birth of healthy children. To the extent that judicial intervention reinforces the notion that a pregnant woman's role is only to carry and deliver a healthy child, it denies her existence as an autonomous individual with legal and constitutional rights and is dangerous to the rights and autonomy of all women. In summary, judicial intervention offers no satisfactory answer to ensuring the well-being of the fetus: it precipitates crisis and conflict, it ignores women's fundamental constitutional and human rights, it contributes to an instrumentalized view of their role in reproduction—with adverse consequences for women individually and as a group—and, most important, it is not effective in achieving its goal of protecting fetal well-being. Society cannot care for a fetus, in the absence of the pregnant woman's cooperation, without taking control of the woman herself. The physical relationship between the fetus and the pregnant woman and the dependency of the fetus on the pregnant woman for sustenance make this impossible. By forcing medical intervention, society would be requiring pregnant women to do something that is asked of no other individual: to undergo medical treatment for the benefit of another. Even a living child has no right to force a parent to undergo medical procedures for the child's benefit, however morally compelling the case might be. This infringement of bodily autonomy and physical integrity is not justified on any grounds. This imperative will not change even as research pushes the boundaries of what can be done to treat a fetus in utero. For example, if and when surgery on a fetus moves beyond the research stage for a range of conditions, there may be increasing pressure on pregnant women to consent to such procedures. Whether these techniques remain experimental or move into the realm of accepted practice, they must be offered only in the context of the ethical and legal considerations set out in this chapter—that is, in the context of the

Royal Commission on New ReproductiveTechnologies: Judicial Intervention in Pregnancy and Birth

pregnant woman's autonomy and with her informed consent, based on full knowledge of the nature and risks of the proposed treatment. An Approach Based on Support and Care

If we reject judicial intervention in pregnancy and birth on moral, practical, and legal grounds, we must return to the question of how to ensure the health and well-being of the fetus and the pregnant woman. How should society respond to a situation where a woman is not caring for her fetus or engaging in behaviour that may harm it? In the Commission's view, the answer lies in examining the reasons for that behaviour and seeking solutions that address them. Some of the situations that give rise to attempts at judicial intervention are among the most difficult and tragic imaginable. The potential for harm is evident; the dangers to a fetus of alcohol abuse, drug addiction, or sexually transmitted diseases are real and potentially devastating. These situations are all the more distressing because the caring and nurturing assumed to be inherent in the relationship between woman and fetus appear to be absent. Although many cases involving refusal to follow medical advice or to accept surgical or other medical treatment have involved drug or alcohol abuse, a woman's reasons for choosing a particular course of action may include her socioeconomic circumstances, her educational level, her religious convictions, her cultural beliefs, her fears, or other deeply held values or personal beliefs. Of relevance in this regard is the fact that most of the women who have been subject to judicial intervention to date have been Aboriginal women and women of colour. Whatever the circumstances, judicial intervention does not provide a solution, because it does nothing to address the circumstances that bring about attempts to intervene or to create the social conditions and support that help to ensure a successful pregnancy and healthy outcome for both the woman and the child. In reaching this conclusion, Commissioners are acutely aware of the tragic nature of some of the situations that give rise to efforts to intervene in a pregnancy. As members of the helping professions, physicians and child welfare workers face situations that call on their basic human instinct to help where possible—an impulse that is rightly very difficult to resist because it is so fundamental to who they are and the job they do. The decision to respect a woman's autonomy and physical integrity and not to intervene must surely be one of the most difficult decisions any human being would ever be called upon to make. That is why the Commissioners' decision in this matter was reached through long and

careful deliberation and consideration of the issues from all sides. We made this decision not because harm to a fetus is acceptable or even tolerable, but because the dangers posed by judicial intervention far outweigh any benefits that a given individual intervention might yield. In line with the ethic of care, we believe that the best approach is to seek ways to ensure that the needs of both the woman and the fetus are met—in other words, to prevent a situation developing in which child welfare, medical, or other authorities might consider judicial intervention appropriate or necessary. The ethic of care offers a means of avoiding the conflicts inherent in judicial intervention by promoting two fundamental values: respect for the rights and autonomy of the pregnant woman and concern for the health and well-being of the fetus. The best way to accomplish this is not by compelling pregnant women to behave in certain ways, but by providing a supportive and caring environment in which they can make informed decisions and choose from among realistic options before and during pregnancy. The situations that lead to judicial intervention are inherently distressing because of the commitment we, as a society and as individuals, have to respect human life and dignity—the life and dignity of the pregnant woman, expressed through her autonomy, and that of the fetus, as a potential person. Judicial intervention sacrifices the human dignity and rights of one for the potential wellbeing of the other. Taking the alternative route of care and assistance means that the human life and dignity of both woman and fetus are respected—and it may even accomplish what legislation or court decisions cannot: establish a caring and nurturing relationship. Clearly, the vast majority of women will act in a way they believe to be the best interests of their fetus. The best way to promote prenatal health is therefore to provide the information and support necessary to enable pregnant women to make healthy choices for the well-being of themselves and their fetuses and informing them—in non-coercive, non-judgemental ways—about the implications of their decisions. This includes providing safe and accessible contraception and abortion services; offering accessible and culturally appropriate prenatal care and social services to pregnant women; counselling pregnant women about healthy lifestyles and ensuring they have the means to make these choices, including financial assistance where necessary, and providing information, outreach, and supports in the forms pregnant women need to make informed choices and realistic decisions about care and treatment, particularly for addictions. The Canadian Bar Association pointed out, in its brief to the Commission, that recourse to judicial intervention should be seen as a failure—a failure to provide policies and programs that sustain a woman's

5 Reproductive Ethics

right to manage her pregnancy and to support her decisions with appropriate services and resources in the community. By itself, prohibiting judicial intervention does not fulfil our responsibility as a society to promote the health and well-being of pregnant women. Meeting this responsibility also requires appropriate programs, services, and outreach designed specifically to support pregnant women who are in the difficult circumstances we have outlined. This is not the case at present. In fact, overall, the behaviour that attracts judicial intervention may be less threatening to fetal and neonatal health than the well-documented effects of poverty on a much larger number of pregnancies. F. . .1 [A] variety of appropriately designed supportive programs for pregnant women can at the same time help to ensure the well-being of the fetus. In particular, with respect to pregnant women who endanger the health of their fetuses by using alcohol or drugs, the Board of Trustees of the American Medical Association has recommended that "[p]regnant substance abusers should be provided with rehabilitative treatment appropriate to their specific physiological and psychological needs." Similar conclusions have been reached by this Commission and by others studying the problem of drug use during pregnancy. What is required is ready access to facilities and services that provide outreach, counselling, and treatment designed specifically for pregnant women that are appropriate to their needs. Conclusion and Recommendations

In summary, trying to use the law and the courts to protect fetal health can only be counterproductive. Such laws may on the surface, have appeal, because we all support the goal of the well-being of the fetus, and enacting them may appear to be a logical extension of society's interest in the health of the fetus. But there is nothing in our experience to demonstrate that such laws work in practice. Indeed, there is strong evidence to the contrary, particularly because the instruments available to the courts—forcing action under penalty of fines or incarceration—are brutally blunt and patently unsuited to the goal of promoting anyone's health or well-being. Clearly, if protecting the fetus is the goal, other methods are needed. A societal interest in pregnancy and birth—to maximize the chances for the birth of a healthy child—is a goal Commissioners strongly endorse; it is an important and worthy goal. But our examination of the legal, ethical, and social implications of judicial intervention leads to the inescapable

conclusion that judicial intervention is neither an acceptable nor an effective method of achieving that goal. Because the woman's consent and cooperation are needed to ensure a positive outcome for the fetus, it follows that the most effective way of caring for the fetus is through appropriate support and caring for the pregnant woman. The Commission therefore recommends that 273. Judicial intervention in pregnancy and birth not be permissible. Specifically, the Commission recommends that a) b)

c) d)

e)

medical treatment never be imposed upon a pregnant woman against her wishes; the criminal law, or any other law, never be used to confine or imprison a pregnant woman in the interests of her fetus; the conduct of a pregnant woman in relation to her fetus not be criminalized; child welfare or other legislation never be used to control a woman's behaviour during pregnancy or birth; and civil liability never be imposed upon a woman for harm done to her fetus during pregnancy.

274. Unwanted medical treatment and other interferences, or threatened interferences, with the physical autonomy of pregnant women be recognized explicitly under the Criminal Code as criminal assault. and that 275. All provinces/territories ensure that they have in place a) information and education programs directed to pregnant women so that they do not inadvertently put a fetus at risk; b) outreach and culturally appropriate support services for pregnant women and young women in potentially vulnerable groups; and c) counselling, rehabilitation, outreach, and support services designed specifically to meet the needs of pregnant women with drug/alcohol addictions. In conclusion, it is the Commission's view that almost all pregnant women will take steps to maximize their chances of a healthy birth if they have ready access to the information, prenatal care, social services, and income support necessary to do so. In

Royal Commission on New Reproductive Technologies: Judicial Intervention in Pregnancy and Birth

the Commission's view, extending care to the fetus by giving the pregnant woman the support she needs provides the best hope for enhancing the health and well-being of both the fetus and the woman carrying it.

General Sources

Dissent on Judicial Intervention in Pregnancy and Birth (Royal Commission on New Reproductive Technologies)

individuals and o f those whom they affect. Their cases should not be predetermined in service to the interests of some other or larger aggregate group, such as women, whose cause (or rather one available sociopolitical interpretation of whose cause) that individual has not explicitly embraced, since no one may be used as a means to an end, however worthy that end. There may be instances in which judicial intervention would enable and defend a woman's best interests, her actual consent and autonomy against the coercion arising from some particular factor in her situation. One such example would be the case of a severely addicted woman who states clearly and explicitly that she wants her child to be healthy but whose withdrawal symptoms would demonstrably drive her to seek the drug she abuses were she not in mandatory treatment. Since only judicial review and possible intervention would allow the nature of her most fundamental consent and the actual expression of her choice and autonomy to be ascertained and enabled, even autonomy would in some instances require the continued existence of the possibility of judicial intervention in pregnancy. The Commission report itself, in what appears to me to be a contradiction of its own position, refers to the appointment of a legal guardian for a person found mentally incompetent. Such a finding and appointment requires court examination of the case and would therefore in fact constitute judicial intervention. It is precisely the ineradicable need for the availability of objective assessment such as this in grave cases that is the point of this dissent. Questions of the existence or non-existence of independent legal or constitutional rights of the fetus are irrelevant to the issue. The state has been declared by the Supreme Court of Canada to have an interest in the fetus, which means that this interest must have some possibility and venue of exercise. The principle of the requirement of consent to treatment, including treatment in the interests of another of any age or relationship, is accepted both in ethics and in law, which means that a woman is protected in general from non-consensual intervention. Positing that the fetus "has no legal existence" and that no third party can volunteer to defend the rights of such

Suzanne Rowell Scorsone

[. .] I do not concur with the recommendation that judicial intervention in pregnancy not be permissible, nor do I concur with the associated legislative measures. Words like "never" are, in my view, far too absolute. Intervention is generally inadvisable, but should not be entirely precluded. The existing possibility of recourse to the courts, a disinterested forum with accepted legitimacy for mediation and resolution of conflict in matters of human welfare, remains necessary in an area so fraught with ambivalence on the part of all parties in very specific and particular personal difficulties. Nor would I support a departure from the normal protections of all individuals from medical or other intervention, whatever their sex. Application of the severe sanctions of the criminal law uniquely to interventions directed at pregnant women appears to me to be unjustified. Intervention in pregnancy is not fundamentally different from other forms of medical or social intervention, and women are not so different from men in their essence or before the law that the protections and sanctions governing them should be of different orders. Such an absolutization could, moreover, have negative and discriminatory implications, calling into question the equality of men and women before the law There are many issues in which attention to the collective status of women and the autonomy of women as women would be of proportionately overriding importance; this is not one of them. The consequences for individual vulnerable human beings, both woman and child, are too severe and personal and the variability of circumstances is too great to be resolved by an absolutized application of a general principle without the possibility of review of individual cases. The resolution of the situations of individuals should be determined in the best interests of those

Rodgers, S. "Juridical Interference with Gestation and Birth." In Research Volumes of the Royal Commission on New Reproductive Technologies, 1993.

Source: From Proceed with Care: Final Report of the Royal Commission on New Reproductive Technologies (1993), 1063-5 Reprinted with permission.

281

11

um I

311

II 1 Li

5 Reproductive Ethics

a being is therefore neither strictly accurate in law nor necessary to the ordinary protection of women. I do fully concur with my fellow Commissioners, however, in recommendations which would

maximize education, service, and care extended to all women, especially to those who are vulnerable or addicted so that risk to both woman and fetus can be avoided.

5.5 Cases

Case 1

Sex Selection in Canada The law in Canada prohibits sex selection of children at preconception, but there are loopholes that permit parents to determine the sex of their children. Since Canada has no laws regulating abortion, a fetus can be terminated if parents discover its sex and do not wish to have a child of that gender. Information about gender can be obtained from a variety of prenatal tests. Although sex disclosure is often discouraged, there is no law against it. Parents can also obtain information about the sex of a fetus by visiting US clinics that will provide this information and assist them with choosing the sex of their child. Recent studies show that the overall ratio of boys to girls born in Canada is about 105 to 100, which is consistent with the standard biological odds for each sex, so it does not appear as if there is a significant problem.' But those studies also provide some evidence of

sex selection favouring boys in some cultural subgroups, in particular, among immigrants from India. It appears that the proportion of boys born to that group in Canada increases with the number of children born to a mother so that by the third birth 138 boys are born for every 100 girls. However, it is not clear how troubling this is if the decisions are motivated to have a child of a different gender from the one(s) already born and there is no overall effect on the normal biological ratio of boys to girls. Australia is currently considering national guidelines for in vitro fertilization that would allow for sex selection for a parent's third child.2 There is also some evidence from those involved in sex selection clinics in the United States that girls are preferred over boys and that preferences are changing toward girls in other countries.3 What changes, if any, there should be to public policy about sex selection in Canada depends crucially on difficult-to-obtain accurate information about current sex selection practices. What policy measures can, or should be, taken in these circumstances?

Notes 1.

2.

Wency Leung, "Some Couples in Canada Practicing Prenatal Sex Selection in Favour of Male Fetuses, Studies Suggest," The Globe and Mail 11 April 2016, httpf/ www.theglobeandmail.comilife/health-and-fitness/ health/some-couples-in-canada-practising-prenatalsex-selection-in-favour-of-male-fetuses-studies-show/ artic1e29583670. Rebecca Sullivan, "Australian Parents May Be Able to Pick the Sex of Their Baby under Proposed 1VF

Case 2

Prenatal Diagnosis and Abortion or Infanticide through Declining Treatment Berthe K is a 35-year-old woman who is pregnant for the first time after years of primary infertility. She is at

3.

Law Changes, News.com.au 25 July 2016, http:// www.news.com.au/lifestyle/parenting/babies/ australian-parents-may-be-able-to-pick-the-sex-oftheir-baby-under-proposed-ivf-law-changes/news-story/ e84aa850333379c6ce26deef80fdc815. Douglas Todd, "Economics Drives Revolution in Sex Roles: And We're Not Ready," The Vancouver Sun 12 July 2010, httplivancouversun.com/news/staff-blogs/economicsdrives-revolution-in-sex-roles-and-were-not-ready.

eight weeks gestation. Since her advanced age increases the risk of having a fetus with Down syndrome, Mrs K is offered an amniocentesis to be done at 15 to 20 weeks gestation. The procedure involves removing and replicating fetal body cells from the amniotic fluid surrounding the fetus in the uterus so that it may be subjected to genetic testing. The test will detect or rule out Down syndrome and other significant genetic anomalies in the

Cases

fetus. Mrs K asks what would happen if the test were positive. She is told that the test is intended to provide her and her husband with information only, and that the choice of what to do with that information would be theirs to make. Their options would be to continue with the pregnancy or to terminate it. Mrs K is very uncomfortable with the prospect of abortion. After discussing the matter with family and friends, the Ks make the decision not to have the amniocentesis. When the baby is born, Mr and Mrs K discover that he does have Down syndrome. He has a heart murmur and a life-threatening intestinal blockage and faces the

likelihood of mental disability in the future. The parents are informed that there is no way of knowing how profound the mental disability will be but that with surgery the infant will have a long life ahead of him. After much soul-searching and many tears, the Ks decide to refuse the surgery. The physician makes the situation clear: 'Without the surgery, your son will die. With it, he can have a long, full life!" Mr K replies: "You would have allowed my wife to have genetic testing and then an abortion halfway through the pregnancy. Whath the difference between that and allowing our son to die now? The condition threatening him is the same, and it's our decision to make."

Case 3

she and her husband desperately want a boy. Theirs is a low-income household, and they are unwilling to take on the costs of caring for another child unless that child is male. If the physician can use prenatal testing to determine that the fetus is male, then she will continue with the pregnancy. If, however, the tests determine that the fetus is female or if she is denied testing that will confirm the sex of the fetus, then she will have an abortion. Should the physician do the test?

Abortion of Suspected Female Fetus Claudia Gonzalez is a 30-year-old married woman with three daughters under the age of 10. Ms Gonzalez is pregnant for the fourth time, currently at 10 weeks gestation, and has come to an appointment with her physician to discuss prenatal testing. She explains that

Case 4

Ms G and Refusal of Treatment while Pregnant In August 1996, 22-year-old Winnipeg resident Ms G, already a mother of three, learned she was pregnant again. Two of her children experienced brain damage caused by Ms Gs abuse of solvents—paint thinner, glue, nail polish remover—during the two pregnancies. Concerned about the welfare of both the unborn fetus and its mother, health care providers brought Ms G's case to the attention of Winnipeg's Child and Family Services, who sought custody of Ms G so they could enforce treatment on her during her pregnancy for the well-being of the fetus. Mr Justice Perry Schulman of the Manitoba Court of Queen's Bench committed Ms G to drug treatment after ruling that she was not mentally competent to refuse it on account of her addiction.'

However, after a court-ordered psychiatric report found Ms G to be competent, she appealed the enforced drug treatment.2 The Manitoba Court of Appeal unanimously reversed the lower court's decision, ruling that courts cannot order a mentally competent person to undergo treatment against her will.3 Winnipeg's Child and Family Services appealed to the Supreme Court of Canada in an attempt to determine whose rights should prevail: those of the mother or those of the fetus.4 By June 1997, when the case reached the Supreme Court of Canada, Ms G had given birth to her child and was pregnant again.5 Additional arguments, on both sides of the issue, were heard from 12 intervenors representing a range of interests—women's rights, Aboriginal rights, human rights, religious rights. In late October 1997, the Supreme Court issued a 7-2 decision stating that a fetus does not have legal rights and that courts cannot, therefore, force a pregnant woman to undergo treatment in order to prevent harm to the fetus.°

Notes 1.

Winnipeg Child and Family Services v G (DF) 1996.

2. "Judge's Order for Care 'Ethical,'" The Globe and Mail 8 August 1996, Al.

283

11111 1

11

5 Reproductive Ethics

3. 4.

"Pregnant Women's Rights Affirmed," The Globe and Mail 13 September 1996, Al. "Top Court to Consider Fetal Rights," Toronto Star 19 October 1996, Al2.

Case 5

Endangering Behaviour in a Pregnant Woman Christina C. is a 24-year-old woman who lives on the streets of Vancouver and works as a prostitute. She has been brought to the hospital as the result of a heroin overdose. Christina is seriously underweight and malnourished, and the needle marks on her body show that she has been addicted for a long time. When Christina revives enough for the HCP5 to take a full medical history she discloses a previous diagnosis of hepatitis C and HIV as well as the possibility that she is pregnant. A test confirms her pregnancy. Because Christina is so badly malnourished, it is difficult to determine the gestational age of the fetus

Case 6

Abortion of a Fetus Due to Diagnosis of Down Syndrome As revealed by Dr Ken Seethram a t the Canadian Society of Fertility and Andrology conference, sometime in 2010 a BC couple discovered that their gestational carrier [GC] or "surrogate mother" was carrying a fetus with Down syndrome.' The GC had been implanted with an embryo created with the commissioning parents' egg and sperm. A first-trimester ultrasound indicated that Down syndrome was likely, which was later confirmed in a diagnostic test. Down syndrome is a genetic condition in which a fetus inherits three copies of chromosome 21 instead of the typical two. It occurs in one in 700-900 births. The resultant cellular changes cause health problems, developmental delays, and learning disabilities.2 More than 40 per cent of affected fetuses will have a lifethreatening congenital heart malformation.3 Eye and hearing problems are common, as are gastrointestinal blockage, underactive thyroid, chronic constipation, sleep apnea, and a tendency to hip dislocation.4 Increased risk of developing leukemia has also been

5. 6.

"Woman in Glue-Sniffing Case is Pregnant Again," Ottawa Citizen 18 June 1997, Al. "Lawmakers Must Decide Rights of Unborn, Top Court Says," Ottawa Citizen 1 November 1997, Al.

from her last menstrual cycle; however, a best-guess estimate suggests that she is about five months pregnant. Christina shows no interest in hearing about proper prenatal care or in taking care of her medical issues: her principal concern is with being discharged so that she can get back to work. The HCP5, conferring among themselves, believe that she will be using drugs again within hours, putting herself and her fetus at risk. They explain to her that drug use during pregnancy will seriously harm the fetus, perhaps causing it to be born with mental disabilities. Christina is indifferent: "Social services took my boy, they'll take this one, too. Are you going to let me go now?" As they watch her wander off into the night, the attending physician shakes his head sadly "There's another child the system is going to have to take care of. If only we could get her into treatment, just until the baby is born, they'd both be better off."

linked to the condition.5 Down syndrome individuals have specific appearance characteristics, like a smallerthan-normal and abnormally-shaped head, excess neck skin, flattened nose, small ears and mouth, upwardslanting eyes, wide short hands with short fingers, and shortened stature. They can expect delayed social and mental development, learning disabilities, and even dementia. Impulsive behavior, poor judgment, short attention span, and slow learning are common problems, and children with Down may exhibit frustration and anger as they grow and become aware of their limitations.4 There is no way of knowing or estimating the full degree of disability at the fetal or even infant stage. This may be why data shows that up to 85 per cent of pregnant women given a Down syndrome diagnosis opt for abortion.° In this case, the commissioning parents urged the GC to abort the affected fetus. She initially refused, being determined to take the pregnancy to term. However, under the agreement the trio had signed, choosing to continue with the pregnancy would mean absolving the commissioning parents of any responsibility for raising the child. This raised thorny legal questions, as Canada's Assisted Human Reproduction Act does not address parentage or parental rights, leaving that up to

Cases 1 285

provincial statutes, which have yet to be written. Before this could become an issue, however, the GC chose to abort the fetus, in part because of her obligations to her family which included two children of her own.' The case drew criticism from various Canadian bioethicists. Francoise Baylis from Dalhousie University said that the case demonstrates the dangers of treating life like a commodity: "The child is seen by the commissioning parents as a product, and in this case a

substandard product because of a genetic condition." Juliet Guichon from the University of Calgary questioned whether the rules of commerce should apply to the creation of children: "It's kind of like stopping the production line: 'Oh, oh, there's a flaw-1 Disability rights' activists were likewise appalled by the suggestion that a fetus is considered inferior or defective because it faces developmental challenges, not worthy of being allowed to live.

Notes 1.

2.

3.

Tom Blackwell, "Couple Urged Surrogate t o Abort Fetus Due to Defect," National Post 6 October 2010, http:// life.nationalpost.com/2010/10/06/couple-urged-surrogatemother-to-abort-fetus-because-of-defect (accessed 26 June 2011). Down Syndrome Research Foundation, http://www.dsrf .org/news_%26_information/information_on_down_ syndrome (accessed 26 June 2011). Canadian Down Syndrome Society, http://wwwcdss.ca/ information.html.

4.

5.

6.

Down Syndrome Trisomy 21 (2010) National Institutes of Health, http://wwwncbi.nlm.nih.gov/pubmedhealth/ PMH0001992 (accessed 26 June 2011). National Cancer Institute, n.d., "Children with Down Syndrome," http://www.cancer.govicancertopics/pdq/ treatment/childAML/HealthProfessional/page8 (accessed June 26, 2011). James L. Natoli, "Prenatal Diagnosis of Down Syndrome: A Systematic Review of Termination Rates," Prenatal Diagnosis 32 (2012) no. 2, 142-53.

5.6 Study Questions 1. What are Christine Overall's consequentialist and deontological arguments for the right to reproductive freedom? What are the limitations to this right? 2. How do you think John A. Robertson would respond to the Royal Commission on New Reproductive Technology's arguments against sex selection of children? Who has the stronger evidence for their position? 3. What is the law on commercial surrogacy i n Canada? What are the arguments for and against it? 4. What is the Conservative Theory? What do you think is the best argument for it? I s a Conservative committed to condemning celibacy and contraception? 5. What is the Liberal Theory? What do you think is the best argument for it? Must a Liberal embrace infanticide? Is infanticide justifiable? 6. What is the Middle Theory? What is the standard objection to it? How does Wayne Sumner try to overcome that problem, and does he succeed? 7. How does Judith Jarvis Thomson try to show that it is not always wrong to kill a human being with a full right to life? Under what conditions (if any) would Thomson say that it is wrong to kill a human fetus with a full right to life? 8. What reason does Susan Sherwin give for saying that a woman has full control over her pregnancy? Do you agree? 9. If a woman behaves in a way that puts her fetus or child-to-be at risk, should judicial intervention ever be invoked? What does the Royal Commission on New Reproductive Technologies say? What does Suzanne Rozell Scorsone say? Who do you think is right, and why?

II III I

u

I

111111

5 Reproductive Ethics

5.7 Suggested Further Reading Busby, Karen, and Delaney Vun. 2010. "Revisiting The Handmaid's Tale: Feminist Theory Meets Empirical Research on Surrogate Mothers." Canadian Journal of Family Law 26: 13-94. Dickens, B.M., "Can Sex Selection Be Tolerated?" 2002. Journal of Medical Ethics 28: 335-6. Feinberg, Joel, ed. 1984. The Problem of Abortion. Belmont, CA: Wadsworth. Flagler, Elizabeth, Francoise Baylis, and Sanda Rodgers. 1997. "Ethical Dilemmas That Arise in the Care of Pregnant Women: Rethinking 'Maternal-Fetal Conflicts."' Canadian Medical Association Journal 157 (12): 1729-32. Hanna, Jason K.M. 2010. "Revisiting Child-Based Objections to Commercial Surrogacy." Bioethics 24: 342-7. Kluge, E.-H., "There Ought to Be a Law" and Zimmerman, David, "No Easy Answer." 1987. BC Medical Journal 29 (1). Murray, Thomas H. 1996. "Moral Obligations to the Not-Yet Born The Fetus as Patient." In Thomas H. Murray, The Worth of a Child, ch. 5,96-114. University of California Press. Nelson, Lawrence", and Nancy Milliken. 1990. "Compelled Medical Treatment of Pregnant Women: Life, Liberty, and Law in Conflict." In Richard T. Hull, ed., Ethical Issues in the New Reproductive Technologies, 224-40. Belmont, CA: Wadsworth. Purdy, Laura M. 1990. "Are Women Fetal Containers?" Bioethics 4 (4): 372-91. Rhoden, Nancy K. 1989. "Cesareans and Samaritans." In John Arras and Nancy Rhoden, eds., Ethical Issues in Modern Medicine, 3rd ed., 321-8. Mountain View, CA: Mayfield. Robertson, John A., and Joseph D. Schulman. 1987. "Pregnancy and Prenatal Harm to Offspring: The Case of Mothers with PKU." Hastings Center Report 17 (4): 23-32. Savalescu, Julian. 2001. "Procreative Beneficence: Why We Should Select the Best Children," Bioethics 15: 413-26. Steinbock, Bonnie. 2009. "Mother—Fetus Conflict." In Helga Kuhse and Peter Singer, eds, A Companion to Bioethics, 2nd edn., 149-60. Blackwell Publishing. Sumner, L.W. 1981. Abortion and Moral Theory. Princeton, NJ: Princeton University Press. Tooley, Michael. 1983. Abortion and Infanticide. Oxford: Clarendon Press. —. 1972. "In Defense of Abortion and Infanticide." Philosophy & Public Affairs 2 (1): 37-65.

End-of-Life Dacision-Making 6.1 Introduction This chapter discusses three related but distinct issues about life and death. The first concerns withholding or withdrawing life-sustaining treatment (wLST). The second is about what is commonly called "assisted suicide and active euthanasia" but is now in Canada called "physician-assisted death" (PAD) or "medical assistance in dying" (MAID). PAD became legal in Canada in 2015, and legislation regulating it was passed in 2016. There is much controversy over both PAD and the ensuing legislation. We will first examine the question of whether PAD should be legal at all and then turn to the question of how the law should regulate PAD. The third issue concerns death itself, and the chapter ends with a short essay on defining death. It will emerge that the problem of deciding when to end life and the problem of defining death have much in common.

Withholding and Withdrawing Life-Sustaining Treatment Canadian law and biomedical ethics have now firmly established that it is permissible to WLST from a wide range of capable and incapable patients. This has not always been so. Section 197 of the Criminal Code of Canada states that no treatment necessary for life can be withheld and section 199 that no treatment necessary for life can be withdrawn. This was read literally and understood to be the law of the land until the late 1970s. The current law slowly evolved from this position through a series of court decisions in Canada and the United States.' These decisions began with the right of capable patients in a terminal condition to refuse life-saving treatment and expanded that right. Treatment could be withdrawn as well as withheld. The patient did not have to be terminal. Foods and fluids as well as ventilation could be withheld or withdrawn. The patient's substitute decision-maker

A Guide to Acronyms Related .W End-of-Life Decision-Making PAD = physician-assisted death MAID = medical assistance in dying WLST = withholding and withdrawing life-sustaining treatment SCC = Supreme Court of Canada DDE = doctrine of double effect

288

6 End-of-Life Decision-Making

or physician could make the decision. Through these developments, Canada arrived at the current law, which allows capable patients and their substitute decision-makers or physicians to WLST if it is not wanted or beneficial. The reading from the Health Law Institute, Dalhousie University, "Withholding and Withdrawal of Potentially Life-Sustaining Treatment" (see page 295) summarizes the current law on WLST. We see a similar story of gradualism repeating when we look at the debate on implementing PAD.

Physician-Assisted Death: Definitions, Ethics, and Law The permissibility of physicians to WLST raises the question of whether they should also be allowed to practise PAD/MAID. However, any systematic discussion of this topic must begin by working through a tangle of definitions of the terms from the box above and others besides. We begin with "euthanasia" and "assisted suicide." Although Canada has abandoned these terms in favour of PAD/MAID, they are in common use in most of the world. Understanding them will enable us to follow the world literature and will also help us to understand the terminology of PAD and MAID. The Scc used the term PAD, whereas the Canadian federal legislation that emerged uses the term MAID. We shall use "PAD" except for cases where the discussion of the Canadian legislation requires using "MAID."

Terminology euthanasia bringing about the death of a person in that person's interest.

passive euthanasia withholding or withdrawing Gfe-sustaining treatment (wLST).

active euthanasia administering a lethal agent to another person in order to kill the person.

assisted suicide providing a person with the means of death.

PAD physician-assisted suicide or active euthanasia performed by a physician.

MAID assisted suicide or active euthanasia performed by a physician or a nurse practitioner, as per amendments to the Criminal Code of Canada in 2016.

Euthanasia consists of bringing about the death of a person in the person's interest. The italicized clause is what distinguishes euthanasia from other reasons for bringing about the death of a person. Usually death is not regarded as a benefit for the patient, but in euthanasia it is, and this makes it possible to make out a moral case for the practice. We can also distinguish between two kinds of euthanasia: "active" and "passive." Passive euthanasia is WLST. Active euthanasia is an intervention such as injecting a patient with a lethal dose. In most jurisdictions, assisted suicide differs from active euthanasia in that it is the patient versus someone else who causes the death—e.g., by drinking something or pushing a button. Under Canadian law, however, assisted suicide does not require that a death occur at all. Assisted suicide is performed when one provides another with the means to death, whether or not the person uses those means. In the terminology of PAD, the distinction between active and passive euthanasia disappears. Passive euthanasia (i.e., wLST) is not euthanasia at all but something to be contrasted with PAD. Likewise, the distinction between assisted suicide and active euthanasia disappears and are both encompassed by the term PAD. The qualifying term "P" (Physician) indicates that these acts are permissible only by physicians. The alternative term MAID ("medical assistance in dying") was proposed to allow assisted suicide or active euthanasia to be performed by a physician or a nurse practitioner, as per amendments to the Criminal Code of Canada in 2016. Cutting across the distinction between PAD and WLST is another one. This concerns whether the aid in dying is "voluntary," "non-voluntary" or "involuntary" PAD and WLST are voluntary if they are done with the consent of the patient or substitute decision-maker, non-voluntary if done without the consent of those parties, and involuntary if done (as proposed in the Rasouli case in Chapter 1; see page 2) contrary to their request. In this chapter we will be concerned primarily with voluntary PAD.

Legal Status of PAD PAD is

legally permitted in Belgium, the Netherlands, Luxembourg (the so-called "Benelux" countries), and Colombia. It is also legal in four US states (Oregon, Washington, Vermont, and Montana) and was ruled legal in Canada in 2015. The conditions under which PAD is

Physician-Assisted Death: Definitions, Ethics, and Law

permitted in these jurisdictions vary. Some allow assisted suicide but not active euthanasia; some restrict PAD to capable individuals, whereas others allow substitute decision-makers to decide. There are also variations in whether advance directives give valid authorization and eligibility requirements such as age and mental capability. We will consider the question of how the legislation is best written later, but for now we look at whether PAD should be legal at all. Two kinds of arguments can be urged in favour of legalizing PAD, which we distinguish as "Philosophical" and "Constitutional," and we will look at them in turn.

Philosophical Argument for Legalizing PAD The philosophical argument for PAD is set out by Dan Brock in "Voluntary Active Euthanasia" (see page 305). According to Brock, "the central argument" for PAD (to put it in Canadian terms) rests on an appeal to two important values: "individual self-determination" and "individual well-being." It is hard to deny the importance of these things, and they combine to make a strong case for legalizing PAD. If we have a moral right to determine our own destiny, and if we have a moral right to minimize our suffering, we must have a moral right to request and receive PAD from any physician who is willing to provide it. Thus, we get a presumptive case for legalizing PAD grounded in values that are recognized as fundamental to biomedical ethics—namely, autonomy and beneficence.

Constitutional Argument for Legalizing PAD The constitutional argument for PAD consists of the claim that prohibiting PAD violates some Charter right or rights. In 1982, a Charter of Rights and Freedoms was added to Canada's Constitution. This gave those who wanted to change the law on PAD access to courts, and it was a constitutional argument that made PAD legal in Canada. The case that decided the issue was the so-called "Carter Case"2 and involved two women, Gloria Taylor and Kay Carter. Gloria Taylor suffered from ALS and did "not want to die slowly, piece by piece" or "wracked with pain." Kay Carter suffered from spinal stenosis, which results in the progressive compression of the spinal cord and ends in immobility and chronic pain. The argument made in the Scc was that the law prohibiting PAD violated section 7 of the Canadian Charter of Rights. That section guarantees Canadians a right to life, liberty, and security of the person. The Scc unanimously found that section 7 rights were infringed by the prohibition of PAD. Life was infringed because the prohibition of PAD has the effect of forcing individuals whose natural death would be dreadful to take their own lives prematurely—that is, before they become so incapacitated by their illness that they cannot commit suicide. Liberty was infringed because without access to PAD, individuals who are unable to commit suicide cannot lawfully do what others can and thus are discriminated against. And security was infringed because without the possibility of PAD, individuals who do not have the ability to kill themselves will sometimes have to endure a dreadful death.

Objections to Legalizing PAD The philosophical and constitutional arguments presented above do not settle the question in favour of legalizing PAD. They only make a presumptive (or, as it is sometimes called, a "prima facie") case for legalizing the practice—that is, a case that settles the question in the absence of countervailing considerations, and there may be some such countervailing considerations present. If the case is made on moral grounds, there may be more weighty moral considerations (e.g., the risk to the vulnerable) on the other side. As well, for a Charter challenge to succeed, it must not only be shown that some article of the Charter is infringed, it must also be shown that the infringement "cannot be justified in a free and democratic society."

289

II

i.111

111141

6 End-of-Life Decision-Making

All prior attempts to legalize PAD in Canada failed for one or the other of these reasons. We must now look at what the arguments against legalizing PAD are and ask whether they are sufficient to cancel the case for legalizing it. Brock identifies two main arguments against making PAD legal. The first claims that the practice is intrinsically wrong. The second, that legalizing it will lead to harmful consequences. We will take these in turn. PAD Is Intrinsically Wrong The case for saying that PAD is intrinsically wrong rests on a view that has roots in the Hippocratic tradition according to which physicians are healers. Their proper business is to protect life and promote health, and hence (so the argument goes) PAD cannot be countenanced by physicians. It is not that PAD is wrong because of bad consequences to patients or society but wrong in itself—intrinsically wrong—because it contradicts the proper aim of medicine. Two arguments can be presented to show that PAD is intrinsically wrong. The first is that PAD is intrinsically wrong because it consists of killing the patient. The second is that PAD is intrinsically wrong because it consists of intentionally killing the patient. Doctors Must Not Kill

direct or proximate cause the cause that is closest to the event to be explained.

indirect orremote cause a cause that causes the proximate cause.

The problem with resting the intrinsic wrongness of PAD on the view that doctors must not kill is that physicians regularly WLST. And since (so the argument goes) WLST causes death and causing death is killing, it is not true that doctors must never kill. They do so whenever they WLST, and WLST is an essential part of medical practice. Brock's article (see page 305) and James Rachels's "Active and Passive Euthanasia" (see page 301) accept this argument and conclude that there is no morally relevant difference between WLST and PAD. They thus argue that if WLST is permissible, so PAD must be. Daniel Callahan in "When Self-Determination Runs Amok" (see page 317) takes a different view He argues that the argument endorsed by Brock and by Rachels rests on a mistake about causation. In Callahan's view, in WLST the doctor does not cause the death of the patient. Rather, the doctor returns the patient to the state of having an untreated disease condition, and it is that condition that causes the death. This view is widely accepted in medicine and the law (see Case 1 on Nancy B., page 341), but it is problematic. We may agree that the untreated disease condition is the direct or proximate cause of the patient's death. However, if being in an untreated disease condition caused the patient's death, and the doctor caused the patient to be in that untreated condition, the doctor caused the death. Callahan may say that withholding or withdrawing treatment is merely an indirect or remote cause. But an indirect cause is still a cause, and it is arbitrary to pick out the direct cause as the cause. The fact is that any event has many causes—conditions "but for" which the event would not have happened—and we typically do not cite them all in explaining an event. Which cause or causes we pick out to explain an event is determined by what we think are most unusual or significant in the circumstances. For example, it may be appropriate to cite the fact that the doctor withdrew treatment as the cause of death if the speaker thought the hearer did not know that treatment was withdrawn. On the other hand, it may be more appropriate to cite the disease condition if the speaker thought the hearer understood that treatment was withdrawn but did not understand why that would cause death. It is also worth noting that citing an indirect cause is often more useful than the direct cause. There is an old joke that says pathology is easy. everyone dies of heart failure. It is true that the direct cause of all death is heart failure, but the task of the pathologist is to say why the heart failed—that is, to identify an indirect cause. Philosophic controversy over the killing/letting die distinction continues to be lively, and there is a large literature on it (see the Suggested Further Reading at the end of this chapter). But many (perhaps most) philosophers dismiss its significance for the PAD debate, and the SCC briskly set it aside with the comment: "the preponderance of evidence from

Physician-Assisted Death: Definitions, Ethics, and Law

ethicists is that there is no ethical distinction between physician-assisted death and other end-of-life practices whose outcome is highly likely to be death."3 Doctors Must Not Intentionally Kill

Even if WI_ST does not constitute killing, there is another obstacle to the view that doctors must never kill their patients. It comes from the fact that it is part of a physician's job to keep patients free of pain as much as possible. It is also understood both in law and as good medical practice that physicians can provide patients with sufficient sedation to quell distress. If they can relieve suffering with non-lethal amounts of drugs, they cannot provide a lethal dose. But if the distress continues to the point at which distress cannot be quelled without the sedation killing the patient, the physician can provide a lethal dose of sedation. When this occurs in the case of dying patients, it is called "terminal sedation." Since it is uncontroversial that physicians can sometimes provide terminal sedation, the case for the intrinsic wrongness of PAD must shift from the claim that physicians must never kill their patients to the claim that they must never intentionally kill them. The question now is whether the prohibition of intentional killing can be reconciled with the practice of terminal sedation. The most common way of trying to reconcile providing terminal sedation with physicians' putative obligations not to intentionally kill their patients is to appeal to the doctrine of double effect (DDE). Joseph M. Boyle explains and defends that doctrine in "Medical Ethics and Double Effect: The Case of Terminal Sedation" (see page 297). The DDE comes into play when an agent is faced with a situation in which a single action has a good effect and a bad effect and the good effect cannot be brought about without bringing about the bad. The DDE tells us that the action can be performed if the good effect is intended and the evil one merely foreseen. This allows physicians to provide terminal sedation if they intend to quell suffering and foresee, but do not intend to cause, the death of the patient. Thus (so the argument goes), physicians can always keep their patients free of distress and yet never intentionally kill them. The DDE is accepted in law in many jurisdictions and by the medical profession, but as Boyle makes clear, the DDE is philosophically open to challenge, and we raise three questions about it. First, to judge the permissibility of actions by intentions requires that the agent's intentions can be identified. But this is not a straightforward task. A doctor providing terminal sedation, for example, may have a number of things in mind such as to reduce suffering and kill the patient and honour a promise and ease a burden on the family. It is unrealistic to suppose that she (or anyone else) will always or even often be able to identify what she intends, what she merely foresees, and the relative importance of her intentions in the way the DDE requires. Second, it is not clear that the DDE enables physicians to provide terminal sedation and not have an intention to cause death. The utilitarian philosopher Jeremy Bentham distinguished two kinds of intentions: "direct" and "oblique." Direct intentions are what the agent aims at. Oblique intentions are what the agent foresees as consequences of her actions.4 If we define "intentions" as "direct intentions," physicians need never intentionally cause death when they provide terminal sedation. On the other hand, if we define "intentions" as (or as including) "oblique intentions," physicians always intend to cause death as long as they foresee death as a consequence. The reconciliation of terminal sedation with the alleged duty not to intentionally kill thus relies on picking out the direct intention as the morally relevant one. But it is not clear what justifies that selection. We ordinarily count foreseen consequences as intended consequences and hold agents responsible for them. What, then, can justify excluding them in cases of terminal sedation? Finally, there is a question of why physicians should be interested in the DDE. That view is only important to physicians who hold that they must never intentionally kill patients in the course of their professional practice. But why should physicians want to hold that view? Intentionally killing another human being is not generally considered always wrong. It is normally thought permissible (even for physicians) to kill in selfdefence, to protect the defenceless, and in war, for example. Why then should physicians

terminal sedation the use of high doses of pain medication to treat a patient's suffering at the end of life, with the awareness that such high dosages will hasten the patient's death.

doctrine of double effect (DDE) the principle that a proposed action that will have benefits but will also cause some harm is permissible if the action itself is morally appropriate and the foreseen harm is not intended.

292

6 End-of-Life Decision-Making

hold that they cannot intentionally kill as a part of their practice? Without a compelling answer to this question, the case for legalizing PAD cannot be rejected by appeal to the claim that doctors must never intentionally kill. There also is disagreement over whether a compelling answer can be given. Brock (see page 305, particularly the section entitled "The Role of Physicians") and Callahan (see page 317, particularly the section entitled "Euthanasia and Medical Practice") take up this question in their articles. Brock argues that physicians should accept intentional killing as part of their job description; Callahan defends the view that they should not.

PAD Will Have Bad Consequences

slippery slope a situation that is likely to lead to an undesirable consequence, whether or not one tries to prevent that.

slippery slope argument an argument against allowing a situation to occur because it is likely to lead to an undesirable situation.

We now turn to the argument that PAD should not be legalized because of potential bad consequences. Brock provides a list of six such consequences (see page 305, especially pages 311-313), and similar lists dot the literature. There are two main concerns about legalizing PAD. The first has to do with risk to individuals requesting it. Here the worry is whether we can ever have sufficient assurance that those requesting PAD are competent and their consent is informed, voluntary, not ambivalent, and not compromised by social vulnerabilities such as age or disability. The second is whether legalizing PAD will lead to a slippery slope. The worry is that legalizing voluntary PAD today will lead to non-voluntary PAD tomorrow and involuntary PAD the day after. What starts out as policy designed to promote the dignity and minimize the suffering of individuals thus may end up as one that aims at eliminating those who are burdens on society. Given this (the argument runs), PAD should not be legalized. The possible risks to individuals requesting PAD and undesirable social consequences have been the main deterrents to legalizing PAD. Until recently, they have been largely based on speculation about what might happen, but now that many jurisdictions permit PAD, the evaluation of risk can be put on an empirical basis. The Supreme Court of BC (from which the Carter case came on appeal) undertook just such an evaluation. After what was arguably to that date the most comprehensive examination of jurisdictions that allow PAD, the court concluded that there is no risk to the vulnerable from misuse or abuse, or of a slippery slope, that would require an absolute prohibition of PAD. At a certain level this was not very surprising, since the same risks equally apply to WLST and PAD. As there is no evidence of systematic abuse of WLST, and no evidence of a slippery slope in this context, many have argued that professional and legal standards that regulate WLST can be expected to apply to PAD without risk The reading from the BC Supreme Court, "Controlling the Risks of PAD" (see page 321) provides the concerns Canada had about such risks and the conclusions of the court. The SCC accepted the BC Supreme Court's evaluation of risks of PAD, and with that, the last obstacle to legalizing PAD—the fear of bad consequences—was overcome. There was thus nothing to counterbalance the infringements of the life, liberty, and security of the person guarantees of section 7 of the Charter, and the SCC ruled, 9-0, that the sections of the Criminal Code prohibiting PAD are unconstitutional. Specifically, the SCC found that legislation cannot prohibit PAD for competent adults who (1) clearly consent to the termination of life and (2) have a grievous medical condition (including an illness, a disease, or a disability) that is irremediable (meaning that it cannot be alleviated by means acceptable to the individual) and causes enduring suffering that is intolerable to the individual.

Implementing PAD: Access, Safeguards, Conscientious Objection The SCC suspended the effect of its declaration for 12 months (later extended by four months) to give the federal and provincial governments time to put legislation in place. To draft the legislation, the federal government of Canada had to decide three issues.

Implementing PAD: Access, Safeguards, Conscientious Objection

These concerned who has access to PAD, what safeguards should be put in place against abuse and misuse of PAD, and what accommodation should be made for those individuals or institutions that have conscientious objection to PAD. Current debates on PAD centre on these issues, and we look at each in turn. In each case we provide a discussion of the issue, report what legislation the government decided on, and identify issues that still have to be decided. The reading from the Department of Justice entitled "Legislative Background: Medical Assistance in Dying (Bill C-14)" (see page 323) explains how the government arrived at its legislation. The reading "Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection" comprises excerpts from recommendations made by the ProvincialTerritorial Expert Advisory Group on Physician-Assisted Dying as well as from the Special Joint Committee on Physician-Assisted Dying (see page 328). These reports were prepared to assist the government in writing the legislation and are included to help in evaluating the legislation and thinking about the issues yet to be settled.

Access The Scc in the Carter case ruled that any capable adult who has a grievous condition, experiences suffering that is irremediable in any way acceptable to the person, and who makes a voluntary and informed choice is entitled to PAD. In Bill C-14, which is the legislation putting PAD into effect, the federal government of Canada translated the judgement in Carter into law by saying that individuals are eligible to request and receive MAID only if: (a) they have a serious and incurable illness, disease or disability, (b) they are in an advanced state of irreversible decline in capability; (c) that illness, disease or disability or that state of decline causes them enduring physical or psychological suffering that is intolerable to them and that cannot be relieved under conditions that they consider acceptable; and (d) their natural death has become reasonably foreseeable, taking into account all of their medical circumstances, without a prognosis necessarily having been made as to the specific length of time that they have remaining.' The Department of Justice "Legislative Background" document mentioned above (see page 323) explains that the above conditions of access are a work in progress and identifies for further study the possibility of providing access to MAID by "mature minors, advance requests, and requests where mental illness is the sole underlying medical condition." Further questions can b e raised about the legislation. Should substitute decision-makers who can make decisions to WLST from incapable adults, infants, children, and adolescents be able to request MAID on their behalf as well? One may also ask why candidates for MAID must have a terminal diagnosis or otherwise have their deaths foreseen. This condition is insisted on by Quebec in its end of life legislation (Act Respecting End of Life Care, CQ LR, c. S-32.0001) and all American states that allow PAD but not by any of the Benelux countries. An argument for the condition is that it limits the risks of MAID by limiting access to it, but in so doing it also excludes persons with disabilities from MAID. One irony of including the foreseeable death condition is that Ms Taylor and Ms Carter would not benefit from the legislation because they were disabled and suffering but not expected to die soon.

Special Safeguards Every jurisdiction in the world that allows PAD surrounds i t with special safeguards such as requiring a second opinion, a mandatory waiting period, witnessing procedures, and so on. The Scc gave encouragement to attaching special safeguards when it cited the conclusion

conscientious objection/ refusal when health care providers refuse on moral grounds to provide a treatment that is permitted by the standards of their profession.

H ..111

294

IIILU

6 End-of-Life Decision-Making

of the trial judge of the BC Supreme Court, saying that the risks inherent in PAD can be substantially minimized through a "carefully designed system imposing stringent limits that are scrupulously monitored and enforced."6 The federal government followed this view in Bill C-14. The preamble to the act states that "robust safeguards are essential to prevent errors and abuse in the provision of MAID," and the act goes on to require that requests for MAID be: (a) made in writing after the person was informed that his or her natural death was reasonably foreseeable; (b) signed and dated in the presence of two independent witnesses; (c) accompanied by a written opinion by a second independent medical practitioner or nurse practitioner confirming the person's eligibility, (d) made at least 15 days before provision of MAID (unless both practitioners agree that death or loss of capacity to consent is imminent); (e) confirmed immediately before providing MAID.? Not everyone, however, thinks special safeguards are required or desirable. An alternative to attaching special safeguards is to leave the decision as a private matter between patients, families, and physicians, just as decisions to WLST are currently left. The Scc also gave encouragement to this approach when it wrote: Logically speaking, there is no reason to think that the injured, ill and disabled who have the option to refuse or to request withdrawal of lifesaving or lifesustaining treatment, or who seek palliative sedation are less vulnerable or less susceptible to biased decision-making than whose who might seek more active assistance in dying.8 Indeed, if there are no greater risks in PAD than in WLST, it would be a logical consequence that either there should be no special safeguards for either or there should be special safeguards for both. Brock acknowledges this (see page 305, especially page 315) and suggests that serious consideration be given to attaching such safeguards to WLST. Defences of going in the opposite direction and rejecting special safeguards for both WLST and PAD are provided by Glanville Williams in The Sanctity of life and the Criminal Law (1957; pp. 339-46 of that publication) and James Rachels in The End of life (1986; Chapter 10 of that publication). These works are cited in the Suggested Further Reading at the end of this chapter.

Conscientious Objection The Scc acknowledged that physicians have a right to refuse to provide PAD if doing so conflicts with their moral values. The question is whether they have to do anything to help those who make the request. Do they have to refer the patient to a physician who will provide PAD, as they would in the case of abortion? If they do not have to refer, is it enough to inform the patient that other physicians will provide PAD, or do they have to also tell him or her how to contact those physicians? Or should there be someone whom an objecting physician must inform and who has an obligation to ensure a referral? Similar questions arise with respect to the obligations of publicly funded institutions that have objections to PAD. Must they allow it on their premises? Or can they transfer patients to another hospital for the procedure? Bill C-14 recommends that the Criminal Code be amended (as it now has been) to specify that no one can be compelled to provide MAID [241.2(9)] but leaves it to physicians' colleges, the federal Parliament, and the provincial legislatures to grapple with the issues of conscientious objection. The question remains unsettled.

Defining Death

Defining Death This chapter has been concerned with the question of when it is appropriate to bring about death. It is thus natural to want to know when death occurs and how it is to be defined. Alister Browne takes up these questions in "Defining Death" (see page 339) and argues that 'when death occurs" is not a time that is discovered (like when the Great Pyramid was built) but rather (like the age of maturity) a time that is decided. On this view, the line between life and death is not laid down by God or nature but decided by human beings for human purposes. Thus, lines that are drawn between life and death at one time can be redrawn in another way at another time. Or, if it seems best (as it does to some), these lines can be erased altogether in favour of descriptions of what can be done to whom and when without using the word "death" at all If so, the questions of PAD and defining death have much in common. Browne's article provides a brief introduction to this topic and an annotated bibliography for those who want to pursue it further. Notes 1. 2. 3. 4. 5. 6. 7. 8.

See Bernard M. Dickens, "Terminal Care and Related Decisions: A Review o f Legal Developments," MODERN MEDICINE of Canada 45 (1990) (4). Carter v Canada (Attorney General), 2015 SCC5. Carter v Canada (Attorney General), 2015 scc5, para. 23. Quoting from the trial judge in Carter v Canada (Attorney General), 2012, para. 335. Jeremy Bentham, The Principles of Morals and Legislation (1823), ch. VIII, sec. IV. Bill C-14, Statutes of Canada 2016, Ch. 3, An Act to Amend the Criminal Code and to Make Related Amendments to Other Acts (Medical Assistance in Dying), Sec. 241.2 (2). Carter v Canada (Attorney General), 2015 SCC5, para. 105. Bill C-14, Statutes of Canada 2016, Ch. 3, An Act to Amend the Criminal Code and to Make Related Amendments to Other Acts (Medical Assistance in Dying), Sec. 241.2 (3), Summary of Key Points. Carter v Canada (Attorney General), 2015 SCC5, para. 115.

6.2 Withholding or Withdrawing Life-Sustaining Treatment and Physician-Assisted Death

Withholding and Withdrawal of Potentially Life-Sustaining Treatment

What Does "Withdrawal of Potentially Life-Sustaining Treatment" Mean?

Health Law Institute, Dalhousie University

Withdrawal of potentially life-sustaining treatment means stopping treatment that has the potential to sustain the life of a patient—for example, stopping ventilatory support for a patient in a persistent vegetative state.

What Does "Withholding of Potentially Life-Sustaining Treatment" Mean? Withholding potentially life-sustaining treatment means not starting treatment that has the potential to sustain the life of a patient—for example, not attempting cardiopulmonary resuscitation (CPR) when a patient's heart suddenly stops beating.

Is the Withholding and Withdrawal of Potentially Life-Sustaining Treatment Legal in Canada? Yes, both canb e legal. In the 1993 Supreme Court of Canada case, Rodriguez v. British Columbia (Attorney General), Justice

Source: From http://asOLucis.dal.ca/dhli/cmp_documents/documents/WW_Brochure.pdf (2017). Reprinted by permission of the author.

II

:111

111111

6 End-of-Life Decision-Making

Sopinka, writing for a majority of the Supreme Court of Canada, made three statements to the effect that there is a common law right to refuse even potentially lifesustaining treatment: . . . That there is a right to choose how one's body will be dealt with, even in the context of beneficial medical treatment, has long been recognized by the common law. To impose medical treatment on one who refuses constitutes battery, and our common law has recognized the right to demand that medical treatment which would extend life be withheld or withdrawn. [. ..] Canadian courts have recognized a common law right of patients to refuse consent to medical treatment, or to demand that treatment, once commenced, be withdrawn or discontinued (Ciarlariello v. Schacter, [1993] 2 S.C.R. 119). This right has been specifically recognized to exist even if the withdrawal from or refusal of treatment may result in death (Nancy B. v. Hotel-Dieu de Quebec (1992), 86 D.L.R. (4th) 385 (Que. S.C.); Malette v. Shulman (1990), 72 O.R. (2d) 417 (C.A.)). . . . ]. • • Whether or not one agrees that the active vs. passive distinction is maintainable, however, the fact remains that under our common law, the physician has no choice but to accept the patient's instructions to discontinue treatment. To continue to treat the patient when the patient has withdrawn consent to that treatment constitutes battery (Ciarlariello and Nancy B., supra). . . . More recently, Justice Smith characterized withholding and withdrawal of potentially life-sustaining treatment as "conventionally ethical end-of-life practice[s]" (at 186) (Carter v Canada (Attorney General), 2012 BCSC 886). Must Healthcare Providers Respect Refusals Made by Patients or Their Substitute Decision-Makers?

The answer to this question depends upon several variables. The Patient Is Competent

If the patient is competent and making a free and informed refusal, health care providers must respect this refusal.

The Patient Is Incompetent but Has a Valid Advance Directive

If the patient has expressed a treatment refusal within a valid advance directive, healthcare providers must respect this refusal. If the patient has appointed a proxy through a proxy directive, health care providers must respect a free and informed refusal by the proxy (unless the healthcare provider believes the proxy is not acting in accordance with their powers under the proxy directive). Provincial legislation dictates the requirements for valid advance directives. The Patient Is a Mature Minor

If the patient is a minor who understands and appreciates the consequences of the decision to be made and has sufficient independence, under common law their decision must be respected. This common law rule can be superseded by statute (as in British Columbia, Alberta, Saskatchewan, Manitoba and Newfoundland and Labrador through their child protection statutes). However, in accordance with the Supreme Court of Canada's ruling in A.C. v Manitoba (Director of Child and Family Services), these statutes should be interpreted such that acting in the minor's "best interests" under the legislation is understood to include respect for the minor's decision-making if the minor is sufficiently mature and independent. In the context of the withholding and withdrawal of potentially life-sustaining treatment, the scrutiny that the court will apply to the minors maturity and independence will be high given the seriousness of the decision's potential impact on the minor's life or health. In provinces and territories without such statutes, the common law mature minor rule will prevaiL The Patient Is an Immature Minor

If the patient is an immature minor, then the healthcare provider must respect a substitute decision-maker's refusal to consent to treatment unless the health care provider believes that the substitute decision-maker is not acting in the patient's best interests. In such a case, the healthcare provider should seek an order to treat from the courts. Do Healthcare Providers Have the Authority to Withhold or Withdraw Potentially LifeSustaining Treatment against the Wishes or Without the Knowledge of a Patient or Their Substitute Decision-Maker?

The law in Canada on the unilateral withholding and withdrawal of potentially life-sustaining treatment by a healthcare provider is not yet clear. Some courts have held that healthcare providers do not have the authority

Brock: Voluntary Active Euthanasia

to unilaterally withhold or withdraw potentially lifesustaining treatment, others have said that they do, and still others have said that the law is unsettled or they have remained silent on the issue. On October 18, 2013 some clarity was brought to the issue when the Supreme Court of Canada released its decision in Brian Cuthbertson, et at v Hassan Rasouli by his Litigation Guardian and Substitute Decision Maher, Parichehr Salasel. Here, a majority of the Scc found that

Ontario's health care consent legislation requires consent from a patient's substitute decision-maker prior to the withdrawal of potentially life-sustaining treatment. Unfortunately, this decision provides little guidance for provinces and territories other than Ontario as it was based on an interpretation of the Ontario legislation (which others do not share). More litigation, or preferably action by provincial legislatures, is needed to resolve the ongoing confusion and controversy.

Voluntary Active Euthanasia

proposal to legalize both voluntary active euthanasia and physician-assisted suicide. Finally, some cases of this kind, such as "It's Over, Debbie," described in the Journal of the American Medical Association, the "suicide machine" of Dr Jack Kevorkian, and the cancer patient "Diane" of Dr Timothy Quill, have captured wide public and professional attention.3 Unfortunately, the first two of these cases were sufficiently problematic that even most supporters of euthanasia or assisted suicide did not defend the physicians' actions in them. As a result, the subsequent debate they spawned has often shed more heat than light. My aim is to increase the light, and perhaps as well to reduce the heat, on this important subject by formulating and evaluating the central ethical arguments for and against voluntary active euthanasia and physician-assisted suicide. My evaluation of the arguments leads me, with reservations to be noted, to support permitting both practices. My primary aim, however, is not to argue for euthanasia, but to identify confusions in some common arguments, and problematic assumptions and claims that need more defence or data in others. The issues are considerably more complex than either supporters or opponents often make out; my hope is to advance the debate by focusing attention on what I believe the real issues under discussion should be. In the recent bioethics literature some have endorsed physician-assisted suicide but not euthanasia.4 Are they sufficiently different that the moral arguments for one often do not apply to the other? A paradigm case of physician-assisted suicide is a patient's ending his or her life with a lethal dose of a medication requested of and provided by a physician for that purpose. A paradigm case of voluntary active euthanasia is a physician's administering the lethal dose, often because the patient is unable to do so. The only difference that need exist between the two is the person who actually administers the lethal dose—the physician or

Dan W. Brock

Since the case of Karen Quinlan first seized public attention 15 years ago, no issue in biomedical ethics has been more prominent than the debate about forgoing life-sustaining treatment. Controversy continues regarding some aspects of that debate, such as forgoing life-sustaining nutrition and hydration, and relevant law varies some from state to state. Nevertheless, I believe it is possible to identify an emerging consensus that competent patients, or the surrogates of incompetent patients, should be permitted to weigh the benefits and burdens of alternative treatments, including the alternative of no treatment, according to the patient's values, and either to refuse any treatment or to select from among available alternative treatments. This consensus is reflected in bioethics scholarship, in reports of prestigious bodies such as the President's Commission for the Study of Ethical Problems in Medicine, The Hastings Center, and the American Medical Association, in a large body of judicial decisions in courts around the country, and finally in the beliefs and practices of health care professionals who care for dying patients.' More recently, significant public and professional attention has shifted from life-sustaining treatment to euthanasia—more specifically, voluntary active euthanasia—and to physician-assisted suicide. Several factors have contributed to the increased interest in euthanasia. In the Netherlands, it has been openly practised by physicians for several years with the acceptance of the country's highest court.' In 1988 there was an unsuccessful attempt to get the question of whether it should be made legally permissible on the ballot in California. In November 1991 voters in the state of Washington defeated a widely publicized referendum

Source: From The Hastings Center Report 22 (1992) (2), 10-22. Reprinted by permission of The Hastings Center and the author.

II .31

111111

6 End-of-Life Decision-Making

the patient. In each, the physician plays an active and necessary causal role. In physician-assisted suicide the patient acts last (for example, Janet Adkins herself pushed the button after Dr Kevorkian hooked her up to his suicide machine), whereas in euthanasia the physician acts last by performing the physical equivalent of pushing the button. In both cases, however, the choice rests fully with the patient. In both the patient acts last in the sense of retaining the right to change his or her mind until the point at which the lethal process becomes irreversible. How could there be a substantial moral difference between the two based only on this small difference in the part played by the physician in the causal process resulting in death? Of course, it might be held that the moral difference is clear and important—in euthanasia the physician kills the patient whereas in physicianassisted suicide the patient kills him- or herself. But this is misleading at best. In assisted suicide the physician and patient together kill the patient. To see this, suppose a physician supplied a lethal dose to a patient with the knowledge and intent that the patient will wrongfully administer it to another. We would have no difficulty in morality or the law recognizing this as a case of joint action to kill for which both are responsible. If there is no significant, intrinsic moral difference between the two, it is also difficult to see why public or legal policy should permit one but not the other, worries about abuse or about giving anyone dominion over the lives of others apply equally to either. As a result, I will take the arguments evaluated below to apply to both and will focus on euthanasia. My concern here will be with voluntary euthanasia only—that is, with the case in which a clearly competent patient makes a fully voluntary and persistent request for aid in dying. Involuntary euthanasia, in which a competent patient explicitly refuses or opposes receiving euthanasia, and nonvoluntary euthanasia, in which a patient is incompetent and unable to express his or her wishes about euthanasia, will be considered here only as potential unwanted side effects of permitting voluntary euthanasia. I emphasize as well that I am concerned with active euthanasia, not withholding or withdrawing lifesustaining treatment, which some commentators characterize as "passive euthanasia." Finally, I will be concerned with euthanasia where the motive of those who perform it is to respect the wishes of the patient and to provide the patient with a "good death," though one important issue is whether a change in legal policy could restrict the performance of euthanasia to only those cases. A last introductory point is that I will be examining only secular arguments about euthanasia, though of course many people's attitudes to it are inextricable

from their religious views. The policy issue is only whether euthanasia should be permissible, and no one who has religious objections to it should be required to take any part in it, though of course this would not fully satisfy some opponents. The Central Ethical Argument for Voluntary Active Euthanasia

The central ethical argument for euthanasia is familiar. It is that the very same two fundamental ethical values supporting the consensus on patient's rights to decide about life-sustaining treatment also support the ethical permissibility of euthanasia. These values are individual self-determination or autonomy and individual well-being. By self-determination as it bears on euthanasia, I mean people's interest in making important decisions about their lives for themselves according to their own values or conceptions of a good life, and in being left free to act on those decisions. Selfdetermination is valuable because it permits people to form and live in accordance with their own conception of a good life, at least within the bounds of justice and consistent with others doing so as well. In exercising self-determination people take responsibility for their lives and for the kinds of persons they become. A central aspect of human dignity lies in people's capacity to direct their lives in this way. The value of exercising self-determination presupposes some minimum of decision-making capacities or competence, which thus limits the scope of euthanasia supported by selfdetermination; it cannot justifiably be administered, for example, in cases of serious dementia or treatable clinical depression. Does the value of individual self-determination extend to the time and manner of one's death? Most people are very concerned about the nature of the last stage of their lives. This reflects not just a fear of experiencing substantial suffering when dying, but also a desire to retain dignity and control during this last period of life. Death is today increasingly preceded by a long period of significant physical and mental decline, due in part to the technological interventions of modern medicine. Many people adjust to these disabilities and find meaning and value in new activities and ways. Others find the impairments and burdens in the last stage of their lives at some point sufficiently great to make life no longer worth living. For many patients near death, maintaining the quality of one's life, avoiding great suffering, maintaining one's dignity, and, insuring that others remember us as we wish them to become of paramount importance and outweigh merely extending one's life. But there is no single,

Brock: Voluntary Active Euthanasia 299

objectively correct answer for everyone as to when, if at all, one's life becomes all things considered a burden and unwanted. If self-determination is a fundamental value, then the great variability among people on this question makes it especially important that individuals control the manner, circumstances, and timing of their dying and death. The other main value that supports euthanasia is individual well-being. It might seem that individual well-being conflicts with a person's self-determination when the person requests euthanasia. Life itself is commonly taken to be a central good for persons, often valued for its own sake, as well as necessary for pursuit of all other goods within a life. But when a competent patient decides to forgo all further life-sustaining treatment then the patient, either explicitly or implicitly, commonly decides that the best life possible for him or her with treatment is of sufficiently poor quality that it is worse than no further life at all. Life is no longer considered a benefit by the patient, but has now become a burden. The same judgment underlies a request for euthanasia: continued life is seen by the patient as no longer a benefit, but now a burden. Especially in the often severely compromised and debilitated states of many critically ill or dying patients, there is no objective standard, but only the competent patient's judgment of whether continued life is no longer a benefit. Of course, sometimes there are conditions, such as clinical depression, that call into question whether the patient has made a competent choice, either to forgo life-sustaining treatment or to seek euthanasia, and then the patient's choice need not be evidence that continued life is no longer a benefit for him or her. Just as with decisions about treatment, a determination of incompetence can warrant not honouring the patient's choice; in the case of treatment, we then transfer decisional authority to a surrogate, though in the case of voluntary active euthanasia a determination that the patient is incompetent means that choice is not possible. The value or right of self-determination does not entitle patients to compel physicians to act contrary to their own moral or professional values. Physicians are moral and professional agents whose own selfdetermination or integrity should be respected as well. If performing euthanasia became legally permissible, but conflicted with a particular physician's reasonable understanding of his or her moral or professional responsibilities, the care of a patient who requested euthanasia should be transferred to another. Most opponents do not deny that there are some cases in which the values of patient self-determination and well-being support euthanasia Instead, they commonly offer two kinds of arguments against it that in

their view outweigh or override this support. The first kind of argument is that in any individual case where considerations of the patient's self-determination and well-being do support euthanasia, it is nevertheless always ethically wrong or impermissible. The second kind of argument grants that in some individual cases euthanasia may not be ethically wrong, but maintains nonetheless that public and legal policy should never permit it. The first kind of argument focuses on features of any individual case of euthanasia, while the second kind focuses on social or legal policy. In the next section I consider the first kind of argument. Euthanasia Is the Deliberate Killing of an Innocent Person

The claim that any individual instance of euthanasia is a case of deliberate killing of an innocent person is, with only minor qualifications, correct. Unlike forgoing life-sustaining treatment, commonly understood as allowing to die, euthanasia is clearly killing, defined as depriving of life or causing the death of a living being. While providing morphine for pain relief at doses where the risk of respiratory depression and an earlier death may be a foreseen but unintended side effect of treating the patient's pain, in a case of euthanasia the patient's death is deliberate or intended even if in both the physician's ultimate end may be respecting the patient's wishes. If the deliberate killing of an innocent person is wrong, euthanasia would be nearly always impermissible. In the context of medicine, the ethical prohibition against deliberately killing the innocent derives some of its plausibility from the belief that nothing in the currently accepted practice of medicine is deliberate killing. Thus, in commenting on the "It's Over, Debbie" case, four prominent physicians and bioethicists could entitle their paper "Doctors Must Not Kill."5 The belief that doctors do not in fact kill requires the corollary belief that forgoing life-sustaining treatment, whether by not starting or by stopping treatment, is allowing to die, not killing. Common though this view is, I shall argue that it is confused and mistaken. Why is the common view mistaken? Consider the case of a patient terminally ill with ALS disease. She is completely respirator-dependent with no hope of ever being weaned. She is unquestionably competent, but finds her condition intolerable and persistently requests to be removed from the respirator and allowed to die. Most people and physicians would agree that the patient's physician should respect the patient's wishes and remove her from the respirator, though this will certainly cause the patient's death. The common

II ill

111111i1

6 End-of-Life Decision-Making

understanding is that the physician thereby allows the patient to die. But is that correct? Suppose the patient has a greedy and hostile son who mistakenly believes that his mother will never decide to stop her life-sustaining treatment and that even if she did her physician would not remove her from the respirator. Afraid that his inheritance will be dissipated by a long and expensive hospitalization, he enters his mother's room while she is sedated, extubates her, and she dies. Shortly thereafter the medical staff discovers what he has done and confronts the son. He replies, "I didn't kill her, I merely allowed her to die. It was her ALS disease that caused her death." I think this would rightly be dismissed as transparent sophistry—the son went into his mother's room and deliberately killed her. But, of course, the son performed just the same physical actions, did just the same thing, that the physician would have done. If that is so, then doesn't the physician also kill the patient when he extubates her? I underline immediately that there are important ethical differences between what the physician and the greedy son do. First, the physician acts with the patient's consent whereas the son does not. Second, the physician acts with a good motive—to respect the patient's wishes and self-determination—whereas the son acts with a bad motive—to protect his own inheritance. Third, the physician acts in a social role through which he is legally authorized to carry out the patient's wishes regarding treatment whereas the son has no such authorization. These and perhaps other ethically important differences show that what the physician did was morally justified whereas what the son did was morally wrong. What they do not show, however, is that the son killed while the physician allowed to die. One can either kill or allow to die with or without consent, with a good or bad motive, within or outside of a social role that authorizes one to do so. The difference between killing and allowing to die that I have been implicitly appealing to here is roughly that between acts and omissions resulting in death.° Both the physician and the greedy son act in a manner intended to cause death, do cause death, and so both kill. One reason this conclusion is resisted is that on a different understanding of the distinction between killing and allowing to die, what the physician does is allow to die. In this account, the mother's ALS is a lethal disease whose normal progression is being held back or blocked by the life-sustaining respirator treatment. Removing this artificial intervention is then viewed as standing aside and allowing the patient to die of her underlying disease. I have argued elsewhere that this alternative account is deeply problematic, in part because it commits us to accepting that what the greedy

son does is to allow to die, not kill.' Here, I want to note two other reasons why the conclusion that stopping life support is killing is resisted. The first reason is that killing is often understood, especially within medicine, as unjustified causing of death; in medicine it is thought to be done only accidentally or negligently. It is also increasingly widely accepted that a physician is ethically justified in stopping life support in a case like that of the ALS patient. But if these two beliefs are correct, then what the physician does cannot be killing, and so must be allowing to die. Killing patients is not, to put it flippantly, understood to be part of physicians' job description. What is mistaken in this line of reasoning is the assumption that all killings are unjustified causings of death. Instead, some killings are ethically justified, including many instances of stopping life support. Another reason for resisting the conclusion that stopping life support is often killing is that it is psychologically uncomfortable. Suppose the physician had stopped the ALS patient's respirator and had made the son's claim, "I didn't kill her, I merely allowed her to die. It was her ALS disease that caused her death." The clue to the psychological role here is how naturally the "merely" modifies "allowed her to die." The characterization as allowing to die is meant to shift felt responsibility away from the agent—the physician—and to the lethal disease process. Other language common in death and dying contexts plays a similar role; "letting nature take its course" or "stopping prolonging the dying process" both seem to shift responsibility from the physician who stops life support to the fatal disease process. However psychologically helpful these conceptualizations may be in making the difficult responsibility of a physician's role in the patient's death bearable, they nevertheless are confusions. Both physicians and family members can instead be helped to understand that it is the patient's decision and consent to stopping treatment that limits their responsibility for the patient's death and that shifts that responsibility to the patient. Many who accept the difference between killing and allowing to die as the distinction between acts and omissions resulting in death have gone on to argue that killing is not in itself morally different from allowing to die.° In this account, very roughly, one kills when one performs an action that causes the death of a person (we are in a boat, you cannot swim, I push you overboard, and you drown), and one allows to die when one has the ability and opportunity to prevent the death of another, knows this, and omits doing so, with the result that the person dies (we are in a boat, you cannot swim, you fall overboard, I don't throw you an available life ring, and you drown). Those who see

Brock: Voluntary Active Euthanasia no moral difference between killing and allowing to die typically employ the strategy of comparing cases that differ in these and no other potentially morally important respects. This will allow people to consider whether the mere difference that one is a case of killing and the other of allowing to die matters morally, or whether instead it is other features that make most cases of killing worse than most instances of allowing to die. Here is such a pair of cases: Case 1. A very gravely ill patient is brought to a hospital emergency room and sent up to the icu. The patient begins to develop respiratory failure that is likely to require intubation very soon. At that point the patient's family members and longstanding physician arrive at the icu and inform the Ku staff that there had been extensive discussion about future care with the patient when he was unquestionably competent. Given his grave and terminal illness, as well as his state of debilitation, the patient had firmly rejected being placed on a respirator under any circumstances, and the family and physician produce the patient's advance directive to that effect. The icu staff do not intubate the patient, who dies of respiratory failure. Case 2. The same as Case 1 except that the family and physician are slightly delayed in traffic and arrive shortly after the patient has been intubated and placed on the respirator. The icu staff extubate the patient, who dies of respiratory failure. In Case 1 the patient is allowed to die, in Case 2 he is killed, but it is hard to see why what is done in Case 2 is significantly different morally than what is done in Case 1. It must be other factors that make most killings worse than most allowings to die, and if so, euthanasia cannot be wrong simply because it is killing instead of allowing to die. Suppose both my arguments are mistaken. Suppose that killing is worse than allowing to die and that withdrawing life support is not killing, although euthanasia is. Euthanasia still need not for that reason be morally wrong. To see this, we need to determine the basic principle for the moral evaluation of killing persons. What is it that makes paradigm cases of wrongful killing wrongful? One very plausible answer is that killing denies the victim something that he or she values greatly—continued life or a future. Moreover, since continued life is necessary for pursuing any of a person's plans and purposes, killing brings the frustration of all of these plans and desires as well. In a nutshell, wrongful killing deprives a person of

a valued future, and of all the person wanted and planned to do in that future. A natural expression of this account of the wrongness of killing is that people have a moral right not to be killed.9 But in this account of the wrongness of killing, the right not to be killed, like other rights, should be waivable when the person makes a competent decision that continued life is no longer wanted or a good, but is instead worse than no further life at all. In this view, euthanasia is properly understood as a case of a person having waived his or her right not to be killed. This rights view of the wrongness of killing is not, of course, universally shared. Many people's moral views about killing have their origins in religious views that human life comes from God and cannot be justifiably destroyed or taken away, either by the person whose life it is or by another. But in a pluralistic society like our own, with a strong commitment to freedom of religion, public policy should not be grounded in religious beliefs which many in that society reject. I turn now to the general evaluation of public policy on euthanasia. Would the Bad Consequences of Euthanasia Outweigh the Good?

The argument against euthanasia at the policy level is stronger than at the level of individual cases, though even here I believe the case is ultimately unpersuasive, or at best indecisive. The policy level is the place where the main issues lie, however, and where moral considerations that might override arguments in favour of euthanasia will be found, if they are found anywhere. It is important to note two kinds of disagreement about the consequences for public policy of permitting euthanasia. First, there is empirical or factual disagreement about what the consequences would be. This disagreement is greatly exacerbated by the lack of firm data on the issue. Second, since on any reasonable assessment there would be both good and bad consequences, there are moral disagreements about the relative importance of different effects. In addition to these two sources of disagreement, there is also no single, well-specified policy proposal for legalizing euthanasia on which policy assessments can focus. But without such specification, and especially without explicit procedures for protecting against wellintentioned misuse and ill-intentioned abuse, the consequences for policy are largely speculative. Despite these difficulties, a preliminary account of the main likely good and bad consequences is possible. This should help clarify where better data or more moral analysis and argument are needed, as well as where policy safeguards must be developed.

II j11

IIILI

6 End-of-Life Decision-Making

Potential Good Consequences of Permitting Euthanasia

What are the likely good consequences? First, if euthanasia were permitted it would be possible to respect the self-determination of competent patients who want it, but now cannot get it because of its illegality. We simply do not know how many such patients and people there are. In the Netherlands, with a population of about 14.5 million (in 1987), estimates in a recent study were that about 1,900 cases of voluntary active euthanasia or physician-assisted suicide occur annually. No straightforward extrapolation to the United States is possible for many reasons, among them, that we do not know how many people here who want euthanasia now get it, despite its illegality. Even with better data on the number of persons who want euthanasia but cannot get it, significant moral disagreement would remain about how much weight should be given to any instance of failure to respect a person's self-determination in this way. One important factor substantially affecting the number of persons who would seek euthanasia is the extent to which an alternative is available. The widespread acceptance in the law, social policy, and medical practice of the right of a competent patient to forgo life-sustaining treatment suggests that the number of competent persons in the United States who would want euthanasia if it were permitted is probably relatively small. A second good consequence of making euthanasia legally permissible benefits a much larger group. Polls have shown that a majority of the American public believes that people should have a right to obtain euthanasia if they want it.m No doubt the vast majority of those who support this right to euthanasia will never in fact come to want euthanasia for themselves. Nevertheless, making it legally permissible would reassure many people that if they ever do want euthanasia they would be able to obtain it. This reassurance would supplement the broader control over the process of dying given by the right to decide about life-sustaining treatment. Having fire insurance on one's house benefits all who have it, not just those whose houses actually burn down, by reassuring them that in the unlikely event of their house burning down, they will receive the money needed to rebuild it. Likewise, the legalization of euthanasia can be thought of as a kind of insurance policy against being forced to endure a protracted dying process that one has come to find burdensome and unwanted, especially when there is no life-sustaining treatment to forgo. The strong concern about losing control of their care expressed by many people who face serious illness likely to end in death suggests that they give substantial importance to the legalization of euthanasia as a means of maintaining this control.

A third good consequence of the legalization of euthanasia concerns patients whose dying is filled with severe and unrelievable pain or suffering. When there is a life-sustaining treatment that, if forgone, will lead relatively quickly to death, then doing so can bring an end to these patients' suffering without recourse to euthanasia. For patients receiving no such treatment, however, euthanasia may be the only release from their otherwise prolonged suffering and agony. This argument from mercy has always been the strongest argument for euthanasia in those cases to which it applies." The importance of relieving pain and suffering is less controversial than is the frequency with which patients are forced to undergo untreatable agony that only euthanasia could relieve. If we focus first on suffering caused by physical pain, it is crucial to distinguish pain that could be adequately relieved with modern methods of pain control, though it in fact is not, from pain that is relievable only by death." For a variety of reasons, including some physicians' fear of hastening the patient's death, as well as the lack of a publicly accessible means for assessing the amount of the patient's pain, many patients suffer pain that could be, but is not, relieved. Specialists in pain control, as for example the pain of terminally ill cancer patients, argue that there are very few patients whose pain could not be adequately controlled, though sometimes at the cost of so sedating them that they are effectively unable to interact with other people or their environment. Thus, the argument from mercy in cases of physical pain can probably be met in a large majority of cases by providing adequate measures of pain relief. This should be a high priority, whatever our legal policy on euthanasia—the relief of pain and suffering has long been, quite properly, one of the central goals of medicine. Those cases in which pain could be effectively relieved, but in fact is not, should only count significantly in favour of legalizing euthanasia if all reasonable efforts to change pain management techniques have been tried and have failed. Dying patients often undergo substantial psychological suffering that is not fully or even principally the result of physical pain." The knowledge about how to relieve this suffering is much more limited than in the case of relieving pain, and efforts to do so are probably more often unsuccessful. If the argument from mercy is extended to patients experiencing great and unrelievable psychological suffering, the numbers of patients to which it applies are much greater. One last good consequence of legalizing euthanasia is that once death has been accepted, it is often more humane to end life quickly and peacefully, when that is what the patient wants. Such a death will often be

Brock: Voluntary Active Euthanasia

seen as better than a more prolonged one. People who suffer a sudden and unexpected death, for example by dying quickly or in their sleep from a heart attack or stroke, are often considered lucky to have died in this way. We care about how we die in part because we care about how others remember us, and we hope they will remember us as we were in "good times" with them and not as we might be when disease has robbed us of our dignity as human beings. As with much in the treatment and care of the dying, people's concerns differ in this respect, but for at least some people, euthanasia will be a more humane death than what they have often experienced with other loved ones and might otherwise expect for themselves. Some opponents of euthanasia challenge how much importance should be given to any of these good consequences of permitting it, or even whether some would be good consequences at all. But more frequently, opponents cite a number of bad consequences that permitting euthanasia would or could produce, and it is to their assessment that I now turn. Potential Bad Consequences of Permitting Euthanasia

Some of the arguments against permitting euthanasia are aimed specifically against physicians, while others are aimed against anyone being permitted to perform it. I shall first consider one argument of the former sort. Permitting physicians to perform euthanasia, it is said, would be incompatible with their fundamental moral and professional commitment as healers to care for patients and to protect life. Moreover, if euthanasia by physicians became common, patients would come to fear that a medication was intended not to treat or care, but instead to kill, and would thus lose trust in their physicians. This position was forcefully stated in a paper by Willard Gaylin and his colleagues: The very soul of medicine is on trial. . . . This issue touches medicine at its moral center, if this moral center collapses, if physicians become killers or are even licensed to kill, the profession—and, therewith, each physician—will never again be worthy of trust and respect as healer and comforter and protector of life in all its frailty. These authors go on to make clear that, while they oppose permitting anyone to perform euthanasia, their special concern is with physicians doing so: We call on fellow physicians to say that they will not deliberately kill. We must also say to each of our fellow physicians that we will not tolerate

killing of patients and that we shall take disciplinary action against doctors who kill. And we must say to the broader community that if it insists on tolerating or legalizing active euthanasia, it will have to find nonphysicians to do its killing.14 If permitting physicians to kill would undermine the very "moral centre" of medicine, then almost certainly physicians should not be permitted to perform euthanasia. But how persuasive is this claim? Patients should not fear, as a consequence of permitting voluntary active euthanasia, that their physicians will substitute a lethal injection for what patients want and believe is part of their care. If active euthanasia is restricted to cases in which it is truly voluntary, then no patient should fear getting it unless she or he has voluntarily requested it. (The fear that we might in time also come to accept nonvoluntary, or even involuntary active euthanasia is a slippery slope worry I address below.) Patients' trust of their physicians could be increased, not eroded, by knowledge that physicians will provide aid in dying when patients seek it. Might Gaylin and his colleagues nevertheless be correct in their claim that the moral centre of medicine would collapse if physicians were to become killers? This question raises what at the deepest level should be the guiding aims of medicine, a question that obviously cannot be fully explored here. But I do want to say enough to indicate the direction that I believe an appropriate response to this challenge should take. In spelling out above what I called the positive argument for voluntary active euthanasia, I suggested that two principal values—respecting patients' self-determination and promoting their well-being—underlie the consensus that competent patients, or the surrogates of incompetent patients, are entitled to refuse any life-sustaining treatment and to choose from among available alternative treatments. It is the commitment to these two values in guiding physicians' actions as healers, comforters, and protectors of their patients' lives that should be at the "moral centre" of medicine, and these two values support physicians' administering euthanasia when their patients make competent requests for it. What should not be at that moral centre is a commitment to preserving patients' lives as such, without regard to whether those patients want their lives preserved or judge their preservation a benefit to them. Vitalism has been rejected by most physicians, and despite some statements that suggest it, is almost certainly not what Gaylin and colleagues intended. One of them, Leon Kass, has elaborated elsewhere the view that medicine is a moral profession whose proper aim is "the naturally given end of health," understood as

Al

II 111

111111

6 End-of-Life Decision-Making

the wholeness and well-working of the human being "for the physician, at least, human life in living bodies commands respect and reverence—by its very nature." Kass continues, "the deepest ethical principle restraining the physician's power is not the autonomy or freedom of the patient; neither is it his own compassion or good intention. Rather, it is the dignity and mysterious power of human life itself."" I believe Kass is in the end mistaken about the proper account of the aims of medicine and the limits on physicians' power, but this difficult issue will certainly be one of the central themes in the continuing debate about euthanasia. A second bad consequence that some foresee is that permitting euthanasia would weaken society's commitment to provide optimal care for dying patients. We live at a time in which the control of health care costs has become, and is likely to continue to be, the dominant focus of health care policy. If euthanasia is seen as a cheaper alternative to adequate care and treatment, then we might become less scrupulous about providing sometimes costly support and other services to dying patients. Particularly if our society comes to embrace deeper and more explicit rationing of health care, frail, elderly, and dying patients will need to be strong and effective advocates for their own health care and other needs, although they are hardly in a position to do this. We should do nothing to weaken their ability to obtain adequate care and services. This second worry is difficult to assess because there is little firm evidence about the likelihood of the feared erosion in the care of dying patients. There are at least two reasons, however, for skepticism about this argument. The first is that the same worry could have been directed at recognizing patients' or surrogates' rights to forgo life-sustaining treatment, yet there is no persuasive evidence that recognizing the right to refuse treatment has caused a serious erosion in the quality of care of dying patients. The second reason for skepticism about this worry is that only a very small proportion of deaths would occur from euthanasia if it were permitted. In the Netherlands, where euthanasia under specified circumstances is permitted by the courts, though not authorized by statute, the best estimate of the proportion of overall deaths that result from it is about 2 per cent.16 Thus, the vast majority of critically ill and dying patients will not request it, and so will still have to be cared for by physicians, families, and others. Permitting euthanasia should not diminish people's commitment and concern to maintain and improve the care of these patients. A third possible bad consequence of permitting euthanasia (or even a public discourse in which strong support for euthanasia is evident) is to threaten the

progress made in securing the rights of patients or their surrogates to decide about and to refuse life-sustaining treatment." This progress has been made against the backdrop of a clear and firm legal prohibition of euthanasia, which has provided a relatively bright line limiting the dominion of others over patients' lives. It has therefore been an important reassurance to concerns about how the authority to take steps ending life might be misused, abused, or wrongly extended. Many supporters of the right of patients or their surrogates to refuse treatment strongly oppose euthanasia, and if forced to choose might well withdraw their support of the right to refuse treatment rather than accept euthanasia. Public policy in the last 15 years has generally let life-sustaining treatment decisions be made in health care settings between physicians and patients or their surrogates, and without the involvement of the courts. However, if euthanasia is made legally permissible greater involvement of the courts is likely, which could in turn extend to a greater court involvement in life-sustaining treatment decisions. Most agree, however, that increased involvement of the courts in these decisions would be undesirable, as it would make sound decision-making more cumbersome and difficult without sufficient compensating benefits. As with the second potential bad consequence of permitting euthanasia, this third consideration too is speculative and difficult to assess. The feared erosion of patients' or surrogates' rights to decide about life-sustaining treatment, together with greater court involvement in those decisions, are both possible. However, I believe there is reason to discount this general worry. The legal rights of competent patients and, to a lesser degree, surrogates of incompetent patients to decide about treatment are very firmly embedded in a long line of informed consent and life-sustaining treatment cases, and are not likely to be eroded by a debate over, or even acceptance of, euthanasia It will not be accepted without safeguards that reassure the public about abuse, and if that debate shows the need for similar safeguards for some life-sustaining treatment decisions they should be adopted there as well. In neither case are the only possible safeguards greater court involvement, as the recent growth of institutional ethics committees shows. The fourth potential bad consequence of permitting euthanasia has been developed by David Velleman and turns on the subtle point that making a new option or choice available to people can sometimes make them worse off, even if once they have the choice they go on to choose what is best for them." Ordinarily, people's continued existence is viewed by them as given, a fixed condition with which they must cope. Making

Brock: Voluntary Active Euthanasia

euthanasia available to people as an option denies them the alternative of staying alive by default. If people are offered the option of euthanasia, their continued existence is now a choice for which they can be held responsible and which they can be asked by others to justify. We care, and are right to care, about being able to justify ourselves to others. To the extent that our society is unsympathetic to justifying a severely dependent or impaired existence, a heavy psychological burden of proof may be placed on patients who think their terminal illness or chronic infirmity is not a sufficient reason for dying. Even if they otherwise view their life as worth living, the opinion of others around them that it is not can threaten their reason for living and make euthanasia a rational choice. Thus the existence of the option becomes a subtle pressure to request it. This argument correctly identifies the reason why offering some patients the option of euthanasia would not benefit them. Velleman takes it not as a reason for opposing all euthanasia, but for restricting it to circumstances where there are "unmistakable and overpowering reasons for persons to want the option of euthanasia," and for denying the option in all other cases. But there are at least three reasons why such restriction may not be warranted. First, polls and other evidence support that most Americans believe euthanasia should be permitted (though the recent defeat of the referendum to permit it in the state of Washington raises some doubt about this support). Thus, many more people seem to want the choice than would be made worse off by getting it. Second, if giving people the option of ending their life really makes them worse off, then we should not only prohibit euthanasia, but also take back from people the right they now have to decide about-life-sustaining treatment. The feared harmful effect should already have occurred from securing people's right to refuse life-sustaining treatment, yet there is no evidence of any such widespread harm or any broad public desire to rescind that right. Third, since there is a wide range of conditions in which reasonable people can and do disagree about whether they would want continued life, it is not possible to restrict the permissibility of euthanasia as narrowly as Velleman suggests without thereby denying it to most persons who would want it; to permit it only in cases in which virtually everyone would want it would be to deny it to most who would want it. A fifth potential bad consequence of making euthanasia legally permissible is that it might weaken the general legal prohibition of homicide. This prohibition is so fundamental to civilized society, it is argued, that we should do nothing that erodes it. If most cases of stopping life support are killing, as I have already

argued, then the court cases permitting such killing have already in effect weakened this prohibition. However, neither the courts nor most people have seen these cases as killing and so as challenging the prohibition of homicide. The courts have usually grounded patients' or their surrogates' rights to refuse life-sustaining treatment in rights to privacy, liberty, self-determination, or bodily integrity, not in exceptions to homicide laws. Legal permission for physicians or others to perform euthanasia could not be grounded in patients' rights to decide about medical treatment. Permitting euthanasia would require qualifying, at least in effect, the legal prohibition against homicide, a prohibition that in general does not allow the consent of the victim to justify or excuse the act. Nevertheless, the very same fundamental basis of the right to decide about life-sustaining treatment—respecting a person's self-determination—does support euthanasia as well. Individual self-determination has long been a wellentrenched and fundamental value in the law, and so extending it to euthanasia would not require appeal to novel legal values or principles. That suicide or attempted suicide is no longer a criminal offence in virtually all states indicates an acceptance of individual self-determination in the taking of one's own life analogous to that required for voluntary active euthanasia. The legal prohibition (in most states) of assisting in suicide and the refusal in the law to accept the consent of the victim as a possible justification of homicide are both arguably a result of difficulties in the legal process of establishing the consent of the victim after the fact. If procedures can be designed that clearly establish the voluntariness of the person's request for euthanasia it would under those procedures represent a carefully circumscribed qualification on the legal prohibition of homicide. Nevertheless, some remaining worries about this weakening can be captured in the final potential bad consequence, to which I will now turn. This final potential bad consequence is the central concern of many opponents of euthanasia and, I believe, is the most serious objection to a legal policy permitting it. According to this "slippery slope" worry, although active euthanasia may be morally permissible in cases in which it is unequivocally voluntary and the patient finds his or her condition unbearable, a legal policy permitting euthanasia would inevitably lead to active euthanasia being performed in many other cases in which it would be morally wrong. To prevent those other wrongful cases of euthanasia we should not permit even morally justified performance of it. Slippery slope arguments of this form are problematic and difficult to evaluate.19 From one perspective, they are the last refuge of conservative defenders of the

II

306

w

11111I

6 End-of-Life Decision-Making

status quo. When all the opponent's objections to the wrongness of euthanasia itself have been met, the opponent then shifts ground and acknowledges both that it is not in itself wrong and that a legal policy which resulted only in its being performed would not be bad. Nevertheless, the opponent maintains, it should still not be permitted because doing so would result in its being performed in other cases in which it is not voluntary and would be wrong. In this argument's most extreme form, permitting euthanasia is the first and fateful step down the slippery slope to Nazism. Once on the slope we will be unable to get off. Now it cannot be denied that it is possible that permitting euthanasia could have these fateful consequences, but that cannot be enough to warrant prohibiting it if it is otherwise justified. A similar possible slippery slope worry could have been raised to securing competent patients' rights to decide about life support, but recent history shows such a worry would have been unfounded. It must be relevant how likely it is that we will end with horrendous consequences and an unjustified practice of euthanasia How likely, and widespread, would the abuses and unwarranted extensions of permitting it be? By abuses, I mean the performance of euthanasia that fails to satisfy the conditions required for voluntary active euthanasia, for example, if the patient has been subtly pressured to accept it. By unwarranted extensions of policy, I mean later changes in legal policy to permit not just voluntary euthanasia, but also euthanasia in cases in which, for example, it need not be fully voluntary. Opponents of voluntary euthanasia on slippery slope grounds have not provided the data or evidence necessary to turn their speculative concerns into well-grounded likelihoods. It is at least clear, however, that both the character and likelihood of abuses of a legal policy permitting euthanasia depend in significant part on the procedures put in place to protect against them. I will not try to detail fully what such procedures might be, but will just give some examples of what they might include: 1. The patient should be provided with all relevant information about his or her medical condition, current prognosis, available alternative treatments, and the prognosis of each. 2. Procedures should ensure that the patient's request for euthanasia is stable or enduring (a brief waiting period could be required) and fully voluntary (an advocate for the patient might be appointed to ensure this). 3. All reasonable alternatives must have been explored for improving the patient's quality of life and relieving any pain or suffering.

4. A psychiatric evaluation should ensure that the patient's request is not the result of a treatable psychological impairment such as depression.2° These examples of procedural safeguards are all designed to ensure that the patient's choice is fully informed, voluntary, and competent, and so a true exercise of self-determination. Other proposals for euthanasia would restrict its permissibility further—for example, to the terminally ill—a restriction that cannot be supported by self-determination. Such additional restrictions might, however, be justified by concern for limiting potential harms from abuse. At the same time, it is important not to impose procedural or substantive safeguards so restrictive as to make euthanasia impermissible or practically infeasible in a wide range of justified cases. These examples of procedural safeguards make clear that it is possible to substantially reduce, though not to eliminate, the potential for abuse of a policy permitting voluntary active euthanasia. Any legalization of the practice should be accompanied by a well-considered set of procedural safeguards together with an ongoing evaluation of its use. Introducing euthanasia into only a few states could be a form of carefully limited and controlled social experiment that would give us evidence about the benefits and harms of the practice. Even then, firm and uncontroversial data may remain elusive, as the continuing controversy over what has taken place in the Netherlands in recent years indicates.2' The Slip into Nonvoluntary Active Euthanasia

While I believe slippery slope worries can largely be limited by making necessary distinctions both in principle and in practice, one slippery slope concern is legitimate. There is reason to expect that legalization of voluntary active euthanasia might soon be followed by strong pressure to legalize some nonvoluntary euthanasia of incompetent patients unable to express their own wishes. Respecting a person's self-determination and recognizing that continued life is not always of value to a person can support not only voluntary active euthanasia, but some nonvoluntary euthanasia as well. These are the same values that ground competent patients' right to refuse lifesustaining treatment. Recent history here is instructive. In the medical ethics literature, in the courts since Quinlan, and in norms of medical practice, that right has been extended to incompetent patients and exercised by a surrogate who is to decide as the patient would have decided in the circumstances if competent? It has been held

Brock: Voluntary Active Euthanasia

unreasonable to continue life-sustaining treatment that the patient would not have wanted just because the patient now lacks the capacity to tell us that. Life-sustaining treatment for incompetent patients is today frequently forgone on the basis of a surrogate's decision, or less frequently on the basis of an advance directive executed by the patient while still competent. The very same logic that has extended the right to refuse life-sustaining treatment from a competent patient to the surrogate of an incompetent patient (acting with or without a formal advance directive from the patient) may well extend the scope of active euthanasia. The argument will be: Why continue to force unwanted life on patients just because they have now lost the capacity to request euthanasia from us? A related phenomenon may reinforce this slippery slope concern. In the Netherlands, what the courts have sanctioned has been clearly restricted to voluntary euthanasia. In itself, this serves as some evidence that permitting it need not lead to permitting the nonvoluntary variety. There is some indication, however, that for many Dutch physicians euthanasia is no longer viewed as a special action, set apart from their usual practice and restricted only to competent persons.23 Instead, it is seen as one end of a spectrum of caring for dying patients. When viewed in this way it will be difficult to deny euthanasia to a patient for whom it is seen as the best or most appropriate form of care simply because that patient is now incompetent and cannot request it. Even if voluntary active euthanasia should slip into nonvoluntary active euthanasia, with surrogates acting for incompetent patients, the ethical evaluation is more complex than many opponents of euthanasia allow. Just as in the case of surrogates' decisions to forgo lifesustaining treatment for incompetent patients, so also surrogates' decisions to request euthanasia for incompetent persons would often accurately reflect what the incompetent person would have wanted and would deny the person nothing that he or she would have considered worth having. Making nonvoluntary active euthanasia legally permissible, however, would greatly enlarge the number of patients on whom it might be performed and substantially enlarge the potential for misuse and abuse. As noted above, frail and debilitated elderly people, often demented or otherwise incompetent and thereby unable to defend and assert their own interests, may be especially vulnerable to unwanted euthanasia. For some people, this risk is more than sufficient reason to oppose the legalization of voluntary euthanasia. But while we should in general be cautious about inferring much from the experience in the Netherlands to what our own experience in the United States might be, there may be one important lesson that we can learn from them. One commentator has noted that in

the Netherlands families of incompetent patients have less authority than do families in the United States to act as surrogates for incompetent patients in making decisions to forgo life-sustaining treatment.24 From the Dutch perspective, it may be we in the United States who are already on the slippery slope in having given surrogates broad authority to forgo life-sustaining treatment for incompetent persons. In this view, the more important moral divide, and the more important with regard to potential for abuse, is not between forgoing life-sustaining treatment and euthanasia, but instead between voluntary and nonvoluntary performance of either. If this is correct, then the more important issue is ensuring the appropriate principles and procedural safeguards for the exercise of decision-making authority by surrogates for incompetent persons in all decisions at the end of life. This may be the correct response to slippery slope worries about euthanasia. I have cited both good and bad consequences that have been thought likely from a policy change permitting voluntary active euthanasia, and have tried to evaluate their likelihood and relative importance. Nevertheless, as I noted earlier, reasonable disagreement remains both about the consequences of permitting euthanasia and about which of these consequences are more important. The depth and strength of public and professional debate about whether, all things considered, permitting euthanasia would be desirable or undesirable reflects these disagreements. While my own view is that the balance of considerations supports permitting the practice, my principal purpose here has been to clarify the main issues.

The Role of Physicians If euthanasia is made legally permissible, should physicians take part in it? Should only physicians be permitted to perform it, as is the case in the Netherlands? In discussing whether euthanasia is incompatible with medicine's commitment to curing, caring for, and comforting patients, I argued that it is not at odds with a proper understanding of the aims of medicine, and so need not undermine patients' trust in their physicians. If that argument is correct, then physicians probably should not be prohibited, either by law or by professional norms, from taking part in a legally permissible practice of euthanasia (nor, of course, should they be compelled to do so if their personal or professional scruples forbid it). Most physicians in the Netherlands appear not to understand euthanasia to be incompatible with their professional commitments. Sometimes patients who would be able to end their lives on their own nevertheless seek the assistance

II ill

IIILL1

6 End-of-Life Decision-Making

of physicians. Physician involvement in such cases may have important benefits to patients and others beyond simply assuring the use of effective means. Historically, in the United States suicide has carried a strong negative stigma that many today believe unwarranted. Seeking a physician's assistance, or what can almost seem a physician's blessing, may be a way of trying to remove that stigma and show others that the decision for suicide was made with due seriousness and was justified under the circumstances. The physician's involvement provides a kind of social approval, or more accurately helps counter what would otherwise be unwarranted social disapproval. There are also at least two reasons for restricting the practice of euthanasia to physicians only. First,

physicians would inevitably be involved in some of the important procedural safeguards necessary to a defensible practice, such as seeing to it that the patient is well informed about his or her condition, prognosis, and possible treatments, and ensuring that all reasonable means have been taken to improve the quality of the patient's life. Second, and probably more important, one necessary protection against abuse of the practice is to limit the persons given authority to perform it, so that they can be held accountable for their exercise of that authority. Physicians, whose training and professional norms give some assurance that they would perform euthanasia responsibly, are an appropriate group of persons to whom the practice may be restricted.

Notes 1.

2. 3.

4.

5.

6. 7.

8.

President's Commission for the Study o f Ethical Problems in Medicine and Biomedical and Behavioral Research, (1983), Deciding to Forego Life-Sustaining Treatment (US Government Printing Office Washington, DC); The Hastings Center, (1987), Guidelines on the Termination of life Sustaining Treatment and Care of the Dying (Indiana University Press: Bloomington); "Current Opinions of the Council on Ethical and Judicial Affairs of the American Medical Association-1989: Withholding or Withdrawing Life-Prolonging Treatment" (1989; American Medical Association: Chicago); George Annas and Leonard Glantz, (1986), "The Right of Elderly Patients to Refuse Life-Sustaining Treatment," in Millbank Memorial Quarterly 64 (suppl. 2): 95-162; Robert E. Weir, (1989), Abating Treatment with Critically Ill Patients (Oxford University Press: New York); Sidney J. Wanzer, et al., (1984), "The Physician's Responsibility toward Hopelessly Ill Patients," in New England Medical Journal 310: 955-9. M.A.M. de Wachter, (1989), "Active Euthanasia in the Netherlands," in jama 262(23): 3315-19. Anonymous, (1988), "It's Over, Debbie," in JAMA 259: 272; Timothy E. Quill, (1990), "Death and Dignity," in New England Journal of Medicine 322: 1881-3. Wanzer, et al., (1989), "The Physicianb Responsibility toward Hopelessly Ill Patients: A Second Look," in New England Journal of Medicine 320: 844-9. Willard Gaylin, Leon R. Kass, Edmund D. Pellegrino, and Mark Siegler, (1988), "Doctors Must Not Kill," in JAMA 259: 2139-40. Bonnie Steinbock, ed., (1980), Killing and Allowing to Die (Prentice Hall: Englewood Cliffs, NJ). Dan W. Brock, (1986), "Forgoing Food and Water: Is It Killing?," in By No Extraordinary Means: The Choice to Forgo Life-Sustaining Food and Water, ed. Joanne Lynn (Indiana University Press: Bloomington), 117-31. James Rachels, (1975), "Active and Passive Euthanasia," in New England Journal of Medicine 292: 78-80; Michael

9.

10.

11. 12.

13. 14. 15.

16.

17. 18.

Tooley, (1983), Abortion and Infanticide (Oxford University Press: Oxford). In my paper "Taking Human Life," (1985), in Ethics 95: 851-65, I argue in more detail that killing in itself is not morally different from allowing to die and defend the strategy of argument employed in this and the succeeding two paragraphs in the text. Dan W. Brock, (1979), "Moral Rights and Permissible Killing," in Ethical Issues Relating to Life and Death, ed. John Ladd (Oxford University Press: New York), 94-117. P Painton and E Taylor, (1990), "Love or Let Die," in Time (19 March): 62-71; Boston Globe/Harvard University Poll, (1991), in Boston Globe (3 Nov.). James Rachels, (1986), The End of life (Oxford University Press: Oxford). Marcia Angell, (1982), "The Quality of Mercy," in New England Journal of Medicine 306: 98-9; M Donovan, P Dillon, and L Mcguire, (1987), "Incidence and Characteristics of Pain in a Sample of Medical-Surgical Inpatients," in Pain 30: 69-78. Eric Cassell, (1991), The Nature of Suffering and the Goals of Medicine (Oxford University Press: New York). Gaylin, et al., "Doctors Must Not Kill." Leon R. Kass, (1989), "Neither for Love Nor Money: Why Doctors Must Not Kill," in The Public Interest 94: 25-46; cf. also his Toward a More Natural Science: Biology and Human Affairs (1985; The Free Press: New York), ch. 6-9. Paul J. Van der Maas, et al., (1991), "Euthanasia and Other Medical Decisions Concerning the End of Life," in Lancet 338: 669-74. Susan M. Wolf, (1989), "Holding the Line on Euthanasia," in Hastings Center Report 19(1 [special supplement]): 13-15. My formulation of this argument derives from David Velleman's statement of it in his commentary on an earlier version of this paper delivered at the American Philosophical Association Central Division meetings; a similar point was made to me by Elisha Milgram in discussion on another occasion. For more general development of the point see Thomas Schelling, (1960), The Strategy of Conflict (Harvard University

Rachels: Active and Passive Euthanasia

Press: Cambridge, MA); and Gerald Dworkin, (1988), "Is More Choice Better Than Less?," in The Theory and Practice of Autonomy (Cambridge University Press: Cambridge). 19. Frederick Schauer, (1985), "Slippery Slopes," in Harvard Law Review 99: 361-83; Wibren van der Burg, (Oct. 1991), "The Slippery Slope Argument," in Ethics 102: 42-65. 20. There is evidence that physicians commonly fail to diagnose depression. See Robert L Misbin, (1991), "Physicians Aid in Dying," in New England Journal of Medicine 325: 1304-7.

21. Richard Fenigsen, (1989), "A Case against Dutch Euthanasia," in Hastings Center Report 19(1 [special supplement]): 22-30. 22. Allen E. Buchanan and Dan W Brock, (1989), Deciding for Others: The Ethics of Surrogate Decisionmaleing (Cambridge University Press Cambridge). 23. Van der Maas, et al., "Euthanasia and Other Medical Decisions." 24. Margaret P Battin, (1990), "Seven Caveats Concerning the Discussion of Euthanasia in Holland," in American Philosophical Association Newsletter on Philosophy and Medicine 89 (2).

Active and Passive Euthanasia

So he asks the doctor for an end to it, and his family joins in the request. Suppose the doctor agrees to withhold treatment, as the conventional doctrine says he may. The justification for his doing so is that the patient is in terrible agony, and since he is going to die anyway, it would be wrong to prolong his suffering needlessly. But now notice this. If one simply withholds treatment, it may take the patient longer to die, and so he may suffer more than he would if more direct action were taken and a lethal injection given. This fact provides strong reason for thinking that, once the initial decision not to prolong his agony has been made, active euthanasia is actually preferable to passive euthanasia, rather than the reverse. To say otherwise is to endorse the option that leads to more suffering rather than less, and is contrary to the humanitarian impulse that prompts the decision not to prolong his life in the first place. Part of my point is that the process of being "allowed to die" can be relatively slow and painful, whereas being given a lethal injection is relatively quick and painless. Let me give a different sort of example. In the United States about 1 in 600 babies is born with Down's syndrome. Most of these babies are otherwise healthy—that is, with only the usual pediatric care, they will proceed to an otherwise normal infancy. Some, however, are born with congenital defects such as intestinal obstructions that require operations if they are to live. Sometimes, the parents and the doctor will decide not to operate, and let the infant die. Anthony Shaw describes what happens then:

James Rachels

The distinction between active and passive euthanasia is thought to be crucial for medical ethics. The idea is that it is permissible, at least in some cases, to withhold treatment and allow a patient to die, but it is never permissible to take any direct action designed to kill the patient. This doctrine seems to be accepted by most doctors, and it is endorsed in a statement adopted by the House of Delegates of the American Medical Association on 4 December 1973: The intentional termination of the life of one human being by another—mercy killing—is contrary to that for which the medical profession stands and is contrary to the policy of the American Medical Association. The cessation of the employment of extraordinary means to prolong the life of the body when there is irrefutable evidence that biological death is imminent is the decision of the patient and/or his immediate family. The advice and judgment of the physician should be freely available to the patient and/or his immediate family. However, a strong case can be made against this doctrine. In what follows I will set out some of the relevant arguments, and urge doctors to reconsider their views on this matter. To begin with a familiar type of situation, a patient who is dying of incurable cancer of the throat is in terrible pain, which can no longer be satisfactorily alleviated. He is certain to die within a few days, even if present treatment is continued, but he does not want to go on living for those days since the pain is unbearable.

When surgery is denied [the doctor] must try to keep the infant from suffering while natural forces sap the baby's life away. As a surgeon whose natural inclination is to use the scalpel to fight

Source From The New England Journal of Medicine 292 (1975) (2), 78-80. Copyright © 1975 Massachusetts Medical Society. All rights reserved.

309

■ Y I

I I .111

111141

6 End-of-Life Decision-Making

off death, standing by and watching a salvageable baby die is the most emotionally exhausting experience I know. It is easy at a conference, in a theoretical discussion, to decide that such infants should be allowed to die. It is altogether different to stand by in the nursery and watch as dehydration and infection wither a tiny being over hours and days. This is a terrible ordeal for me and the hospital staff—much more so than for the parents who never set foot in the nursery' I can understand why some people are opposed to all euthanasia, and insist that such infants must be allowed to live. I think I can also understand why other people favour destroying these babies quickly and painlessly. But why should anyone favour letting "dehydration and infection wither a tiny being over hours and days"? The doctrine that says that a baby may be allowed to dehydrate and wither, but may not be given an injection that would end its life without suffering, seems so patently cruel as to require no further refutation. The strong language is not intended to offend, but only to put the point in the clearest possible way. My second argument is that the conventional doctrine leads to decisions concerning life and death made on irrelevant grounds. Consider again the case of the infants with Down syndrome who need operations for congenital defects unrelated to the syndrome to live. Sometimes, there is no operation, and the baby dies, but when there is no such defect, the baby lives on. Now, an operation such as that to remove an intestinal obstruction is not prohibitively difficult. The reason why such operations are not performed in these cases is, clearly, that the child has Down syndrome and the parents and doctor judge that because of that fact it is better for the child to die. But notice that this situation is absurd, no matter what view one takes of the lives and potentials of such babies. If the life of such an infant is worth preserving, what does it matter if it needs a simple operation? Or, if one thinks it better that such a baby should not live on, what difference does it make that it happens to have an unobstructed intestinal tract? In either case, the matter of life and death is being decided on irrelevant grounds. It is the Down syndrome, and not the intestines, that is the issue. The matter should be decided, if at all, on that basis, and not be allowed to depend on the essentially irrelevant question of whether the intestinal tract is blocked. What makes this situation possible, of course, is the idea that when there is an intestinal blockage, one can "let the baby die," but when there is no such defect there is nothing that can be done, for one must not

"kill" it. The fact that this idea leads to such results as deciding life or death on irrelevant grounds is another good reason why the doctrine should be rejected. One reason why so many people think that there is an important moral difference between active and passive euthanasia is that they think killing someone is morally worse than letting someone die. But is it? Is killing, in itself, worse than letting die? To investigate this issue, two cases may be considered that are exactly alike except that one involves killing whereas the other involves letting someone die. Then, it can be asked whether this difference makes any difference to the moral assessments. It is important that the cases be exactly alike, except for this one difference, since otherwise one cannot be confident that it is this difference and not some other that accounts for any variation in the assessments of the two cases. So, let us consider this pair of cases: In the first, Smith stands to gain a large inheritance if anything should happen to his six-year-old cousin. One evening while the child is taking his bath, Smith sneaks into the bathroom and drowns the child, and then arranges things so that it will look like an accident. In the second, Jones also stands to gain if anything should happen to his six-year-old cousin. Like Smith, Jones sneaks in planning to drown the child in his bath. However, just as he enters the bathroom Jones sees the child slip and hit his head, and fall face down in the water. Jones is delighted; he stands by ready to push the child's head back under if it is necessary, but it is not necessary. With only a little thrashing about, the child drowns all by himself, "accidentally," as Jones watches and does nothing. Now, Smith killed the child, whereas Jones "merely" let the child die. That is the only difference between them. Did either man behave better, from a moral point of view? If the difference between killing and letting die were in itself a morally important matter, one should say that Jones's behaviour was less reprehensible than Smith's. But does one really want to say that? I think not. In the first place, both men acted from the same motive, personal gain, and both had exactly the same end in view when they acted. It may be inferred from Smith's conduct that he is a bad man, although that judgment may be withdrawn or modified if certain further facts are learned about him—for example, that he is mentally deranged. But would not the very same thing be inferred about Jones from his conduct? And would not the same further considerations also be relevant to any modification of this judgment? Moreover, suppose Jones pleaded, in his own defence, "After all, I didn't do anything except just stand there and watch the child drown. I didn't kill him; I only let him die." Again, if letting die were

Rachels: Active and Passive Euthanasia

in itself less bad than killing, this defence should have at least some weight. But it does not. Such a "defence" can only be regarded as a grotesque perversion of moral reasoning. Morally speaking, it is no defence at all. Now, it may be pointed out, quite properly, that the cases of euthanasia with which doctors are concerned are not like this at all. They do not involve personal gain or the destruction of normal, healthy children. Doctors are concerned only with cases in which the patient's life is of no further use to him, or in which the patient's life has become or will soon become a terrible burden. However, the point is the same in these cases: the bare difference between killing and letting die does not, in itself, make a moral difference. If a doctor lets a patient die, for humane reasons, he is in the same moral position as if he had given the patient a lethal injection for humane reasons. If his decision was wrong—if, for example, the patient's illness was in fact curable—the decision would be equally regrettable no matter which method was used to carry it out. And if the doctor's decision was the right one, the method used is not in itself important. The AMA policy statement isolates the crucial issue very well; the crucial issue is "the intentional termination of the life of one human being by another." But after identifying this issue, and forbidding "mercy killing," the statement goes on to deny that the cessation of treatment is the intentional termination of life. This is where the mistake comes in, for what is the cessation of treatment, in these circumstances, if it is not "the intentional termination of the life of one human being by another"? Of course it is exactly that, and if it were not, there would be no point to it. Many people will find this judgment hard to accept. One reason, I think, is that it is very easy to conflate the question of whether killing is, in itself, worse than letting die, with the very different question of whether most actual cases of killing are more reprehensible than most actual cases of letting die. Most actual cases of killing are clearly terrible (think, for example, of all the murders reported in the newspapers), and one hears of such cases every day On the other hand, one hardly ever hears of a case of letting die, except for the actions of doctors who are motivated by humanitarian reasons. So one learns to think of killing in a much worse light than of letting die. But this does not mean that there is something about killing that makes it in itself worse than letting die, for it is not the bare difference between killing and letting die that makes the difference in these cases. Rather, the other factors—the murderer's motive of personal gain, for example, contrasted with the doctor's humanitarian motivation—account for different reactions to the different cases.

I have argued that killing is not in itself any worse than letting die; if my contention is right, it follows that active euthanasia is not any worse than passive euthanasia. What arguments can be given on the other side? The most common, I believe, is the following The important difference between active and passive euthanasia is that, in passive euthanasia, the doctor does not do anything to bring about the patient's death. The doctor does nothing, and the patient dies of whatever ills already afflict him. In active euthanasia, however, the doctor does something to bring about the patient's death: he kills him. The doctor who gives the patient with cancer a lethal injection has himself caused his patient's death; whereas if he merely ceases treatment, the cancer is the cause of the death. A number of points need to be made here. The first is that it is not exactly correct to say that in passive euthanasia the doctor does nothing, for he does do one thing that is very important: he lets the patient die. "Letting someone die" is certainly different, in some respects, from other types of action—mainly in that it is a kind of action that one may perform by way of not performing certain other actions. For example, one may let a patient die by way of not giving medication, just as one may insult someone by way of not shaking his hand. But for any purpose of moral assessment, it is a type of action nonetheless. The decision to let a patient die is subject to moral appraisal in the same way that a decision to kill him would be subject to moral appraisal: it may be assessed as wise or unwise, compassionate or sadistic, right or wrong. If a doctor deliberately let a patient die who was suffering from a routinely curable illness, the doctor would certainly be to blame for what he had done, just as he would be to blame if he had needlessly killed the patient. Charges against him would then be appropriate. If so, it would be no defence at all for him to insist that he didn't "do anything." He would have done something very serious indeed, for he let his patient die. Fixing the cause of death may be very important from a legal point of view, for it may determine whether criminal charges are brought against the doctor. But I do not think that this notion can be used to show a moral difference between active and passive euthanasia. The reason why it is considered bad to be the cause of someone's death is that death is regarded as a great evil—and so it is. However, if it has been decided that euthanasia—even passive euthanasia—is desirable in a given case, it has also been decided that

uu I

I I .111

111111

6 End-of-Life Decision-Making

in this instance death is no greater an evil than the patient's continued existence. And if this is true, the usual reason for not wanting to be the cause of someone's death simply does not apply. Finally, doctors may think that all of this is only of academic interest—the sort of thing that philosophers may worry about but that has no practical bearing on their own work. After all, doctors must be concerned about the legal consequences of what they do, and active euthanasia is clearly forbidden by the law. But even so, doctors should also be concerned with the fact that the law is forcing upon them a moral doctrine that may well be indefensible, and has a considerable effect on their practices. Of course, most doctors are not now in the position of being coerced in this matter, for they do not regard themselves as merely going along with what the law requires. Rather, in statements such as the

AMA policy statement that I have quoted, they are

endorsing this doctrine as a central point of medical ethics. In that statement, active euthanasia is condemned not merely as illegal but as "contrary to that for which the medical profession stands," whereas passive euthanasia is approved. However, the preceding considerations suggest that there is really no moral difference between the two, considered in themselves (there may be important moral differences in some cases in their consequences, but, as I pointed out, these differences may make active euthanasia, and not passive euthanasia, the morally preferable option). So, whereas doctors may have to discriminate between active and passive euthanasia to satisfy the law, they should not do any more than that. In particular, they should not give the distinction any added authority and weight by writing it into official statements of medical ethics.

Notes 1.

Shaw, A. 1972. "Doctor, Do We Have a Choice?," in The New York Times Magazine (30 Jan.): 54.

When Self-Determination Runs Amok Daniel Callahan

The euthanasia debate is not just another moral debate, one in a long list of arguments in our pluralistic society. It is profoundly emblematic of three important turning points in Western thought. The first is that of the legitimate conditions under which one person can kill another. The acceptance of voluntary active euthanasia would morally sanction what can only be called "consenting adult killing." By that term I mean the killing of one person by another in the name of their mutual right to be killer and killed if they freely agree to play those roles. This turn flies in the face of a longstanding effort to limit the circumstances under which one person can take the life of another, from efforts to control the free flow of guns and arms, to abolish capital punishment, and to more tightly control warfare. Euthanasia would add a whole new category of killing to a society that already has too many excuses to indulge itself in that way. The second turning point lies in the meaning and limits of self-determination. The acceptance of euthanasia would sanction a view of autonomy holding that

individuals may, in the name of their own private, idiosyncratic view of the good life, call upon others, including such institutions as medicine, to help them pursue that life, even at the risk of harm to the common good. This works against the idea that the meaning and scope of our own right to lead our own lives must be conditioned by, and be compatible with, the good of the community, which is more than an aggregate of selfdirecting individuals. The third turning point is to be found in the claim being made upon medicine: it should be prepared to make its skills available to individuals to help them achieve their private vision of the good life. This puts medicine in the business of promoting the individualistic pursuit of general human happiness and well-being. It would overturn the traditional belief that medicine should limit its domain to promoting and preserving human health, redirecting it instead to the relief of that suffering which stems from life itself, not merely from a sick body. I believe that, at each of these three turning points, proponents of euthanasia push us in the wrong direction. Arguments in favour of euthanasia fall into four general categories, which I will take up in turn: (1) the moral claim of individual self-determination and

Source: From The Hastings Center Report 22, 2 (March/April 1992), 52-5. Reprinted by permission of The Hastings Center and the author.

Callahan: When Self-Determination Runs Amok

well-being; (2) the moral irrelevance of the difference between killing and allowing to die; (3) the supposed paucity of evidence to show likely harmful consequences of legalized euthanasia; and (4) the compatibility of euthanasia and medical practice. Self-Determination Central to most arguments for euthanasia is the principle of self-determination. People are presumed to have an interest in deciding for themselves, according to their own beliefs about what makes life good, how they will conduct their lives. That is an important value, but the question in the euthanasia context is: What does it mean and how far should it extend? If it were a question of suicide, where a person takes her own life without assistance from another, that principle might be pertinent, at least for debate. But euthanasia is not that limited a matter. The self-determination in that case can only be effected by the moral and physical assistance of another. Euthanasia is thus no longer a matter only of self-determination, but of a mutual, social decision between two people, the one to be killed and the other to do the killing. How are we to make the moral move from my right of self-determination to some doctor's right to kill me—from my right to his right? Where does the doctor's moral warrant to kill come from? Ought doctors to be able to kill anyone they want as long as permission is given by competent persons? Is our right to life just like a piece of property, to be given away or alienated if the price (happiness, relief of suffering) is right? And then to be destroyed with our permission once alienated? In answer to all those questions, I will say this: I have yet to hear a plausible argument why it should be permissible for us to put this kind of power in the hands of another, whether a doctor or anyone else. The idea that we can waive our right to life, and then give to another the power to take that life, requires a justification yet to be provided by anyone. Slavery was long ago outlawed on the ground that one person should not have the right to own another, even with the other's permission. Why? Because it is a fundamental moral wrong for one person to give over his life and fate to another, whatever the good consequences, and no less a wrong for another person to have that kind of total, final power. Like slavery, duelling was long ago banned on similar grounds: even free, competent individuals should not have the power to kill each other, whatever their motives, whatever the circumstances. Consenting adult killing, like consenting adult slavery or degradation, is a strange route to human dignity.

There is another problem as well. If doctors, once sanctioned to carry out euthanasia, are to be themselves responsible moral agents not simply hired hands with lethal injections at the ready then they must have their own independent moral grounds to kill those who request such services. What do I mean? As those who favour euthanasia are quick to point out, some people want it because their life has become so burdensome it no longer seems worth living. The doctor will have a difficulty at this point. The degree and intensity to which people suffer from their diseases and their dying, and whether they find life more of a burden than a benefit, has very little directly to do with the nature or extent of their actual physical condition. Three people can have the same condition, but only one will find the suffering unbearable. People suffer, but suffering is as much a function of the values of individuals as it is of the physical causes of that suffering. Inevitably in that circumstance, the doctor will in effect be treating the patient's values. To be responsible, the doctor would have to share those values. The doctor would have to decide, on her own, whether the patient's life was "no longer worth living." But how could a doctor possibly know that or make such a judgment? Just because the patient said so? I raise this question because, while in Holland at the euthanasia conference, . . . the doctors present agreed that there is no objective way of measuring or judging the claims of patients that their suffering is unbearable. And if it is difficult to measure suffering, how much more difficult to determine the value of a patient's statement that her life is not worth living? However one might want to answer such questions, the very need to ask them, to inquire into the physician's responsibility and grounds for medical and moral judgment, points out the social nature of the decision. Euthanasia is not a private matter of self-determination. It is an act that requires two people to make it possible, and a complicit society to make it acceptable. Killing and Allowing to Die Against common opinion, the argument is sometimes made that there is no moral difference between stopping life-sustaining treatment and more active forms of killing, such as lethal injection. Instead I would contend that the notion that there is no morally significant difference between omission and commission is just wrong. Consider in its broad implications what the eradication of the distinction implies: that death from disease has been banished, leaving only the actions of physicians in terminating treatment as the cause of death. Biology, which used to bring about death, has

■ Y

II

11

I 1 :.111

111111

6 End-of-Life Decision-Making

apparently been displaced by human agency. Doctors have finally, I suppose, thus genuinely become gods, now doing what nature and the deities once did. What is the mistake here? It lies in confusing causality and culpability, and in failing to note the way in which human societies have overlaid natural causes with moral rules and interpretations. Causality (by which I mean the direct physical causes of death) and culpability (by which I mean our attribution of moral responsibility to human actions) are confused under three circumstances. They are confused, first, when the action of a physician in stopping treatment of a patient with an underlying lethal disease is construed as causing death. On the contrary, the physicianb omission can only bring about death on the condition that the patient's disease will kill him in the absence of treatment. We may hold the physician morally responsible for the death, if we have morally judged such actions wrongful omissions. But it confuses reality and moral judgment to see an omitted action as having the same causal status as one that directly kills. A lethal injection will kill both a healthy person and a sick person. A physician's omitted treatment will have no effect on a healthy person. Turn off the machine on me, a healthy person, and nothing will happen. It will only, in contrast, bring the life of a sick person to an end because of an underlying fatal disease. Causality and culpability are confused, second, when we fail to note that judgments of moral responsibility and culpability are human constructs. By that I mean that we human beings, after moral reflection, have decided to call some actions right or wrong, and to devise moral rules to deal with them. When physicians could do nothing to stop death, they were not held responsible for it. When, with medical progress, they began to have some power over death, but only its timing and circumstances, not its ultimate inevitability, moral rules were devised to set forth their obligations. Natural causes of death were not thereby banished. They were, instead, overlaid with a medical ethics designed to determine moral culpability in deploying medical power. To confuse the judgments of this ethics with the physical causes of death which is the connotation of the word kill is to confuse nature and human action. People will, one way or another, die of some disease; death will have dominion over all of us. To say that a doctor "kills" a patient by allowing this to happen should only be understood as a moral judgment about the licitness of his omission, nothing more. We can, as a fashion of speech only, talk about a doctor killing a patient "by omitting treatment he should have provided." It is a fashion of speech precisely because it is the underlying

disease that brings death when treatment is omitted; that is its cause, not the physician's omission. It is a misuse of the word killing to use it when a doctor stops a treatment he believes will no longer benefit the patient when, that is, he steps aside to allow an eventually inevitable death to occur now rather than later. The only deaths that human beings invented are those that come from direct killing when, with a lethal injection, we both cause death and are morally responsible for it. In the case of omissions, we do not cause death even if we may be judged morally responsible for it. This difference between causality and culpability also helps us see why a doctor who has omitted a treatment he should have provided has "killed" that patient while another doctor performing precisely the same act of omission on another patient in different circumstance does not kill her, but only allows her to die. The difference is that we have come, by moral convention and conviction, to classify unauthorized or illegitimate omissions as acts of "killing." We call them "killing" in the expanded sense of the term: a culpable action that permits the real cause of death, the underlying disease, to proceed to its lethal conclusion. By contrast, the doctor who, at the patient's request, omits or terminates unwanted treatment does not kill at all. Her underlying disease, not his action, is the physical cause of death; and we have agreed to consider actions of that kind to be morally licit. He thus can truly be said to have "allowed" her to die. If we fail to maintain the distinction between killing and allowing to die, moreover, there are some disturbing possibilities. The first would be to confirm many physicians in their already too powerful belief that, when patients die or when physicians stop treatment because of the futility of continuing it, they are somehow both morally and physically responsible for the deaths that follow That notion needs to be abolished, not strengthened. It needlessly and wrongly burdens the physician, to whom should not be attributed the powers of the gods. The second possibility would be that, in every case where a doctor judges medical treatment no longer effective in prolonging life, a quick and direct killing of the patient would be seen as the next, most reasonable step, on grounds of both humaneness and economics. I do not see how that logic could easily be rejected. Calculating the Consequences

When concerns about the adverse social consequences of permitting euthanasia are raised, its advocates tend to dismiss them as unfounded and overly speculative. On the contrary, recent data about the Dutch experience suggests that such concerns are right on target.

Callahan: When Self-Determination Runs Amok

From my own discussions in Holland, and from the articles on that subject, I believe we can now fully see most of the likely consequences of legal euthanasia. Three consequences seem almost certain, in this or any other country: the inevitability of some abuse of the law; the difficulty of precisely writing, and then enforcing, the law; and the inherent slipperiness of the moral reasons for legalizing euthanasia in the first place. Why is abuse inevitable? One reason is that almost all laws on delicate, controversial matters are to some extent abused. This happens because not everyone will agree with the law as written and will bend it, or ignore it, if they can get away with it. From explicit admissions to me by Dutch proponents of euthanasia, and from the corroborating information provided by the Remmelink Report and the outside studies of Carlos Gomez and John Keown, I am convinced that in the Netherlands there are a substantial number of cases of nonvoluntary euthanasia, that is, euthanasia undertaken without the explicit permission of the person being killed. The other reason abuse is inevitable is that the law is likely to have a low enforcement priority in the criminal justice system. Like other laws of similar status, unless there is an unrelenting and harsh willingness to pursue abuse, violations will ordinarily be tolerated. The worst thing to me about my experience in Holland was the casual, seemingly inherent attitude toward abuse. I think that would happen everywhere. Why would it be hard to precisely write, and then enforce, the law? The Dutch speak about the requirement of "unbearable" suffering, but admit that such a term is just about indefinable, a highly subjective matter admitting of no objective standards. A requirement for outside opinion is nice, but it is easy to find complaisant colleagues. A requirement that a medical condition be "terminal" will run aground on the notorious difficulties of knowing when an illness is actually terminal. Apart from those technical problems there is a more profound worry. I see no way, even in principle, to write or enforce a meaningful law that can guarantee effective procedural safeguards. The reason is obvious yet almost always overlooked. The euthanasia transaction will ordinarily take place within the boundaries of the private and confidential doctor—patient relationship. No one can possibly know what takes place in that context unless the doctor chooses to reveal it. In Holland, less than 10 per cent of the physicians report their acts of euthanasia and do so with almost complete legal impunity. There is no reason why the situation should be any better elsewhere. Doctors will have their own reasons for keeping euthanasia secret, and some patients will have no less a motive for wanting it concealed.

I would mention, finally, that the moral logic of the motives for euthanasia contain within them the ingredients of abuse. The two standard motives for euthanasia and assisted suicide are said to be our right of self-determination and our claim upon the mercy of others, especially doctors, to relieve our suffering. These two motives are typically spliced together and presented as a single justification. Yet if they are considered independently and there is no inherent reason why they must be linked they reveal serious problems. It is said that a competent, adult person should have a right to euthanasia for the relief of suffering. But why must the person be suffering? Does not that stipulation already compromise the principle of selfdetermination? How can self-determination have any limits? Whatever the person's motives may be, why are they not sufficient? Consider next the person who is suffering but not competent, who is perhaps demented or mentally retarded. The standard argument would deny euthanasia to that person. But why? If a person is suffering but not competent, then it would seem grossly unfair to deny relief solely on the grounds of incompetence. Are the incompetent less entitled to relief from suffering than the competent? Will it only be fluent, middleclass people, mentally fit and savvy about working the medical system, who can qualify? Do the incompetent suffer less because of their incompetence? Considered from these angles, there are no good moral reasons to limit euthanasia once the principle of taking life for that purpose has been legitimated. If we really believe in self-determination, then any competent person should have a right to be killed by a doctor for any reason that suits him. If we believe in the relief of suffering, then it seems cruel and capricious to deny it to the incompetent. There is, in short, no reasonable or logical stopping point once the turn has been made down the road to euthanasia, which could soon turn into a convenient and commodious expressway. Euthanasia and Medical Practice

A fourth kind of argument one often hears both in the Netherlands and in this country is that euthanasia and assisted suicide are perfectly compatible with the aims of medicine. I would note at the very outset that a physician who participates in another person's suicide already abuses medicine. Apart from depression (the main statistical cause of suicide), people commit suicide because they find life empty, oppressive, or meaningless. Their judgment is a judgment about the value of continued life, not only about health (even if they are sick). Are doctors now to be given the right to

II

:al

1111:LI

6 End-of-Life Decision-Making

make judgments about the kinds of life worth living and to give their blessing to suicide for those they judge wanting? What conceivable competence, technical or moral, could doctors claim to play such a role? Are we to medicalize suicide, turning judgments about its worth and value into one more clinical issue? Yes, those are rhetorical questions. Yet they bring us to the core of the problem of euthanasia and medicine. The great temptation of modern medicine, not always resisted, is to move beyond the promotion and preservation of health into the boundless realm of general human happiness and well-being. The root problem of illness and mortality is both medical and philosophical or religious. "Why must I die?" can be asked as a technical, biological question or as a question about the meaning of life. When medicine tries to respond to the latter, which it is always under pressure to do, it moves beyond its proper role. It is not medicine's place to lift from us the burden of that suffering which turns on the meaning we assign to the decay of the body and its eventual death. It is not medicine's place to determine when lives are not worth living or when the burden of life is too great to be borne. Doctors have no conceivable way of evaluating such claims on the part of patients, and they should have no right to act in response to them. Medicine should try to relieve human suffering,

but only that suffering which is brought on by illness and dying as biological phenomena, not that suffering which comes from anguish or despair at the human condition. Doctors ought to relieve those forms of suffering that medically accompany serious illness and the threat of death. They should relieve pain, do what they can to allay anxiety and uncertainty, and be a comforting presence. As sensitive human beings, doctors should be prepared to respond to patients who ask why they must die, or die in pain. But here the doctor and the patient are at the same level. The doctor may have no better an answer to those old questions than anyone else; and certainly no special insight from his training as a physician. It would be terrible for physicians to forget this, and to think that in a swift, lethal injection, medicine has found its own answer to the riddle of life. It would be a false answer, given by the wrong people. It would be no less a false answer for patients. They should neither ask medicine to put its own vocation at risk to serve their private interests, nor think that the answer to suffering is to be killed by another. The problem is precisely that, too often in human history killing has seemed the quick, efficient way to put aside that which burdens us. It rarely helps, and too often simply adds to one evil still another. That is what I believe euthanasia would accomplish. It is selfdetermination run amok.

Medical Ethics and Double Effect: The Case of Terminal Sedation

procedure within palliative care. But sometimes the amount of sedative needed to control suffering has the effect of shortening the patient's life. This creates worries that the requirements of appropriate palliative care mandate actions indistinguishable from euthanasia, which is illegal and morally objectionable to many health care professionals. Invoking double effect addresses these worries: the intent of the physician prescribing the lifeshortening analgesics is to control the suffering, not to shorten life. Evidence of physician intent can be found in notations on the patient's chart and in the recorded dosages and titration of analgesics. Consequently, this action is not euthanasia but palliative care.2 A consensus was reached among a small but representative group of Canadian intensivists and a similar group of coroners that this application of double effect provided proper ethical guidance concerning terminal sedation. Possibly, therefore, there is consensus or the prospect of the emergence of a consensus

Joseph M. Boyle

The ethics of the medical profession incorporates some application of the moral doctrine of the double effect, particularly to govern medical decisions whose outcomes include shortening a patient's life. Very roughly. double effect provides that it can be morally good to shorten a patient's life as a foreseen and accepted but unintended side effect of an action undertaken for a good reason, even if it is agreed that intentionally killing the patient or shortening the patient's life is wrong. The medical profession's use of this moral doctrine has some support in legal decisions.' The use of terminal sedation to control the intense discomfort of dying patients appears to be an established

Source: From Theoretical Medicine and Bioethics: Philosophy of Medical Research and Practice 25 (2004) (1), 51-60, Boyle, J. © 2004, Springer Netherlands.

Boyle: Medical Ethics and Double Effect:The Case of Terminal Sedation

on this application of double effect among intensivists and other physicians, including those with oversight responsibility for deaths related to medical decisions.3 Whether or not the consensus reported in this study can be further validated or extended, it pointedly raises the central ethical issues involved in the prospect that this application of double effect could be an established part of medical ethics—and, consequently, legally enforced. Addressing these issues is distinct from determining the consensus of practitioners. The central issue is perhaps this: double effect is a general moral doctrine, not an ad hoc device to deal with terminal sedation or other difficult life and death decisions. This doctrine emerged within the casuistry of Roman Catholicism. How much of this distinctive moral view does the medical profession and the law implicitly accept by accepting this application of double effect? To answer this question it is useful to consider further the idea of double effect and its rationale within Catholic moral theology. The expression "double effect" was first used by St Thomas Aquinas (1225-74) to refer to the duality of the results of a single human action. A person performing an action that is foreseen to have a multiplicity of results can have very different interests in them—from serious commitment to bringing about a result to reluctant acceptance of a result that is unwanted but unavoidable. Since actions are purposeful, at least one of the results of an action must be intended, but others, although knowingly and voluntarily caused, can be outside the agent's intention; these latter results I will call "accepted side effects." Aquinas made this distinction in discussing killing in self-defence: one who uses lethal force for defence against attack need not intend the assailant's death; that can be outside the agent's intention.4 Following Aquinas, Roman Catholic moral theology has attributed a specific, and very important, moral significance to this distinction between what a person intends in acting and what a person accepts as a side effect of intentionally acting for another result. The significance is this: the factors sufficient to make simply wrong actions involving the intention of some result are not also sufficient to make simply wrong actions involving accepting, but not intending, a result of the same kind. In other words, the impermissibility of an action that is based upon the agentb intention of a certain result does not render impermissible actions having a result of that kind, if the result is not intended, but accepted as a side effect. In Aquinas's example of killing in selfdefence, the moral issue was framed by the acceptance of the Augustinian prohibition of killing in self-defence on the part of private persons, those lacking public authority to kill. That prohibition, Aquinas maintained, applied only to intentional killing, not to killing brought about and accepted as a side effect. Consequently, the

acceptance of the Augustinian limitation on those who are permitted to kill does not imply the impermissibility of lethal self-defence, where the death is not intentional. The manuals of moral theology of the nineteenth century refined and formulated Aquinas's reasoning into a set of rules useful for giving moral advice and conducting casuistry. These rules became known as the doctrine of double effect, or the principle of double effect. They were generally formulated as three or four necessary conditions for the permissibility of actions similar in some of their results to actions that are impermissible because of the general categories into which they fall. Thus, the classic formulation of the Jesuit moralist J.-B. Gury, in the mid-nineteenth century: It is licit to posit a cause which is either good or indifferent from which there follows a twofold effect, one good the other evil, if a proportionately grave reason is present, and if the end of the agent is honorable—that is, if he does not intend the evil.' Gury elaborated these three conditions into four, by construing the condition of honourable intention as two. The first addressed the distinction between a means and a side effect: if the bad effect—that is, the result which would render that action simply wrong were it intended—is the means to the good effect, then it cannot be a side effect and is intended. Thus, the key requirement that the good effect be brought about "immediately," that is, not by means of the bad effect. The second of these extrapolated conditions—that one intend only the good effect—excludes cases in which the bad effect is not brought about as a means to the good effect, but is nevertheless intended because it functions as an independent goal. An example would be "bonus" effects—results that emerge as side effects of bringing about a goal but then recognized as independently useful or beneficial and so (ordinarily) intended. Gury first two conditions—that the "cause" be morally good or indifferent and that there be a proportionately grave reason for doing what brings about evil side effects—refer to the further moral considerations that are needed for a complete assessment of an action meeting the conditions for upright intention. These conditions address the two areas where Gury thought an action that cleared the intentionally focused conditions might still fail morally. The first area of concern arises from the possibility that, prior to any consideration of further results that might be intended or accepted as side effects but not intended, some actions might be simply wrong. Perhaps his thought here is that since the movements a person chooses for the sake of self-defence have both defensive and destructive results, one can distinguish the chosen

I I ill

111111

6 End-of-Life Decision-Making

performance from the results and ask of it whether that performance is morally permissible. In some cases, telling a lie or committing adultery, for example, the action is impermissible on account of considerations logically prior to those concerning intended or accepted results. In the case at hand, a physician's prescribing analgesics, described in just that way, is morally indifferent; therefore, the results, intentions, and other circumstances of this chosen behaviour will determine its permissibility or impermissibility. In this respect the action is unlike acts of adultery or lying, which as so described are wrong. This condition may be strictly redundant (and perhaps also confused) since an intended result is intrinsic to an action as a chosen performance, as the lying and adultery cases indicate. Nevertheless, this condition highlights an important aspect of the style of moral reasoning involved in double effect: if this condition, or either of the intentionally focused conditions is not met, then the act is simply and indefeasibly impermissible. These are absolute judgments that cannot be overturned by further considerations of the action's particular circumstances: thus, in Aquinas's example of killing in self-defence, for any person who is not publicly authorized, intentionally killing the assailant is simply, that is, indefeasibly, wrong; nothing further one can discover about the action will reverse that moral judgment.6 This is the absolutism of double effect. The second area of concern is that even if the intentional conditions are met, and the action is not excluded as simply wrong for some prior reason, the full consideration of its circumstances might still turn up a morally excluding factor. The requirement of a proportionately grave reason explicitly addresses this second area; it presupposes that bringing about as a side effect what would be wrong to bring about intentionally is likely harmful or otherwise morally suspect, and so in need of wider justification. Thus, the requirement is that any other considerations relevant to the moral assessment of the action should be brought to bear. In Aquinas's example of self-defence, two concerns were addressed: whether one defending oneself has a duty to refrain from harming the assailant, and the extent of the violence of the defensive action. He argued that one duty to refrain is generally less pressing than the duty to protect one's life, and only that level of violence needed for the defence is justified.' The determination of what constitutes a proportionately grave matter is not essential for assessing what is distinctive in the doctrine of double effect. That determination is likely to be as complex and as variable in outcome as any other reasoning leading to an all-things-considered assessment of the morality of an action!' But the existence of this condition, both in the

manualist formulations and in Aquinas's reasoning in his classic statement, underlines the fact that double effect does not imply that it is permissible to bring about bad results if and only if they are not intended. In the case of terminal sedation, this condition of proportionality seems to be easily met, and the existing consensus assumes that. The need for palliation of some dying patients is substantial and is assumed generally to justify terminal sedation if moral and legal worries about euthanasia are satisfactorily addressed.9 This brief survey of the Catholic sources of double effect shows what it means, and at least roughly how it works. But the summary does not provide a justification of the special moral significance attributed to the distinction between what a person intends in acting and accepts as a side effect (hereafter the intended/ accepted distinction). That is because there appears to be no developed justification in the tradition. It seems that Aquinas and the theological tradition regarded the ethical significance of the intended/accepted distinction of double effect as simply obvious. Aquinas's unargued assertion that what is intended is morally per se and what is outside or beyond the intention is per accidens suggests as much.l° Some applications of double effect are certainly intuitively compelling, for example, those distinguishing terror bombing in warfare from carefully targeted bombing of military targets where some civilians will likely be killed as side effects. But such intuitive convictions are not readily generalized to all cases of double effect, particularly to those where peoples' normative convictions about the cases distinguished is more variable and uncertain than in the bombing cases.0 The distinction between terminal sedation and euthanasia may distinguish just such cases: some may accept terminal sedation but regard efforts to distinguish it from euthanasia to be sleight of hand; some may, for reasons independent of reasoning about terminal sedation, accept euthanasia as morally legitimate and so fail to see the point of introducing double effect's distinctions. Moreover, people's convictions about especially clear cases are not sufficiently focused to justify the precise significance double effect requires the intended/accepted distinction to bear. Thus, some think that double effect implies that the distinction between bad outcomes that are intended and those that are accepted is that bringing about the former is, other things being equal, morally worse than bringing about the latter.12 This is not obviously true.. the reckless or unjustified inflicting of harm as a side effect is not clearly worse than inflicting it intentionally. More importantly, the implication of double effect is different: namely, that the wrongness of the former does not guarantee that of the latter.

Boyle: Medical Ethics and Double Effect:The Case of Terminal Sedation

Perhaps the moral significance of this distinction appeared obvious to older Catholic moralists because of its analogy to the structure of divine creative activity: God creates only good, but allows the evils of his creation—evils he does not intentionally cause but only permits as privations that flaw his good creation.° This analogue suggests a justification. Just as God creates only what is good, humans should voluntarily pursue in their actions only what is humanly good. And just as God permits the evils flawing his creation, so humans must accept some evil consequences they should not intend. The idea is that God, if he is to create some universe he has good reason to create, must permit the evils which inevitably arise as side effects of his creating that universe—for example, the misuse of free will by rational creatures' immoral choices. Applied to human action, the analogy suggests that in acting, humans can and should aim exclusively at the good, but that there will inevitably be some bad side effects of doing that. Independently of the suggestive theological analogy, there is a limitation on the human capacity to pursue the good, and that limitation is precisely an incapacity to avoid evil side effects, not an incapacity to choose and intend only the good. The limitation is this: in all the situations calling for human choice, no matter what a person chooses to do, some instance of a human good will be harmed, destroyed or at least knowingly neglected (hereafter I will refer to all such harms as simply "harming a good"). Thus, it is beyond human power to act in such a way that one's action does nothing more than promote human good; in all human action some instances of human goods are promoted and served while others are, at the very least, not promoted (as when a person leaves some of her talents undeveloped to pursue a career), or, very often, more or less harmed (as one risks health for the sake of sports), or in some cases even destroyed (as when one kills, whether intentionally or as a side effect). This limitation is essentially a limitation on the human capacity to avoid some bad side effects of good choices, and not a limitation on the human capacity to avoid choosing precisely for the sake of bad goals. In the choice to act for some goal, namely, in an intentional action, it always remains in the agent's power to choose not to do it. So, when something humanly evil— harming a good—is the intended result of one's action, one always has the choice of not doing that action. But one does not have a choice about whether or not there will be some bad side effects of whatever one chooses to do. Accepting bad side effects, therefore, is unavoidable; choosing to pursue results that involve harming a good is always avoidable—though often at a high price.

Since some bad side effects are inevitable, a morality based on concern for the human good does not justify an impossible prohibition of bringing them about Rather, the relevant moral guidance concemingbringing about bad side effects addresses questions such as which bad side effects are to be accepted and on whom the harms will fall—matters considered under the proportionality condition of double effect. But if the underlying moral principle is taken to be allegiance to the human good, a persons intentional actions can always be aimed at the good, since one can always choose to avoid any that are not Prohibition against intentionally harming a human good is not an impossible prohibition. As already noted above, the prohibitions against intentionally harming human goods presupposed by double effect are taken to be indefeasible or absolute. That puts the morality in which double effect developed at odds with much of current moral opinion. What now counts as common sense morality largely rejects indefeasible prohibitions of actions of generally described kinds; and consequentialism rejects moral absolutes as inadequate devices for promoting the good. But if allegiance to the good is morally basic, and if instances of goods are not commensurable in goodness, then rational concern for the good readily justifies absolute prohibitions of intentionally harming them. For if instances of human goods are incommensurable in their goodness, the respect for each instance of such goodness is required because there cannot be any good or sum of goods that would capture precisely the goodness of that instance. On these assumptions, therefore, absolute prohibitions of harms to the most basic goods alone do justice to the reality of the human good.14 But although double effect presupposes the truth of some such indefeasible norms, its function in moral thought is not to justify them but rather to limit their application to intentionally harming the goods of human beings. If the preceding account of the justification of double effects use of the intended/accepted distinction is sound, then within the framework of traditional morality, as understood by the older Catholic moralists, double effect is a legitimate moral doctrine. Within that context, it plainly is a moral doctrine, not a specifically Catholic or religious doctrine, since it is the implication of the character of a limitation in human action and willing joined with the implications of a kind of goods based ethical theory. Can the justification be exported from its strictly moral framework to the regulatory framework of medical ethics? In one respect, double effect cannot be exported as is. A person's intentions in doing an action may be inferred from the action and other aspects of its context, most importantly, how a person explains his or her actions But a person's intentions themselves are at least in part inaccessible to others, and sometimes difficult

■u

I

II :III

111111

6 End-of-Life Decision-Making

even for the acting person to articulate accurately and reflectively. Opponents of double effect argue that these epistemological difficulties are insurmountable, indeed, that they suggest that the basic concepts of double effect cannot be rationally applied." However, difficult applications of double effect can be debated with the possibility of confident agreement, and not all applications are controversial. Moreover, an individual's efforts to make upright choices will involve personal moral scrutiny and discussion with others that can lead to judgments that are sufficiently confident to guide choices for many cases." But in the context of the regulation of behaviour by law or by professional ethics, the third-person perspective of judges and juries becomes more central, and concerns about people's deepest moral orientations less important. Here the relevant volitions are not those inaccessible in a person's heart, but only those of which there is evidence accessible to others besides the agent. For example, the evidence of intent in terminal sedation cases is the sum of notations on charts, dosages and titration of analgesics, and so on. A physician who would prefer to perform euthanasia but who remains constrained in his or her actions by the requirements of intent as publicly accessible indicates that he or she does not intend to end life, even if he or she wishes to do that. And that implies that those who co-operate with such a physician's action, are co-operating only in a common act of palliative care, not of euthanasia. These considerations show that the question about the exportability of double effect to medical ethics must be reformulated. The inherent limitations involved in using a moral doctrine for public regulatory purposes must be recognized. The question remains: supposing they are recognized, can the moral doctrine be exported?

If there are some kinds of behaviour which society or a profession judges unacceptable, then altogether banning that behaviour may be tempting. But a general ban on behaviour having certain results likely prohibits too much, for reasons already noted: for example, causing death or bodily harm is not reasonably prohibited by medical ethics when unavoidable, and such results are unavoidable in some clinical conditions. However, banning intentional killing is not prohibiting too much in this way. Furthermore, there appear to be good reasons why the medical profession would want to uphold such a ban," and, therefore, if practitioners are confident of their ability publicly to determine intent in the relevant cases, then such a prohibition would be as justified as the prohibitions in the strictly moral cases. The fact that the ban would not be justified if applied to the acceptance of side effects has, of itself, no tendency to call into question the narrower exclusion of intentional killing. The logic of double effect, therefore, has application in medical ethics and the law, quite independently of the particular moral framework in which it was developed and has a natural function in moral reasoning. Double effect does not provide the justification of norms excluding intentionally harming a person's good, but reminds us that when such norms are taken as true or appropriate, whether as moral norms or social regulations, they cannot reasonably prohibit harming as a side effect. And so, the inference common in debates about euthanasia is not sound: it is not the case that the fact that we accept bringing about death or an earlier death as a side effect of choosing something else gives rational grounds for judging intentional killing to be justified.18

Notes 1. Vacco v. Quill, 117 Supreme Court Reporter 2293 (1997);

2.

3.

4. 5. 6. 7.

J. Finnis, G. Grisez, and J. Boyle, (2001), "'Direct' and `Indirect': A Reply to Critics of Our Action Theory," in The Thomist 65: 1-44. L. Hawryluck and W. Harvey, (2000), "Analgesia, Virtue, and the Principle of Double Effect," in Journal of Palliative Care 16 (supplement): S24—S30. L Hawryluck, W. Harvey, L. Lemieux-Charles, and P Singer, (2002), "Consensus Guidelines on Analgesis and Sedation in Dying Intensive Care Patients," in BMC Medical Ethics [online], available at www.biomedcentral.com/14726939/3/3. St Thomas Aquinas, Summa Theologiae, q. 64, a 7. J. Boyle, (1980), "Toward Understanding the Principle of Double Effect," in Ethics 90: 527-38. Cf. note 4, above. Cf. note 4, above.

8. G. Grisez, (1997), The Way of the LordJesus, Volume 3: Difficult Moral Questions, Appendix 1: "Human Acts and Moral Judgments" (Franciscan Press: Quincy, IL), 849-70. 9. Cf. note 2 and note 3, above. 10. Cf. note 4, above. 11. J. Boyle, (1991), "Who Is Entitled to Double Effect?," in The Journal of Medicine and Philosophy 16: 475-94. 12. Cf. note 11, above. 13. J. Boyle, (1997), "Intentions, Christian Morality and Bioethics: Puzzles of Double Effect," in Christian Bioethics 3: 87-8. 14. J. Finnis, J. Boyle, and G. Grisez, (1987), "A Sounder Theory of Morality," in Nuclear Deterrence, Morality and Realism (Oxford University Press: Oxford), 275-96; see also note 11 and note 13, above. 15. T. Quill, R Dresser, and D. Brock, (1997), "The Rule of Double Effect: A Critique of Its Role in End-of-Life Decision Making," in New EnglandJournal of Medicine 337: 1768-71.

Boyle: Medical Ethics and Double Effect: The Case of Terminal Sedation

16. J. Finnis, G. Grisez, and J. Boyle, (2001), —Direct' and 'Indirect': A Reply to Critics of Our Action Theory," in The Thomist 65: 1-44; see also note 5, above. 17. D. Sulmasy and E. Pellegrino, (1999), "The Rule of Double Effect: Clearing Up the Double Talk," in Archives of

Internal Medicine 159: 545-50; see also Vacco v. Quill, note 1, above. 18. Cf. D. Sulmasy and E. Pellegrino, "The Rule of Double Effect: Clearing Up the Double Talk," note 17, above. See also note 13, above.

6.3 Physician-Assisted Dying in Canada

Controlling the Risks of PAD British Columbia Supreme Court

•

Key areas of risk [748-7541 Conclusions re: feasibility of controlling the risks [798-8541 Overall conclusion [8831

Ix. Safeguards: Feasibility

A. Introduction [7481 It is Canada's position [i.e., the position of the government of Canada which opposed the legalization of PAD in the SCCI that without an absolute prohibition on physician-assisted suicide, there is a risk of "wrongful deaths." Canada identifies the following key areas of risk. [7491 First, Canada adverts to the deaths of incompetent persons. It can be difficult to determine whether a person is competent to request an assisted death, particularly in the context of a short-term physician— patient relationship. It can be difficult to perform obj ective assessments uninfluenced by assumptions about the quality of the patient's life and the reasonableness of her desire for death. It can be just as difficult to assess whether and to what extent impaired cognitive function may have affected a person's decision-making capacity. [7501 Second, Canada identifies the risk of involuntary deaths. It points to the need to be certain that a patient's request for an assisted death is voluntary that is, free from coercion, pressure, undue inducement, and psychological or emotional manipulation. Some of the external forces that can influence a patient's decision include illness (such as depression—even where it is not sufficient to impair capacity, depression can

still influence an individual's decision by impacting his or her outlook and perceptions of his or her circumstances); lack of Information about options; concerns about burdening, or pressure from, family members; the physician's influence, particularly in light of the power differential that exists between physician and patient; and society's approval of physician-assisted dying. Such forces can elude detection. [7511 Third, Canada says there is a risk of deaths of individuals with treatable conditions. The desire for a hastened death is often associated with underlying conditions that may be amenable to therapeutic intervention, including depression, hopelessness, physical distress and the perceived difficulty of accessing health care resources to relieve pain or other physical suffering [7521 Fourth, Canada refers to the possible deaths of ambivalent individuals. The desire for death is often transitory Canada says. Individuals frequently change their minds about suicide, and even life-sustaining treatment options, when they have the opportunity. [7531 Fifth, Canada says there is a risk of deaths of misinformed individuals. It asserts a very real danger that individuals, during difficult periods of illness or disability, may choose assisted suicide based on an inaccurate prognosis, and thereby be deprived of a life they did not realize at the time they could have had. [7541 Finally, Canada identifies the risk of deaths of vulnerable individuals, in particular, the elderly and people with disabilities. [. . .1 1. Competence [7981 Weighing the evidence as a whole, I [Justice Lynn Smith, the trial judge of the BCSCI conclude that it is feasible for properly-qualified and experienced physicians reliably to assess patient competence, including in the context of life-and-death decisions, so long as

Source: From Carter v Canada (Attorney General), 2012 BCSC 886. Reprinted with permission.

I

II

ill

111111

6 End-of-Life Decision-Making

they apply the very high level of scrutiny appropriate to the decision and proceed with great care. E. . .1 2. Voluntariness

[815] Although I accept the evidence of Dr. Gallagher, Dr. Chochinov, Professor Helsel and Professor Frazee that influence can be subtle and exercised at an unconscious level, I accept the evidence of Dr. Ganzini and Dr. Donnelly that coercion and undue influence can be detected as part of a capacity assessment. To be accurate and reliable, clinicians who perform such assessments would have to be aware of the risks of coercion and undue influence, of the possibility of subtle influence, and of the risks of unconscious biases regarding the quality of the lives of persons with disabilities or persons of advanced age. 3. Informed Consent

[816] I will deal under this heading with two of the concerns that Canada raises: the deaths of individuals who are misinformed, and the deaths of individuals with treatable conditions. [. . .] [831] The evidence as to informed consent permits me to conclude that, just as physicians routinely assess the requirements for informed consent in patients seeking or refusing medical treatment, it would be feasible to require informed consent for physician-assisted death. In that context, particular care would be required to ensure a patient is properly informed of her diagnosis and prognosis, given the seriousness of the decision. As well, the range of treatment options described would have to encompass all reasonable palliative care interventions, including those aimed at loss of personal dignity 4. Ambivalence

Battin, while it is possible for a person who is grievously and irremediably ill to be "suicidal" in the traditional sense, this is not necessarily the case for those seeking physician-assisted death, and it cannot be assumed to be so. Instead, the decisions often reflect long-held, deep-seated values and are rational, consultive, informed, and considered. [843] The evidence on the question of ambivalence leads me to the conclusion that it is feasible to screen out from physician-assisted death patients who are ambivalent, by assessing capacity and requiring some time to pass between the decision and its implementation. As to the argument that it is impossible to screen out people who may, in the future, come to change their minds, I agree with Professor Battin's observations about the need to respect the decisions that people make, even when future circumstances are not fully certain. [. . .] 5. Vulnerable Individuals

a) Elderly people [847] I accept that elderly persons are vulnerable to abuse and that the assessment of voluntariness of elderly people must incorporate an understanding of that reality. As discussed earlier, however, there is no evidence that the elderly access physician-assisted dying in disproportionate numbers in permissive jurisdictions (Professor Battin, Dr. Ganzlnl, Dr. van Delden, Professor Deliens), and Professor Deliens observed that the number of patients over the age of 80 whose deaths resulted from IAWER [i.e., non-voluntary euthanasia] in Belgium was not disproportionate. [. . .] b) People with disabilities [853]I accept that persons with disabilities face prejudice and stereotyping and that there is a risk of unconscious bias about the quality of life of a person with a disability. However, while I accept Professor Frazee's evidence and take into account the submissions of the defendants and the Euthanasia Prevention Coalition, I am not persuaded that the risks to persons with disabilities are such that they cannot be avoided through practices of careful and well-informed capacity assessments by qualified physicians who are alert to those risks.

[832] In Professor Mishara's view, one of the prime characteristics of suicide is that ambivalence is omnipresent. A very large number of people change their minds after considering suicide. Even among those who attempt suicide, the vast majority change their minds after initiating the attempt, he says. The experience in other jurisdictions, and some experience with patients who initially determined to refrain from life sustaining treatment options, shows how often people change their minds.

C. Addressing the Risks

[833] Expert witnesses called by the plaintiffs, however, emphasize that in their opinion the decision-making process for assisted dying cannot be equated with that involved in suicide outside that context. According to those witnesses, such as Professor Werth and Professor

[854] This review of the evidence permits no conclusion other than that there are risks inherent in permitting physician-assisted death, and that the utmost care would be needed in designing and managing a system which would allow it, in order to avoid those risks. [. • .]

Justice: Legislative Background: Medical Assistance in Dying (Bill C-14)

D. Conclusion

[8831 My review of the evidence in this section, and in the preceding section on the experience in permissive jurisdictions, leads me to conclude that the risks

Legislative Background: Medical Assistance in Dying (Bill C-14) Department of Justice [...] Part 1—Description of Proposed Legislation on Medical Assistance in Dying

The development of the proposed legislation on medical assistance in dying (Bill C-14) was informed by the evidence before all levels of court in the Carter case, by available Canadian and international research, social science evidence, governmental reports and parliamentary studies. It was also informed by the experience of existing international medical assistance in dying regimes, as well as by numerous recent consultative activities on such assistance, including the work of the Special Joint Committee on Physician-Assisted Dying, the External Panel on Options for a Legislative Response to Carter v Canada, the Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying, the Canadian Medical Association, the College of Family Physicians of Canada, and the work of provincial colleges of physicians and surgeons, among others. Bill C-14 would strike an appropriate balance between the autonomy of those individuals seeking access to medical assistance in dying and the interests of vulnerable persons and of society, through amendments to the Criminal Code to allow physicians and nurse practitioners to provide assistance in dying to eligible competent adults in accordance with specified safeguards. The legal effect of the new legislation would be to decriminalize medical assistance in dying and leave further regulation of the practice to the provinces and territories (Pi's) should they so choose. Medical assistance in dying has aspects that fall under both federal and provincial jurisdiction. The criminal law aspects of such assistance fall under exclusive federal jurisdiction and would apply consistently across the country The Prs can legislate in relation to the health care aspects and civil law implications so long as PT legislation does not conflict with

inherent in permitting physician-assisted death can be identified and very substantially minimized through a carefully-designed system imposing stringent limits that are scrupulously monitored and enforced. [. . .1

the criminal law. From a federal perspective, a reasonable degree of consistency across and within provinces and territories would support the underlying values and principles of the Canada Health Act—that is, that all Canadians should have access to needed health care services, which could include services related to medical assistance in dying, without financial or other barriers. L Legislative Objectives

The objectives of the proposed legislation are expressly stated in the preamble of Bill C-14 and include: •

•

•

•

•

• •

recognizing the autonomy of persons who have a grievous and irremediable medical condition that causes them enduring and intolerable suffering to seek medical assistance in dying; recognizing that robust safeguards, which reflect the irrevocable nature of ending a life, are essential to prevent error and abuse in the provision of medical assistance in dying; affirming the inherent and equal value of every person's life and avoiding encouraging negative perceptions of the quality of life of persons who are elderly, ill or disabled; protecting vulnerable persons from being induced, in moments of weakness, to end their lives; recognizing that suicide is a significant public health issue that can have lasting and harmful effects on individuals, families and communities; recognizing that those who wish to access medical assistance in dying should be able to do so without adverse legal consequences on their families; respecting the personal convictions of health care providers; and, recognizing that a consistent approach to medical assistance in dying across Canada is desirable.

The legislative objectives in the preamble speak to the circumstances for which medical assistance in dying would be made available, but they also speak to the reasons why access would not be permitted in other circumstances. For example, the objective of recognizing that suicide is a public health issue

Source Minister of Justice and Attorney General of Canada (2016). Reprinted with permission.

II

,a1

III LL.I

6 End-of-Life Decision-Making

helps to explain why medical assistance in dying is not presently being contemplated for people who are not approaching a natural death. To permit it in circumstances where a person is not approaching natural death could be seen as undermining suicide prevention initiatives and normalizing death as a solution to many forms of suffering. That said, the preamble also recognizes the Government of Canada's commitment to explore additional circumstances in which a person may seek access to medical assistance in dying, namely requests by mature minors, advance requests, and where mental illness is the sole underlying medical condition. These circumstances are complex, and require additional study and consideration. As well, the preamble affirms the Government's commitment to uphold the principles set out in the Canada Health Act and to develop non-legislative measures to support the improvement of end-of-life care in the country. In this regard, the Minister of Health has offered to work in collaboration with the Yrs on the development of an endof-life care coordination system to provide information on options and facilitate patient access to care. II. Definitions

The Bill would enact new definitions in the Criminal Code for the purpose of the national medical assistance in dying regime (see Annex I for a clause by clause guide to Bill C-14). The proposed new section 241.1 (clause 3 of the Bill) is a central provision of the regime as it would define the umbrella term "medical assistance in dying" as encompassing what is commonly called voluntary euthanasia (i.e., the administration by a medical practitioner or nurse practitioner of medication that will cause a person's death at their request) and assisted suicide (i.e., the prescription or provision by a medical practitioner or nurse practitioner of medication that a person could self-administer to cause their own death). It should be noted that, in the case of what is commonly called voluntary euthanasia, the provision of medical assistance in dying would result directly in the patient's death, whereas in the case of assisted suicide, the provision of such assistance would result in the patient obtaining medication that they could choose to use—or not—to end their life. This difference would be reflected in the new criminal exemptions, as set out below. With respect to health care professionals, who are regulated under PT responsibility, the new section 241.1 would define the terms "medical practitioner" and "pharmacist" as those who are entitled to practise medicine or pharmacology under provincial laws (e.g., licensed professionals who are members of provincial colleges of physicians and surgeons or provincial

colleges of pharmacists). Although the term "physician" or "doctor" is more plain language in English, the term "medical practitioner" is already used in several places in the Criminal Code, and so is adopted to ensure consistency within the Criminal Code. The term "nurse practitioner" is defined in the Bill as a registered nurse who is designated as a nurse practitioner or other equivalent title, and who is entitled to autonomously make diagnoses, interpret tests, prescribe medications and treat individuals. III. Criminal Exemptions Culpable Homicide It is a crime to intentionally cause the death of another

person, even if they consent to die (section 14 of the Criminal Code). The Bill would therefore enact a new exemption from criminal liability for culpable homicide for medical practitioners and nurse practitioners who provide medical assistance in dying in the form of what is commonly called voluntary euthanasia, i.e., the administration of medication to a person, at their request, that causes their death (new section 227(1) of the Criminal Code, in clause 2 of the Bill). The Bill would also create an additional and related exemption for any other person who would do anything in order to help a physician or nurse practitioner provide such assistance. The latter category would include, for instance, a social worker who is asked to meet with the patient to help assess the voluntariness of their request, a lawyer to a hospital who is asked to review the documents to verify compliance with the law, or a pharmacist who fills the prescription for the medication to be administered by a medical practitioner or nurse practitioner to the patient. The Bill would also re-enact the legal rule (section 14 of the Criminal Code—clause 1 of the Bill) that says that a person's consent to die is not a defence for someone who inflicts death on them. This rule was found to be unconstitutional in Carter. At the same time, the Bill would clarify that this legal rule does not apply in the case of a person who receives medical assistance in dying in conformity with the new regime (new subsection 227(4) of the Criminal Code—clause 2 of the Bill). In all other circumstances, it would continue to be a crime to cause the death of a person who consented to die. Aiding a Person to Die by Suicide It is a crime to assist a person to die by suicide, whether or not suicide ensues (paragraph 241 (b) of the Criminal Code). Assistance can be in the form of providing information about how to end their life, or providing a tool or other means that could be used. The Bill would enact criminal exemptions for

Justice: Legislative Background: Medical Assistance in Dying (Bill C-14)

the offence of aiding a person to die by suicide for medical practitioners, nurse practitioners and people who would assist them in providing or dispensing medication to eligible persons. The person could selfadminister the medication to cause their own death, either at that time or at a later time of their choosing (new subsections 241(2) and 241(3) of the Criminal Code—clause 3 of the Bill). The Bill recognizes that after a person would receive the prescription from the physician or nurse practitioner (which meets the definition of medical assistance in dying in the Criminal Code as explained above), they may still need various forms of assistance depending on their state of health. For instance, they may need someone to collect the medication from the pharmacy to open the bottle that contains such medication, or to lift a glass of water to their mouth so that they can swallow it. As these forms of conduct are prohibited under the offence of aiding a person to die by suicide, additional exemptions are provided in the Bill to address this type of assistance. Specifically, pharmacists who fill a prescription for the purpose of medical assistance in dying and give the medication directly to the patient or to another person on their behalf would also be exempted for this conduct (new subsection 241(4)—clause 3 of the Bill). Similarly, any person who helps the patient to self-administer the medication would also be exempted from criminal responsibility (new subsection 24(5)—clause 3 of the Bill). Any person helping someone to self-administer the medication should exercise extreme caution, as the decision to self-administer, and the final actions of doing so, must be those of the patient for whom the medication was prescribed. In any other circumstance, there would remain a risk of criminal prosecution. Finally, the Bill would re-enact the offence of aiding a person to die by suicide, and this conduct will continue to be criminal in all circumstances other than those described in the medical assistance in dying regime under the Criminal Code. "Reasonable but Mistaken Belief"

For greater certainty, the Bill would codify a common law principle to clarify that the exemptions would still apply even if the practitioners, or other exempted individuals, have a reasonable but mistaken belief about some relevant fact; for example, whether the person is 18 years of age and is therefore eligible to receive medical assistance in dying (new subsections clauses 227(3) and 241(6) in clauses 2 and 3 of the Bill). These would essentially function as a "good faith" defence for those who would participate in medical assistance in dying.

IV. Eligibility Criteria for Medical Assistance in Dying

Bill C-14 would enact a new section 241.2 of the Criminal Code, which in essence sets out the criminal rules surrounding the provision of medical assistance in dying. New subsections 241.2(1) and 241.2(2) would set out the eligibility criteria for such assistance in Canada. Under the proposed legislation, medical assistance in dying would be available to a person who meets all of the following criteria (subsection 241.2(1)): •

• •

• •

being an adult (at least 18 years old) who is mentally competent ("capable") to make health care decisions for themselves; having a grievous and irremediable medical condition (as defined under subsection 241.2(2)); making a voluntary request for medical assistance in dying which does not result from external pressure; giving informed consent to receive medical assistance in dying; and, being eligible for health services funded by a government.

A grievous and irremediable medical condition would be expressly defined under the Bill as (subsection 241.2(2)): • • •

•

having a serious and incurable illness, disease or disability, and, being in an advanced state of irreversible decline in capability; and, experiencing enduring physical or psychological suffering, due to the illness, disease, disability or state of decline, that is intolerable to the person and cannot be relieved in a manner that they consider acceptable; and, where the person's natural death has become reasonably foreseeable, taking into account all of their medical circumstances, without requiring a specific prognosis as to the length of time the person has left to live.

The proposed eligibility criteria would enable individuals who are intolerably suffering, in an advanced state of irreversible decline in capability, and who are on a trajectory towards their natural death, to have the option of a peaceful medically-assisted dying process, instead of having to endure a painful, prolonged or undignified one. It would enable them to make a fundamentally personal decision concerning their bodily integrity, autonomy and dignity, which could also help

uw

I

I I .111

111111

6 End-of-Life Decision-Making

prevent them from ending their lives prematurely, by access medical assistance in dying if they met all of providing reassurance that they will have access to the eligibility criteria. medical assistance in dying at a time when they may The requirement for the person to be eligible to be unable to end their own life without assistance. In- receive publicly funded health services is intended to dividuals would have to be able to provide informed prevent foreigners from visiting Canada to obtain medconsent when eligibility is assessed and confirm their ical assistance in dying. consent one last time immediately before medical assistance in dying is provided. V. Safeguards In Carter, the Court expressly stated that the scope In Carter, the Court acknowledged that there are inof its declaration was "intended to respond to the fac- herent risks in permitting medical assistance in dying, tual circumstances" of the case and that it made "no but agreed with thetrialjudge that these risks"can be pronouncement on other situations where physician- identified and very substantially minimized through assisted dying may be sought." The proposed eligibility a carefully-designed system imposing stringent limits criteria would directly respond to the factual circum- that are scrupulously monitored and enforced." stances raised in Carter. The individuals whose cases As enacted in other jurisdictions that have medical were considered by the Court were either in physical assistance in dying, robust procedural safeguards decline and nearing death in the late stages of a fatal are a critical component of any carefully-designed disease, or were otherwise nearing the end of their lives regime and are essential to prevent error and abuse while in decline due to a condition that was not fatal from occurring and to protect vulnerable persons. in itself. This approach would also be consistent with Consistent with the Court ruling and the Special the Court's comparison of medical assistance in dying Joint Committee's recommendations on safeguards, with other forms of end-of-life care, i.e., medical as- Bill C-14 would enact mandatory procedural safesistance in dying would become another end-of-life guards that medical practitioners and nurse practioption, in addition to palliative care and palliative sed- tioners would be required to follow before providing ation for instance, for intolerably suffering individuals medical assistance in dying to a person. These safewhose deaths are reasonably foreseeable. The Court guards would require the medical practitioner or also recognized that assisted dying is a complex issue nurse practitioner to: confirm that the person meets involving various interests, that a number of solutions all eligibility criteria for medical assistance in dying; were possible, and that Parliament's response would ensure that the person's request was made in writing receive a high degree of deference. (See Annex B for after the person was informed that his or her natural more information.) death was reasonably foreseeable and be satisfied The criterion of reasonable foreseeability of that it was signed and dated in the presence of two death is intended to require a temporal but flex- independent witnesses; ensure that the person was ible connection between the person's overall med- informed that they may withdraw their request at ical circumstances and their anticipated death. As any time; ensure that a second independent medsome medical conditions may cause individuals to ical practitioner or nurse practitioner provided a irreversibly decline and suffer for a long period of written opinion confirming the person's eligibility; time before dying, the proposed eligibility criteria ensure that a period of at least 15 days has elapsed would not impose any specific requirements in between the moment the written request was signed terms of prognosis or proximity to death (e.g., a six and the provision of medical assistance in dying month prognosis as the US states' medical assistance (unless both practitioners agree that death or loss in dying laws require). The medical condition that of capacity to consent is imminent); and, immediis causing the intolerable suffering would not need ately before providing such assistance, confirm the to be the cause of the reasonably foreseeable death. person's consent. In other words, eligibility would not be limited to In order to be considered independent, the witthose who are dying from a fatal disease. Eligibil- nesses could not be a beneficiary under the will of that ity would be assessed on a case-by-case basis, with person or otherwise benefit from their death, or be flexibility to reflect the uniqueness of each person's involved directly in giving care to the person, among circumstances, but with limits that require a natural other criteria (new subsection 241.2(5)—clause 3 of death to be foreseeable in a period of time that is the Bill). With respect to the independence of the first not too remote. It should be noted that people with medical practitioner or nurse practitioner from the a mental illness or physical disability would not be second one, the Bill provides that they could not be excluded from the regime, but would only be able to connected to each other in any way that could impair

Justice: Legislative Background: Medical Assistance in Dying (Bill C-14)

their objectivity, such as by being in a business or mentoring relationship with each other. They would also need to be independent of the patient, in the sense that they could not be a beneficiary under his or her will, or be otherwise connected to them in a manner that could affect their objectivity (new subsection 241.2(6)). However, the legislation makes clear that standard compensation for providing medical services would not affect the physician or nurse practitioner's independence. The Bill would make provision for individuals who are unable to sign their own request, by enabling them to ask a capable adult, who is not directly involved in providing health care services to them, to sign and date their request on their behalf and in their presence (new subsection 241.2(4)). Finally, in order to fall within the exemption, the medical practitioner or nurse practitioner would also be required to act with reasonable knowledge, care and skill, in accordance with applicable provincial laws, rules or standards (new subsection 241.2(7)), and to inform the pharmacist whenever medication is sought or prescribed for the purposes of medical assistance in dying (new subsection 241 .2 (8)). VI. Monitoring System

As recognized by the Court in Carter, by the Special Joint Committe8 report, as well as by many stakeholders, a pan-Canadian monitoring system to collect and analyze data on the provision of medical assistance in dying across Canada, to monitor trends and provide information to the public on the implementation of the new law, would be a critical component of the new Canadian regime. Such a monitoring system would be essential to foster transparency and public trust in the system.

Filing Obligations and Related Offences

To enable the monitoring regime to operate effectively, the proposed legislation would also create a legal obligation for medical and nurse practitioners who receive a written request for medical assistance in dying to provide information as prescribed in regulations (new subsection 241.31(1)). Pharmacists would also be required to provide information in relation to their provision of medications used for such assistance (new subsection 241.31(2)). Bill C-14 proposes a hybrid offence—punishable by a maximum of 2 years imprisonment—for failure to comply with the reporting obligations, and for any person who knowingly breaches the regulations. The reporting obligations and related offences would be brought into force at a later date than the rest of the Bill, once the detailed regulations on the monitoring regime would be in place. VII. Other Offences

To ensure compliance with the new medical assistance in dying legislation, and given the serious and irrevocable nature of helping people die and the potential for criminal liability for medical professionals, the Bill would enact new hybrid offences for failure to comply with the mandatory safeguards (new section 241.3), and for forging or destroying documents related to assistance requests with criminal intent (new section 241.4). For instance, a person might destroy a written medical assistance in dying request in order to block someone's access to such assistance or to interfere with a medical practitioner's ability to rely on an exemption, or they might forge the signature of a person they were trying to influence to seek assistance. These offences would be liable to a maximum term of imprisonment of five years, where prosecuted on indictment, and to a maximum term of 18 months on summary conviction.

Regulation-Making Power

The proposed legislation would create a power for the Minister of Health to make regulations to establish a system for monitoring requests for, and the provision of, medical assistance in dying in Canada. Such regulations may include: information to be provided by medical practitioners or nurse practitioners to a designated body; the form, manner and circumstances in which the information has to be provided; details about how the data would be analyzed, interpreted, and reported to the public; and rules for the protection and disposal of such information (new subsection 241.31(3)—clause 4 of the Bill). The Minister of Health would consult with PTs and others in developing the regulations.

VIII. Related Amendments

Bill C-14 would enact related amendments to other statutes to ensure that recourse to medical assistance in dying does not affect pensions under the Pension Act or benefits under the Canadian Forces Members and Veterans Re-establishment and Compensation Act. It would also amend the Corrections and Conditional Release Act to ensure that no investigation need be conducted under section 19 of that Act as a result of a person dying with medical assistance. IX.Parliamentary Review

The Bill includes a mandatory legislative review to take place five years after its coming into force.

6 End-of-Life Decision-Making

xThe medical assistance in dying legislation would be referred to a designated parliamentary committee of the House of Commons, the Senate or both Houses of Parliament. The committee would review all provisions of Bill C-14 and submit a report to Parliament. It could also consider any reports that would be published under the monitoring regime. The parliamentary review could assess whether the new regime is meeting its legislative objectives, how medical assistance in dying is being implemented across Canada, and identify areas of potential changes and improvements, if necessary.

Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection This reading consists o f excerpts from two reports compiled to compare their recommendations on four issues presented below in sections titled "Mature Minors," "Mental Illness," "Advance Directives," and "Conscientious Objection." The reports are the following: •

•

Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying: Final Report, November 2015 Medical Assistance in Dying: A Patient-Centred Approach: Report of the Special Joint Committee on Physician-Assisted Dying, February 2016

Below, the reports are referred to as "Expert Advisory Group" and "Special Joint Committee," respectively. Mature Minors Expert Advisory Group (p. 34) Adult Person RECOMMENDATION 17: Access to physician-assisted dying should not be impeded by the imposition of arbitrary age limits. Provinces and territories should recommend that the federal government make it clear in its changes to the Criminal Code that eligibility for physician-assisted dying is to be based on competence rather than age.

X. Areas for Further Study

The parliamentary review would be a distinct process from the Government of Canada's commitment to study additional complex circumstances in which a person may seek access to medical assistance in dying, namely requests by mature minors, advance requests, and requests where mental illness is the sole underlying medical condition. The results of these studies could be considered by the parliamentary committee as part of its broader legislative review of the new regime. F. . .1

The Carter decision applies to a "competent adult person," but does not include a definition of adult. In assessing whether someone is an adult person, an arbitrary age limit such as 18 years old provides no valid safeguard. Instead, it is important that willing physicians carefully consider the context of each request to determine whether the person has the information needed, is not under coercion or undue pressure, and is competent to make such a decision. Competence and consent to treatment laws vary among provinces and territories. In some cases, decisional competence is based on age, but in others, age is irrelevant to whether the patient has the legal authority to make decisions about his or her treatment. An ideal regulatory framework for physicianassisted dying would have uniform eligibility rules. Indeed, many provider groups asked the Advisory Group to recommend a consistent national approach to eligibility for physician-assisted dying and patient advocacy groups also sought an easily understandable criterion for accessing the service regardless of where in the country a patient lives. The federal government could facilitate this by affirming that the definition of "adult" in the context of physicianassisted dying relates to a patient's ability to give consent rather than a particular age cut-off. That is, to allow access to patients who are competent, regardless of age. [. . .1 Special Joint Committee (pp. 18-21) D. Age

The Carter decision dealt with plaintiffs who were adults, so no decision was made with regard to minors' eligibility for MAID. However, as Prof. Pelletier made clear, Parliament can choose to allow minors to access MAID.'

Source: From Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying Final Report (2015) and from Medical Assistance in Dying: A Patient-Centred Approach: Report of the Special Joint Committee on Physician-Assisted Dying (2016) 42nd Parliament, 1st Session.

Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection

In response to a comment stating that it was up to the Committee to determine what the age of consent would be in relation to a specific offence, Prof. Hogg replied:

who was head of psychiatry at Vancouver's Children's Hospital for 30 years, argued for a competence-based approach, saying

Yes, I think that's right. The Supreme Court, in its order, spoke of a "competent adult person." I don't think it would be open to you, for example, to have 16 as an age of consent for this purpose, because that would not be a competent adult person. Between 18 and 21, I would think you would have some leeway within the word "adult" to decide that.2

I have worked with many teenagers over the years and I have worked with a number who have been facing death, and I think they would be competent in the legal sense to consent to physician-assisted dying as they would be legally competent to agree to other kinds of medical care.

Certain witnesses, such as the Nova Scotia College of Physicians and Surgeons, chose not to take a position on this issue, simply asking for greater clarity to be provided.3 Some witnesses who appeared before the Committee and the External Panel recommended that legislation define an age below which MAID would not be available (generally 18 but one submission suggested as old as 25).4 Other witnesses wanted all individuals who are competent to make the decision to be eligible for MAID.5 One witness also flagged the need to consider the suffering of children who are not competent, though did not go so far as to suggest including them in any MAID regime.° To date, Belgium and the Netherlands are the only two jurisdictions that allow minors access to MAID.7 Stakeholders who prefer competency-based criteria, such as the Provincial/Territorial Expert Advisory Group and the British Columbia Civil Liberties Association, argue that the trend is toward increased recognition of the competence of minors in health care decision-making and that age limits are arbitrary.8 Prof. Pelletier stated that suffering is suffering, regardless of age, and that there is a risk that the provisions may be challenged on the basis of section 15 of the Charter (equality rights) if minors are excluded.9 The Canadian Paediatric Society advocated against including minors, regardless of competence, in any MAID regime. The organization made this argument for a number of reasons, including the lack of evidence before the court in Carter regarding minors; the fact that an age limit is not arbitrary, and the lack of social consensus with respect to MAID for minors. The organization also rejected the idea that a constitutional challenge by excluded minors would clearly be successful. It suggested addressing whether to allow minors to access MAID at a later date, after there has been time to gather data, as was the case in Belgium which legalized minors' access to MAID in 2014, 12 years after adults were granted access." In contrast, Dr. Derryck Smith, Chair of the Physicians Advisory Council of Dying with Dignity Canada

. . . Why would we want teenagers to suffer, but we're prepared to relieve adults of suffering?" Other witnesses such as Margaret Birrell, President of the Alliance of People with Disabilities Who Are Supportive of Legal Assisted Dying Society, and Dr. John Soles, President of the Society of Rural Physicians of Canada, were open to minors possibly having access, but felt that this should not be allowed at the present time." Dr. Hartley Stern, Executive Director and CEO of the Canadian Medical Protective Association, said that if "mature minors" are to be entitled to MAID, clarification is needed as to how their competency will be assessed." Quebec's An Act respecting end-of-life care restricts "medical aid in dying" to "a person of full age."" The Committee understands the concerns of many witnesses regarding the capacity of minors to understand the implications of such a serious decision. However, it is important to remember, as noted in the External Panel's report, that the Supreme Court has stated that minors have a right "to a degree of decision-making autonomy that is reflective of their evolving intelligence and understanding."15 Allowing competent minors access to MAID would not be eliminating the requirement for competence. Given existing practices with respect to mature minors in healthl° and the obvious fact that minors can suffer as much as any adult, the Committee feels that it is difficult to justify an outright ban on access to MAID for minors. As with issues of mental health, by instituting appropriate safeguards, health care practitioners can be relied upon to identify appropriate cases for MAID and to refuse MAID to minors that do not satisfy the criteria. The Committee acknowledges that a competent mature minor who has a grievous and irremediable medical condition should not be forced to endure intolerable suffering. Moreover, there are serious questions whether a restriction of the right to MAID only to competent adults would be consistent with the Charter. However, the Committee realizes that witnesses and briefs received were of differing opinions on the subject of extending the right to MAID to mature minors, and that these differences reflect a divergence of opinion

II

.111

111111

6 End-of-Life Decision-Making

among the Canadian public. After reflecting on the issue, the Committee recommends the following: RECOMMENDATION 6 That the Government of Canada implement a two-stage legislative process, with the first stage applying immediately to competent adult persons 18 years or older, to be followed by a second stage applying to competent mature minors, coming into force at a date no later than three years after the first stage has come into force; and That the Government of Canada immediately commit to facilitating a study of the moral, medical and legal issues surrounding the concept of "mature minor" and appropriate competence standards that could be properly considered and applied to those under the age of 18, and that this study include broadbased consultations with health specialists, provincial and territorial child and youth advocates, medical practitioners, academics, researchers, mature minors families and ethicists before the coming into force of the second stage. Mental Illness Special Joint Committee (pp. 13-15)

2. Mental Illness Because the individuals that brought the case in Carter did not have mental health issues, the Court made no pronouncement with respect to MAID and psychiatric conditions. Jeanette Ettel, Senior Counsel, Human Rights Law Section at the Department of Justice, told the Committee that it was open to the Committee to consider whether to include psychiatric illnesses in the conditions that could result in a right to MAID.17 As was the experience of the External Panel, the Committee heard widely diverging views on how to address mental health in the context of MAID. Benoit Pelletier, member of the External Panel and an expert in constitutional law, noted that the External Panel identified greater support from Canadians for MAID in the context of a physical illness but that, prima facie, the Carter criteria would also apply to psychiatric conditions. 18 Professor Downie and others supported this position: [MI ental illness should not be an exclusion criterion. It was not excluded by the Supreme Court, and not all individuals with mental illness are incompetent. Physicians already routinely determine whether someone is competent, even when they have a mental illness. Furthermore, the suffering that can accompany mental illness can be as

excruciating as any suffering that can accompany physical illness. Finally, I would argue that excluding individuals on the basis of mental illness would violate the charter." A number of witnesses and submissions expressed concern about mental illness in the context of mAiD.2° Dr. K. Sonu Gaind, President of the Canadian Psychiatric Association, outlined some of the challenges that will need to be addressed: In terms of what is "irremediable," careful consideration needs to be given about what this means in the context of mental illness. Irremediable, of course, cannot simply mean incurable. Many conditions in psychiatry and medicine are considered chronic and not curable, but things may be done to remediate or improve the situation.21 Dr. Tarek Rajji, Chief of Geriatric Psychiatry at the Centre for Addiction and Mental Health, told the Committee that: [MI ental illness may be grievous to an individual, and symptoms can cause enduring psychological and sometimes physical suffering. However, suffering should not be equated with an irremediable nature, and the lack of inevitable and predictable death by natural history provides us with an opportunity to deliver recovery-based treatment. [P] eople with mental illness may be vulnerable to the impact of the social determinants of health. They may live in poverty, have poor housing, and lack social support. These circumstances may exacerbate suffering and a person's perception that their illness is irremediable . . . with a clinical recovery-based environment, there is always the potential for mental illness to be remediable. 22 In response, Professor Downie reminded the Committee of the following aspect of the Carter judgment: "Irremediable," it should be added, does not require the patient to undertake treatments that are not acceptable to the individual.23 The Committee recognizes that there will be unique challenges in applying the eligibility criteria for MAID where the patient has a mental illness, particularly where such an illness is the condition underlying the request. However, where a person is competent and fits the other criteria set out by law, the Committee does not see how that individual could be denied

Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection

a recognized Charter right based on his or her mental health condition. Furthermore, we do not understand the Carter decision to exclude mental illnesses. Any individual applying for MAID would need to satisfy all the criteria, including irremediability and capacity. As several witnesses reminded the Committee, health professionals will need to strike an appropriate balance between the rights of all Canadians to access this constitutionally protected right, and the protection of those vulnerable persons who might be coerced into requesting MAID. Cases involving mental illness may prove challenging to address for health care practitioners, but the Committee has faith in the expertise of Canadian health care professionals to develop and apply appropriate guidelines for such cases. The difficulty surrounding these situations is not a justification to discriminate against affected individuals by denying them access to MAID. The Committee expects that cases where the underlying condition is a mental health condition will be rare, as is the case in other jurisdictions that have legalized MAID.24 A more detailed discussion of appropriate safeguards can be found below. The Committee therefore recommends: RECOMMENDATION 3 That individuals not be excluded from eligibility for medical assistance in dying based on the fact that they have a psychiatric condition.

are satisfied before accepting a request for MAID and recommends: RECOMMENDATION 4 That physical or psychological suffering that is enduring and intolerable to the person in the circumstances of his or her condition should be recognized as a criterion to access medical assistance in dying. Advance Directives Expert Advisory Group (pp. 30-31) Timing of Completion of Declaration

RECOMMENDATION 12: Provinces and territories should request that the federal government make it clear in its changes to the Criminal Code that at any time following the diagnosis of a grievous and irremediable condition, a request for physicianassisted dying made through a valid patient declaration form may be fulfilled when suffering becomes intolerable. RECOMMENDATION 13: Within one year, provinces and territories, in collaboration with the federal government, should study whether patient declaration forms completed prior to the diagnosis of a grievous and irremediable medical condition might also be considered valid.

B. Suffering

The Supreme Court did not specify in Carter whether suffering is limited to physical suffering. Witnesses voiced different opinions, with some advocating for the inclusion of physical suffering only and others recommending that psychological suffering be included as well. 25 The Committee received at least one submission arguing that mental suffering is as severe as physical suffering and should not be excluded from MAID.26 In addition, the Supreme Court referred to suffering "from the knowledge that they lack the ability to bring a peaceful end to their lives at a time and in a manner of their own choosing," which would be psychological in nature.27 The requirements that the suffering is enduring and intolerable to the person are safeguards to ensure that someone in temporary or minor pain does not make a rash decision to die. In addition, the suffering must relate to a grievous and irremediable condition. Where mental illness is an issue, Dr. Gaind noted that what is considered enduring and intolerable suffering may be affected by the mental illness itself.28 The Committee has confidence that health care professionals will proceed cautiously in such cases, as in all cases, and ensure that all criteria

We have carefully considered the question of when the requirement for competency must be met. We have considered four possibilities and determined that physician-assisted dying should be permitted in three of those situations. Provinces and territories should advise the federal government to make clear in the Criminal Code that physician-assisted dying is permissible where: a) the patient is competent at all times from the initial request to the moment of provision of assistance; b) although competent and having a grievous and irremediable condition and experiencing intolerable suffering at the time of the completion of the patient declaration form, the patient lost competence between the completion of the patient declaration form and the provision of assistance., or although competent and having been diagnosed c) with a grievous and irremediable condition, but not yet experiencing intolerable suffering, at the time of the completion of a patient declaration form, the patient lost competence between the

I Iui

111111

6 End-of-Life Decision-Making

completion of the patient declaration form and the onset of the enduring intolerable suffering. We note that Carter requires that physician-assisted dying be provided to a competent adult. Scenario (a) above is clearly allowed under the Supreme Court of Canada's decision. However, the Supreme Court's decision is silent on scenarios (b) and (c) above. The first situation is the most straightforward: a patient who has a grievous and irremediable condition and is experiencing intolerable suffering and is competent until the moment of provision. This person should qualify for physician-assisted dying. The second situation involves a patient who, although competent and having a grievous and irremediable condition and experiencing intolerable suffering at the time of the completion of the declaration form, loses competency between the completion of the form and the provision. In this case, we believe that the person should be eligible for physician-assisted dying. The third situation is more complex. It may occur when a patient has been diagnosed with a condition that will remove competency, such as dementia, and has concerns about the quality of his or her end-oflife in that future. The patient is not experiencing intolerable suffering, but has been diagnosed with a grievous and irremediable medical condition. Should this patient be permitted to complete a patient declaration indicating that he or she would like to receive physician-assisted dying when certain conditions that the patient believes would constitute enduring

intolerable suffering are met, even if the patient will no longer be competent when those conditions are met? Our answer is yes. Particularly for certain degenerative diseases, we feel that to require that these patients be competent at the time physician-assisted dying is delivered would force many of them to trigger the act earlier in the course of their disease than necessary. Some people may be more comfortable with their diagnosis knowing that physician-assisted dying will be available to them after they lose competency. In such cases, physicians will need to take extra care to ensure that the patient understands and freely consents to his or her decision. The physician and patient should discuss the normal progression of the disease and how patients often respond and adapt to the disease. These conversations should continue throughout the progression of the disease (or up to the point of losing competence) to assess and confirm that the patient maintains his or her decision with respect to the future provision of physician-assisted dying. Special Joint Committee (pp. 21-4) E. Advance Request The Carter decision dealt with plaintiffs who would

remain competent while they faced significant physical decline. It did not address whether an individual who is not competent at the time of death could identify the circumstances in which he or she would choose MAID in advance. With respect to advance requests for MAID, witnesses and briefs outlined diverging opinions, from recommending not to

Situation

Situation

Situation

1

2

3

Patient meets eligibility criteria and is competent until the moment of provision.

Patient meets eligibility criteria and was competent at the moment of the request, but loses competency between the request and the provision.

Patient is competent and has been diagnosed with a grievous irremediable medical condition that will remove cognitive capacity potentially before the condition of intolerable suffering has been met.

Figure 1 Timing of the Completion of a Declaration

Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection

allow such requests, to allowing them only after an individual is diagnosed, to allowing advance requests to be written prior to any illness. There was general agreement however that, if requests are to be allowed in advance, the individual must be competent at the time the advance request is drafted. An advance request could be considered in three different situations:

•

where a person's request has been accepted but the individual loses competence before MAID takes place; where a person has been diagnosed with a grievous and irremediable condition but is not yet experiencing enduring and intolerable suffering and prior to diagnosis.

Professor Downie recommended that advance requests be permitted in the first two cases, but not the third. She argued that advance requests prevent the suffering of someone who has been approved for MAID but then loses competence and must continue to suffer. It also prevents individuals from ending their lives earlier than they would otherwise in order to avoid losing competence before the suffering becomes intolerable, something which was a major factor in the Carter case.29 Finally, an advance request allows the process to be undertaken before the suffering is enduring and intolerable. Otherwise, the person would have to continue to endure the suffering during the processing of the request and any waiting period.3° Linda Jarrett and the Hon. Steven Fletcher, both living with disabilities, also told the Committee they believed advance requests should be respected.3' Other witnesses also voiced support for advance requests.32 Wanda Morris, outgoing CEO of Dying with Dignity Canada, argued in a similar vein to Professor Downie: In their decision, the Supreme Court justices wrote that to force someone to choose between undergoing a premature, perhaps violent, death and enduring prolonged suffering is a cruel choice. We submit that unless the committee recommends that informed consent be allowed by advance consent, the injustice will continue. Nowhere does this play out more than around the issue of dementia . . . I think that what we do will actually be lifeaffirming if we are able to provide a clear advance consent mechanism.33 Ms. Morris explained that objective, verifiable criteria must be included in any such request to assist a health care team in assessing whether the criteria outlined in

the advance request have been satisfied. As examples, she listed being bedridden, being unable to feed, wash, or shave oneself or being unable to speak for 30 days or more. The same safeguards for a contemporaneous request must be in place for an advanced request to confirm informed consent and capacity.34 Jean-Pierre Menard, representing the Barreau du Quebec, noted that the panel of legal experts appointed by Quebec's government, of which he was a member, recommended allowing advance requests but that the legislation adopted in Quebec does not permit them for MAID. He explained that there were a number of questions that were raised about how to assess competence at the time an advance request is made: whether the individual would fully understand the decisions being made; how to know whether the individual had changed his or her mind; and whether a third party could act against the interests of the patient. He concluded that there was much debate, with valid arguments on both sides, and that a decision was made in Quebec to prioritize protection of the vulnerable.35 Jay Cameron from the Justice Centre for Constitutional Freedoms expressed concerns that advance requests could result in abuse if a patient becomes incompetent and that it would not be possible to verify if the request was made under duress. He also argued that it is too difficult to know how one will feel once in a changed state, such as when one is experiencing the symptoms of dementia.36 Michael Bach, Executive Vice-President of the Canadian Association for Community Living, argued that the requirement for the suffering of the patient to be intolerable "in the circumstances of his or her condition" bars the use of advance requests.37 Similarly, Prof. Trudo Lemmens from the Faculty of Law and Dalla Lana School of Public Health at the University of Toronto expressed concern in written submissions after his appearance before the Committee. He felt that someone with dementia who is still enjoying life could end up dying by MAID because he or she met the criteria related to suffering, such as not recognizing family members, that was included in his or her advance request for MAID.38 Dr. Jeff Blackmer, Vice-President, Medical Professionalism at the Canadian Medical Association, noted that the organization had not consulted its membership on the issue of advance requests because the issue was not addressed in Carter. However, he did say that implementing advance directives is "incredibly complex and difficult, because it's very hard to capture all of the nuances and the specifics of a very complicated medical condition and intervention ."39

11 111 i

II

11

.111

IIILLI

6 End-of-Life Decision-Making

Dr. Douglas Grant, Registrar and CEO of the College of Physicians and Surgeons of Nova Scotia, without taking a position on whether advance requests should be permitted, also noted that "a myriad of new issues" would need to be addressed if such requests were permitted." The Committee understands these challenges but is deeply concerned that by exduding individuals who want access to MAID but have lost competence, such individuals will be left to suffer or end their lives prematurely. This situation was exactly what the Carter decision sought to avoid. Allowing advance requests also provides comfort to individuals, reducing their psychological suffering, knowing that their lives will not end in a way that is against their wishes.41 Limiting the option of advance directives to individuals who already have a diagnosis makes it easier to ascertain that there was informed consent. At that point, the person knows more about what he or she may expect in the future to provide relevant direction in the request. The same safeguards to ensure competence and consent must be in place for advance requests, and consideration could be given to additional safeguards. Thought should be given to encouraging and possibly requiring health care practitioners to communicate regularly with their patients while they are still competent to ensure that their advance requests continue to reflect their wishes. The concerns of Dr. Blackmer, Dr. Grant and others will need to be examined as the system is put in place to minimize the risk of abuse and error, but the Committee is confident that this can and must be done to ensure the autonomy of Canadians and the protection of the vulnerable. The Committee therefore recommends: RECOMMENDATION 7 That the permission to use advance requests for medical assistance in dying be allowed any time after one is diagnosed with a condition that is reasonably likely to cause loss of competence or after a diagnosis of a grievous or irremediable condition but before the suffering becomes intolerable. An advance request may not, however, be made, prior to being diagnosed with such a condition. The advance request is subject to the same procedural safeguards as those in place for contemporaneous requests. Conscientious Objection Expert Advisory Group (pp. 44-7) Duty to Inform

All end-of-life options (including palliative care and physician-assisted dying) should be discussed with

RECOMMENDATION 31: Conscientiously objecting health care providers should be required to inform patients of all end-of-life options, including physicianassisted dying, regardless of their personal beliefs.

patients in a respectful way regardless of the provider's beliefs. Such requirements are often already included in professional codes of ethics. For example, the Canadian Medical Association's Code of Ethics requires that physicians provide patients with the information they need to make informed decisions about their medical care.

RECOMMENDATION 32: Conscientiously objecting health care providers should be required to appropriately inform their patients of the fact and implications of their conscientious objection to physician-assisted dying. Any ongoing treatment of the patient must be provided in a nondiscriminatory manner.

Conscientiously objecting providers have a duty to inform new patients and, when appropriate, current patients of their position on physician-assisted dying. This allows patients to make an informed choice with respect to enrolling with or retaining that provider. Regulatory authorities and governments should produce educational materials for physicians and patients to help navigate these discussions. We also encourage all physicians to initiate and regularly revisit conversations with patients around advance care planning, the goals of care, and the patient's end-of-life wishes. Duty to Care for the Patient

RECOMMENDATION 33: Conscientiously objecting health care providers should be required to either provide a referral or a direct transfer of care to another health care provider or to contact a third party and transfer the patient's records through the system described in Recommendation 4.

To facilitate access to physician-assisted dying, conscientiously objecting health care providers must refer, transfer care or contact a third party in a timely manner and transfer records to another health care provider or third party, as required. There is no expectation that they will assess the patient's eligibility for physician-assisted dying.

Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection

We heard from some stakeholders who believed Duties of Institutions that a referral in the context of physician-assisted dying All institutions have certain responsibilities i n common, implied that the conscientiously objecting physician many of which align with our recommendations related had assessed the patient, found them eligible for to physicians. physician-assisted dying and was transferring care for the purposes of providing the service. While this is an RECOMMENDATION 34: All institutions should be option available to those who object to the provision of required to inform patients/residents of any Instituservice, we did not want to imply that assessment is an tional position on physician-assisted dying, includobligation, or that this was the only option available to ing any and all limits on its provision. providers. When we discuss a direct transfer of care, we envisage a situation where a patient has requested This recommendation will ensure that patients have physician-assisted dying and has already discussed clarity on what is permitted within the facility. They end-of-life options with their health care provider. We can then make informed decisions with respect to recognize that some providers view a transfer of care as whether to enter or remain in the facility. morally preferable to referral because, unlike referral, it is taken to neither explicitly nor implicitly affirm the RECOMMENDATION 35: Provinces and territories service sought by the patient. A health care provider should prohibit any requirement by institutions that could transfer the patient to another health care propatients give up the right to access physician-assisted vider for the assessment and treatment of the patient's dying as a condition of admission. medical condition and, if the patient meets the eligibility criteria, provision of physician-assisted dying. The receiving health care provider must be someone who is willing and able to accept the person as a patient and RECOMMENDATION 36: Provinces and territories does not conscientiously object to physician-assisted should prohbit any requirement by institutions that dying. Following the transfer of care, the receiving physicians refrain from the provision of physicianhealth care provider would follow the pathway outassisted dying external to the non-participating institulined earlier in this report. tion. In addition, employment conditions or privileges For those conscientiously objecting health care should not be negatively impacted in any way. providers who are not willing to provide a direct transfer of care on conscience or religious grounds, we provide the option of contacting and transfer- Taken together, the two recommendations above limit ring the patient's records through a publicly-funded the power of institutions to restrict the autonomy of pasystem designed to ensure that patients are able to tients who may wish to seek physician-assisted dying at access a health care provider willing to accept them a later date, and providers, who may wish to provide as a patient for assessment and, if the eligibility cri- physician-assisted dying outside a faith-based institution teria are met, to provide physician-assisted dying. If that does not offer physician-assisted dying. They reprethis system is not in place, conscientiously objecting sent an appropriate balance between patient and physician providers should be required to make a direct transfer autonomy, patient access, and institutional autonomy. of care to a health care provider who is willing and able to accept the person as a patient and does not Duties of Non Faith-Based Institutions conscientiously object to physician-assisted dying. It is imperative that the burden of the transfer to anRECOMMENDATION 37: Non faith-based instituother physician, institution or third party not fall on tions, whether publicly- or privately-funded, must the patient seeking the care. not prevent physician-assisted dying from being As described in Recommendation 4, we recomprovided at their facilities. mend a system where regional health authorities develop a publicly-funded care coordination system to help facilitate access to physician-assisted dying. One Governments have significantly more levers to inof this system's key functions should be to receive re- fluence the policies and practices of institutions that are quests from unwilling or unable health care providers funded in whole or in part by public funds. However, as and transfer patients to providers who are willing and a point of principle, we believe that physician-assisted able to receive new patients. F. . .1 dying should be available wherever people are living and

II

ui

IIILLJ

6 End-of-Life Decision-Making

dying. This includes privately-funded institutions. Recognizing that not all institutions will feel capable of providing physician-assisted dying for a variety of reasons, including size, geography ar d mandate, we concluded that institutions should be permitted to offer a patient transfer to another institution, as long as the receiving institution can and will provide a health care provider who is willing and able to accept the person as a patient, to assess whether the criteria for access to physician-assisted dying have been met, and provide physician-assisted dying where the criteria for access have been met. Duties of Faith-Based Institutions

RECOMMENDATION 38: raith-based institutions must either allow physician-assisted dying within the institution or make arrangements for the safe and timely transfer of the patient to a non-objecting institution for assessment and, potentially, provision of physician-assisted dying. The duty of care must be continuous and non-discriminatory.

Faith-based institutions have a duty to care for and not abandon the patients within their institution. While they should not be recuired to provide access to physician-assisted dying, they must still ensure access for patients who wish to seek it. When a patient makes a request for physician-assisted dying, faith-based institutions should be required to either allow for the assessment and provision of physician-assisted dying within the institution or make arrangements for an effective transfer of the patient to a non-objecting institution. This transfer must also include the transfer of all relevant records, and must be made to a non-objecting institution where the patient's medical condition can be assessed and treated by a health care provider who is willing and able to assess whether the patient meets the eligibility criteria for physician-assisted dying and, if so, can provide assistance. If a safe and timely transfer to a non-objecting institution is not possible, the objecting institution must allow an outside health care provide: to assess the patient and, if the eligibility criteria for physician-assisted dying are met, to provide assistance. The receiving outside health care provider would follow the pathway for physicianassisted dying as outlined previously in this report. Special Joint Committee (pp. 25-27) A. Conscientious Objection to Participating in Medical Assistance in Dying

The External Panel's report noted that "the medical profession is divided over the issue of MAID."42 Many

witnesses who appeared before the Committee, and briefs/letters that were submitted to the Committee, discussed the extent to which health care practitioners should be able to refuse to participate in MAID for reasons of conscience.' No one was of the opinion that a health care practitioner should be obliged to perform MAID. As the Supreme Court of Canada stated in Carter, "Rin our view, nothing in the declaration of invalidity which we propose to issue would compel physicians to provide assistance in dying."44 It was argued by some witnesses that strong protections for health care practitioners who refuse to participate for reasons of conscience need to be put in place, including the possibility that such protection be established in legislation45 Other witnesses were concerned about the effect a practitioner's refusal to participate in MAID would have both on the individual who was seeking an assisted death and on the availability of MAID more broadly. As Vyda Ng from the Canadian Unitarian Council told the Committee, " [i] t's very much in keeping with Canadian values to put the needs and wishes of Canadians ahead of the values of individual doctors and institutions, and to respect each person's dignity at the most traumatic period of their lives."46 Some witnesses and submissions to the Committee recommended that a practitioner who conscientiously objects to MAID should be required to provide an effective referral or transfer of care for their patient,47 while some felt that referring the individual to a third-party organization should be sufficient.48 Joanne Klineberg from the Department of Justice noted that provinces and territories "have legislation and policies in relation to the rights of physicians to refuse to partake in certain types of medical practices, so it is definitely something that the provinces and territories already are responsible for."49 In Quebec, a physician must notify a designated individual if he or she refuses to participate in MAID so that a willing physician may be identified.5° The Committee notes that the Supreme Court of Canada in Carter stated that the Charter rights of patients and physicians will need to be reconciled.5' The Committee believes that having health care professionals who conscientiously object to MAID provide an effective referral for a patient who seeks MAID is an appropriate balancing of the rights of patients and the conscience rights of physicians. The Committee therefore recommends: RECOMMENDATION 10 That the Government of Canada work with the provinces and territories and their medical regulatory bodies to establish a process that respects a health care practitioner's freedom of conscience while at the same time respecting the needs of a patient who seeks medical assistance

Expert Advisory Group/Special Joint Committee: Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection

in dying. At a minimum, the objecting practitioner must provide an effective referral for the patient. Witnesses and briefs also addressed whether a health care facility should be permitted to refuse to either provide MAID or to allow MAID to be provided on its premises. One witness told the Committee that in Quebec, hospices (which the witness stated are largely privately funded) sought and received an exemption from having to provide MAID.52 A number of witnesses

argued, and the Committee also believes, that if a health care facility is publicly funded, it must provide MAID.53 The difficulty in transferring a patient from one facility to another was highlighted.54 The Committee recommends therefore: RECOMMENDATION 11 That the Government of Canada work with the provinces and territories to ensure that all publicly funded health care institutions provide medical assistance in dying.

References 1. PDAM, Evidence, 26 January 2016, 1805 (Pelletier). 2. PDAM, Evidence, 25 January 2016, 1240 (Hogg). 3. PDAM, Evidence, 2 February 2016, 1940 (Grant). 4. See, for example, PDAM, Evidence, 4 February 2016, 1640

5.

6. 7. 8. 9. 10.

11. 12.

13.

14. 15. 16.

(Carmela Hutchison, President of Disabled Women's Network of Canada) (DAWN); PDAM, Evidence, 4 February 2016, 1920 (Bach). Regarding the suggestion of age 25, see Colette Squires, Physician Assisted Dying Public Consultation, January 30, 2016, in Langley, B.C., submission to the Committee, p. 4. Provincial-Territorial Report, Recommendation 17. Also see, for example, PDAM, Evidence, 2 February 2016, 1735 (Dr. Derryck Smith, Chair of Physicians Advisory Council, Dying with Dignity Canada). PDAM, Evidence, 2 February 2016, 1735 (Smith). PDAM, Evidence, 18 January 2016, 1415 (Klineberg). PDAM, Evidence, 26 January 2016, 1835 (Gibson); PDAM, Evidence, 1 February 2016, 1100 (Pastine). PDAM, Evidence, 26 January 2016, 1810 (Pelletier). PDAM, Evidence, 3 February 2016 (Mary Shariff, Associate Professor of Law and Associate Dean Academic, University of Manitoba, Canadian Paediatric Society). PDAM, Evidence, 2 February 2016, 1815 (Smith). PDAM, Evidence, 4 February 2016, 1730 (Margaret Birrell, President, Alliance of People with Disabilities Who Are Supportive of Legal Assisted Dying Society); 4 February 2016, 1900 (Dr. John Soles, President, Society of Rural Physicians of Canada). PDAM, Evidence, 4 February 2016, 1915 (Dr. Hartley Stem, Executive Director and CEO, the Canadian Medical Protective Association). An Act respecting end-of-life care, section 5. A.C. v Manitoba (Director of Child and Family Services), 2009, sec 30, para. 69. In A.C. v Manitoba (Director of Child and Family Services), 2009 scc 30, the Supreme Court of Canada discusses the ability of minors to consent to medical treatment in the context of protective legislation that allows a court to authorize treatment for a child when it deems it to be in the child's best interest. At para. 46, Justice Abella (for the majority) states: The latitude accorded to adults at common law to decide their own medical treatment

17.

18.

19.

20.

had historically narrowed dramatically when applied to children. However the common law has more recently abandoned the assumption that all minors lack decisional capacity and replaced it with a general recognition that children are entitled to a degree of decision-making autonomy that is reflective of their evolving intelligence and understanding. This is known as the common law "mature minor" doctrine. As the Manitoba Law Reform Commission noted, this doctrine is "a well-known, wellaccepted and workable principle which . . . raise's] few difficulties on a day-to-day basis" (Minors' Consent to Health Care (1995), Report #91, at p. 33). The doctrine addresses the concern that young people should not automatically be deprived of the right to make decisions affecting their medical treatment. It provides instead that the right to make those decisions varies in accordance with the young person's level of maturity, with the degree to which maturity is scrutinized intensifying in accordance with the severity of the potential consequences of the treatment or of its refusal PDAM, Evidence, 18 January 2016, 1530 (Jeanette Ettel, Senior Counsel, Human Rights Law Section, Department of Justice). PDAM, Evidence, 26 January 2016, 1750 (Benoit Pelletier, Member, External Panel on Options for a Legislative Response to Carter v. Canada). PDAM, Evidence, 28 January 2016, 1850 (Downie). See also, for example, Centre for Inquiry, p. 4, and PDAM, Evidence, 26 January 2016, 1900 (Taylor). See, for example, Living with Dignity, Recommendations

for the Special Joint Committee on Physician Assisted Dying, p. 3, written submission to the Committee; Derek B.M. Ross & Johnathan R. Sikkema, Christian Legal Fellowship, Submission of the Christian Legal Fellowship to the Special Joint Committee on Physician-Assisted Dying, 1 February 2016, p. 4. 21. PDAM, Evidence, 27 January 2016, 1935 (Dr. K. Sonu Gaind, President, Canadian Psychiatric Association). 22. PDAM, Evidence, 3 February 2016, 1805 (Dr. Tarek Rajji, Chief, Geriatric Psychiatry, Centre for Addiction and Mental Health). 23. Carter, 2015, para. 127.

H Ju

111111

6 End-of-Life Decision-Making

24. Regarding Belgium, for example: neuropsychiatric 38. disorders were 1.2% of cases in 2004/05, 2.8% (or 58 cases) in 2010/11 and 3.7% (or 67 cases) in 2013/14 according to Trudo Lemmens, Why Canada Should Avoid 39. a Belgian-Style Regulatory Regime for Physician Assisted 40.

Dying, Memorandum for the Special Joint Committee on Physician-Assisted Dying, written submission to the Com- 41. mittee, p. 6. 42. 25. See, for example, PDAM, Evidence, 4 February 2016, 43. 1705 (Somerville), regarding limiting MAID to physical suffering; and PDAM, Evidence, 1 February 2016, 1155 (Wanda Morris, Dying with Dignity), regarding psychological suffering. 44. 26. Marcia Hogan, Brief to the Joint Committee on Phys- 45. ician Assisted Dying, submission to the Committee, p. 2. 27. Carter, 2015, para. 14. 28. PDAM, Evidence, 27 January 2016, 1935 (Gaind). 29. Carter, 2015, para. 57-58. 30. PDAM, Evidence, 28 January 2016, 1850 (Downie). 46. 31. PDAM, Evidence, 28 January 2016, 1735 (Jarrett); 1805 47. (Fletcher). 32. See, for example, PDAM, Evidence, 1 February 2016, 1105 (Paterson); PDAM, Evidence, 4 February 2016, 1655 (Angus Gunn, Counsel, Alliance of People with 48. Disabilities Who Are Supportive of Legal Assisted Dying Society); Centre for Israel and Jewish Affairs, Brief for the Special Joint Committee on Physician Assisted Dying, 49. 50. submission to the Committee, p. 4-5. 33. PDAM, Evidence, 1 February 2016, 1120 (Morris). 34. Ibid., 1155. 51. 35. PDAM, Evidence, 25 January 2016, 1255 (Menard). 52. 36. PDAM, Evidence, 1 February 2016, 1205 (Jay Cameron, 53. Barrister and Solicitor, Justice Centre for Constitutional Freedoms). 37. PDAM, Evidence, 4 February 2016, 1925 (Bach). See also, for example, POAM, Evidence, 4 February 2016, 1930 (Gerald Chipeur, Lawyer, as an individual). 54.

Trudo Lemmens, Response to Comments Made during the Committee Hearings of January 28, 2016, 2 February 2016, submission to the Committee. PDAM, Evidence, 27 January 2016, 1750 (Blackmer). PDAM, Evidence, 2 February 2016, 1925 (Grant). See also, for example, PDAM, Evidence, 3 February2016, 1850 (Radji). Provincial-Territorial Report, p. 31. External Panel Report, p. 98. The issue of freedom of conscience of pharmacists was raised by the Canadian Pharmacists Association (CPbA), PDAM, Evidence, 27 January 2016 (Phil Emberley, Canadian Pharmacists Association). Carter, 2015, para. 132. See, for example, PDAM, Evidence, 27 January 2016, 1725 (Branigan); PDAM Evidence, 3 February 2016, 1905 (Hashmi); 1705 (Collins); Letter to Minister WilsonRaybould and Minister Philpott, Canadian Conference of Catholic Bishops, 20 January 2016. PDAM, Evidence, 3 February 2016, 1705 (Ng). See for example, PDAM, Evidence, 26 January 2016, 1900 (Gibson); PDAM, Evidence, 1 February 2016, 1125 (Moms); Rhonda Morison, submission to the Committee, 30 January 2016. PDAM, Evidence, 1 February 2016, 1105 (Paterson), submission to the Committee, Vivre dans la Dignite 2016; Ellen Agger, submission to the Committee, 1 February 2016. PDAM, Evidence, 18 January 2016, 1535 (Klineberg). An Act Respecting End-of-Life Care, RSQ c S-32.0001, section 31. Carter, 2015, para. 132. PDAM, Evidence, 2 February 2016, 1845 (Baxter). See, for example, PDAM, Evidence, 3 February 2016, 1700 (Ng); British Columbia Humanist Association, Allow Assisted

Dying for All Who Choose It. A Brief for the Special Joint Committee on Physician-Assisted Dying, 25 January 2016; Lori Goodwin, submission to the Committee, 30 January 2016. Evidence, 2 February 2016, 1850 (Baxter).

PDAM,

6.4 The Concept of Death and Its Practical Implications

Defining Death Mister Browne The end of life is death, and death is typically thought to be a time of great behavioural significance. For example, it is normally though: that we can do certain things such as routinely remove all life-support, mine an organ donor's body for transplantable organs, or bury a person after they are dead but not before, and thus that determining when exactly a person is dead This article was written specifically for this book.

is a matter of importance. It is, however, surprisingly difficult to identify that time. The aim of this essay is to explain why there is a problem and to canvass the main solutions that have been offered.

Definitions of Death The earliest understanding of death was in terms of the loss of heartbeat and respiration. There was always a problem of saying how long these functions must be absent before a person is dead, and

Browne: Defining Death

since that exact interval was not (and still is not) known, death was defined as the "prolonged" absence of these functions. This gave rise to worries about premature burial, but the real demand for greater precision came from the invention of ventilators and the advent of transplantation. Ventilators made it possible to sustain cardiorespiratory function long after it was no longer spontaneous. This provoked the question of whether such a patient is alive or a corpse being ventilated, and the possibility of transplantation made it important to know. If the person could be declared dead while on the ventilator, transplantation could begin immediately with organs kept in the freshest possible condition. If not, the ventilator would have to be disconnected and the patient declared dead by cardiocirculatory criteria before organs could be excised. The whole-brain definition of death was invented to solve this problem. On this view, death occurs when the upper brain (which sponsors consciousness) and lower brain (which controls heartbeat and respiration) have permanently ceased to function. This means that once a person no longer has spontaneous respiration and heartbeat, they are dead. Death can thus be declared either clinically by the permanent cessation of spontaneous heartbeat and respiration or neurologically by the irreversible cessation of all brain function. Patients on ventilators with artificially supported respiration and heartbeat could thus be declared dead. This allowed for an uncontroversial termination of treatment and (more importantly) facilitated transplantation. Thus, whole-brain death became the standard definition of death. But it had critics. Some argued that it diagnosed death too soon on the ground that a person with artificially supported heartbeat and respiration was still alive, perhaps not alive in a state worth preserving, but alive nonetheless and not dead until all spontaneous and artificially supported cardiorespiratory function had irreversibly ceased. Others argued that the whole-brain definition pegs the time of death too late. On this view, a person is dead as soon as the possibility of consciousness is permanently lost, and that occurs as soon as all higher-brain function has irreversibly shut down. Other brain activity is no more relevant than continued cellular activity in non-vital parts of the body is for the whole-brain account. All the above definitions define death as an irreversible state, but "irreversibility" is an undefined term, and it is not easy to define it. Does it mean that the state will not reverse itself or cannot be reversed, and if the latter, is the impossibility logical, empirical,

technical, legal, or moral? The literature on this is large and inconclusive, and it is worrisome that a key term that cannot be eliminated remains undefined. A fourth definition of death has been added to the list in the past 15 years—cardiocirculatory death for DCD—and is valid only for what is known in Canada as "donation after cardiocirculatory death (DCD)" and elsewhere as "non-heart-beating organ donation (NHBoD)." The aim of this protocol is to increase the supply of transplantable organs by enabling death to be declared before whole-brain death is diagnosed and thus enlarging the donor pool. In DCD, after the patient or family has consented to the procedure, the ventilator is removed, and once heartbeat and respiration cease, the patient is continuously observed for two to ten (the interval varies from centre to centre, the most common being five) minutes. If there has been no pulse for that time, death is declared, and transplantation starts to take place.

Selecting a Definition That is the menu of choices of when to declare death. To have a uniform definition of death—i.e., a single all-purpose definition of death for a society—one of the first three of these (the last is only for DCD) must be chosen, but it is not clear how to select one. It is sometimes suggested that a committee of physicians or health care professionals should choose, but the decision is not a matter for empirical investigation. Science can tell when cardiorespiratory, whole-brain, and higher-brain death occur, but it cannot tell us which one should be picked as the death of a human being. Deciding which of these candidates is the time of death is, rather, more like deciding what the age of majority is, and one may suggest that the question should be decided in the same way. We decide on the age of majority by deciding when it is appropriate to allow people to vote, drive, drink in public, etc. and identify that time as that age. Similarly, one may contend that we should proceed by determining when death-behaviour is appropriate—i.e., behaviour that is appropriate if and only if a person is dead—and identify that time as the time of death. This, however, does not yield a clear answer. Death-behaviour covers a wide range, including disparate things such as when it is possible to routinely withdraw all health care, take organs for transplantation, use the body for experimental or teaching purposes, begin an autopsy, initiate burial or cremation procedures, prosecute a crime as murder, inherit an estate, and remarry without divorce. It is not clear that all these items of behaviour

MI Bli

II

II

Al

Al

111111

6 End-of-Life Decision-Making

become appropriate at the same time. Indeed, it is pretty clear that they do not. But if they do not, this way of selecting between the candidates for the time of death will not work. Moreover, even if all death-behaviour did coalesce at a single point, the original aim in seeking a definition of death would be lost. That aim was to define death so that deathbehaviour could be initiated, and on this account we have to know when death-behaviour is appropriate before we can define death. This difficulty in choosing between candidates for the time of death has led some to abandon a search for a single uniform definition of death and accept instead different definitions for different purposes. The age of majority differs depending on whether we are talking about the right to drive, drink in public, vote, marry, live on one's own, and so forth, and the suggestion is that we can have different definitions of death for different items of death-behaviour. This has in turn led to a more radical suggestion—namely, to allow individuals to pick th e times that they would like various items of death-behaviour initiated for them rather than have them designated by the state. Robert Veatch, for example, has suggested that society adopt whole-brain deat:i as a default definition but allow individuals to deviate from that by selecting other definitions of death for different deathbehaviour. On his proposal, :ndividuals would not be able to choose a definition of death earlier than the irreversible cessation of consciousness or later than the loss of all circulatory and respiratory functions, but within that range they should be free to say what death-behaviour can be visited on them and when. An even more radical view has also been proposed. On this view, we should give up the so-called "dead donor rule." Rather than adjust the definition of death to allow for death-behaviour to occur at different times for different purposes, the suggestion is to uncouple the definition of death from its traditional death-behaviour. If it can be determined what can be done to whom and when, legislation permitting those items of behaviour could be passed without using the word "death" at all. For example, instead of saying that transplant proceedings can only begin when a person is dead (or dead in the sense that would license transplantation), we could say that transplantation cannot begin until the heart has stopped beating with or without mechanical assistance (or whatever conditions one thinks apt). The same would apply for deciding when it is appropriate to distribute an estate, remarry without committing bigamy, and all the other items of traditional death-behaviour. In this spirit, Robert D. Truog ("Is It Time to Abandon Brain

Death?" cited in the section "Defining Death" in the Suggested Further Readings) has proposed that an individual could authorize organ retrieval under general anaesthetic without first undergoing an orchestrated withdrawal of life support in which the patient is declared dead. On this view, organ procurement would be controlled by considerations of consent and harm to the donor rather than the concept of death. When the patient does not want to live in that state and wants to be an organ donor, this could be viewed as a species of euthanasia—euthanasia by organ donation, or if one prefers, organ donation by euthanasia. Death in Canada This ends our quick and very incomplete tour of the hard and complex question of defining death. The debate over whether to seek a uniform definition of death or individualize death, and what would be the best way to provide either, continues without a clear end in sight. The Canadian Law Reform Commission in 1981 recommended that Canada adopt the whole-brain definition, but that was not accepted, and Canada remains without a legislated definition. Nonetheless, in Canada (and the world generally) death is understood either as whole-brain death—diagnosed by its cardiorespiratory criteria for the vast majority of patients, its neurological criteria for organ donors on ventilation—or the modified cardiocirculatory criteria used for DCD. Further Reading The following readings will be useful to those who would like to pursue the topic of defining death. References to the articles by the authors mentioned here are in the Suggested Further Readings under the heading "Defining Death." Robert Truog provides an explanation and critique of whole brain death and gives an exceptionally clear survey of the debate on defining death. James L Bernat brings us to the frontier of the development of DCD and identifies issues that are yet to be resolved and yet to come. Readings representing the main positions on defining death are: "Higher-Brain Death," Green and Wikler; "The Individualization of Death," Emanuel; "Eliminating the Concept of Death," Dworkin. The Canadian version of DCD is stated by the Canadian Council for Donation and Transplantation (ccDT) and the American version by the Institute of Medicine (IOM). The reading by Gail A. Van Norman identifies some of the many medical and moral complexities that arise with DCD.

Cases

6.5 Cases

Case 1

Nancy B.: Withdrawing Life-Sustaining Treatment Nancy B. (her last name was never disclosed) was a 25-year old woman who suffered from Guillain-Barre syndrome, a disabling neurological disorder. In January 1991, she was informed that her condition was irreversible. Her respiratory muscles had atrophied, and degeneration of the motor nerves left her tied to her hospital bed, entirely dependent for survival on mechanical ventilation by intubation. In the months following this, Nancy B. determined that she was unwilling to survive in her intubated, dependent existence. She was not diagnosable as clinically depressed but was distressed that life could afford her nothing more than limited head movement and utter dependency for comfort on others and a machine. Her intellectual faculties were intact, and she was lucid and logical in expression of her wish to be disconnected from the mechanical respirator. She knew that the consequence would be that she would die in a very short time. Her mother said that her family had come to accept her wish, and the Quebec Superior Court judge (who himself attended her in hospital) found that Nancy B.'s preference that Source: Taken from Bernard M. Dickens, "Medically Assisted Death: Nancy B. v. Hotel-Dieu de Quebec," McGill LawJournal 38 (1993), 1053-70.

Case 2

Dr Nancy Morrison: Nonvoluntary Active Euthanasia of an Adult In 1996, 65-year-old Halifax, Nova Scotia, resident Paul Mills was suffering from terminal esophageal cancer that had required previous removal of his esophagus and repositioning of his stomach in an attempt to repair the gap. By mid-October of 1996, Mr Mills had experienced nine additional operations at two different hospitals due to a severe post-surgical infection and was profoundly depressed, expressing a wish to die. Between 15 October and 6 November, his weight dropped by 19 kg (42 lbs), and the infection had developed to such a degree that

artificial respiration be discontinued was "real and enlightened." The judge (Dufour J.) posed himself the questions: "Can the conduct of a physician who stops the respiratory support treatment of his patient at the freely given and informed request of the patient, and so that nature may take its course, be characterized as unreasonable? Or does such conduct denote wanton and reckless disregard?' Dufour J. answered both questions, "I do not believe so." The resulting judgment was that, on expiration of the time for appeal from the judgment, the court (1) permits Nancy B.'s attending physician to stop the respiratory support treatment when Nancy B. so desires and (2) permits the physician to request from the hospital the necessary assistance in circumstances such as these so that everything can take place in a manner respecting her dignity. Dufour J. found that the circumstances of Nancy B.'s proposed death would constitute neither homicide nor suicide but rather natural death. In support of this, Dufour J. cited the extremely influential New Jersey Supreme Court case of Re Conroy to the effect that: "Declining life-sustaining medical treatment may not properly be viewed as an attempt to commit suicide. Refusing medical intervention merely allows the disease to take its natural course; if death were eventually to occur, it would be the result, primarily, of the underlying disease, and not the result of a self-inflicted injury" The citation for the case is: Nancy B. v. Hotel-Dieu de Quebec [1992] R.J.Q. 361, 86D.L.R. (4th) 385 (Sup. Ct.)

healing from his prior surgeries had become impossible. Pus continuously oozed from a gap in the chest wall, which could not be covered. Estimates suggest that there were approximately 10 chest tubes, stomach tubes, and Ns connecting Mr Mills's body to machines and equipment. A DNR (Do Not Resuscitate) order was put in place, and on 9 November, when it became clear that there was no hope of recovery the family agreed to withdraw active life support in favour of palliative care. On the morning of 10 November, Mr Mills remained in his heavily sedated state and was still receiving narcotics for pain. His feed tubes and antibiotics were withdrawn, and he was given increasing doses of Dilaudid for pain. At approximately 12:30 p.m., the respirator was withdrawn; however, the patient did not then die

342

6 End-of-Life Decision-Making

as expected but rather struggled for air for quite some time. Pus oozed from his incisions as he laboured for breath. His attending nurse, Elizabeth Bland-Maclnnes, described his struggle for air as "a horrible and hideous scene" and said that "it was beyond the shadow of a doubt the worst death I've ever witnessed." The only HCP consistently present at the bedside during these hours, Bland-Maclnnes had no doubt that Mr. Mills was suffering. However, the icu resident, Dr Cohen, only described Mr Mills as "in distress" and "apparently in discomfort" but would not commit to stating that the patient was conscious and therefore aware and suffering. There is speculation that Mr Mills lost consciousness when the tubes were removed that morning and therefore some question as to whether he was experiencing an agonizing death or not. Whether or not Mr Mills was conscious and requesting assistance to die is relevant to determining whether this was a case of voluntary as opposed to nonvoluntary active euthanasia. What is clear is that Nurse Bland-Maclnnes was deeply disturbed by Mr Mills's apparent distress. She communicated to Dr Nancy Morrison her exasperation that sedatives and narcotics were ineffective. Reports indicate that the nurse begged Dr Morrison to help end the patient's suffering. At 2:52 pm, Dr Morrison injected nitroglycerin into the Iv line in an unsuccessful attempt to decrease blood pressure and thereby end the patient's suffering. Seven minutes later, Dr Morrison returned to inject Mr Mills with potassium chloride, and within minutes he was dead.

In May 1997, a hospital physician who had seen an internal document about the case and was afraid that the hospital would attempt to cover it up reported the incident to the police as an instance of active euthanasia. This resulted in Dr Morrison being charged with first-degree murder on 6 May. In February 1998, Judge Hughes Randall declined to commit Dr Morrison to stand trial. He noted that Mr Mills had been given extraordinary amounts of Dilaudid, morphine, and other painkillers prior to Dr Morrison's involvement in his treatment. These potentially lethal drugs had been legally administered earlier in an unsuccessful attempt to relieve Mr Mills's apparently intense suffering after being removed from the respirator and could have been the cause of death rather than the potassium chloride Dr Morrison injected. Because Dr Morrison used drugs that have no pain-killing, anaesthetic, or sedative properties, Nova Scotia's College of Physicians and Surgeons stated that the use of these drugs had no place in Mr Mills's medical management.2 In delivering their formal reprimand to Dr Morrison, the college labelled her actions as "inappropriate and outside the bounds of acceptable medical practice."1 The college would have preferred Dr Morrison to have chosen one or more of these alternatives: (1) consulting with other intensive care physicians; (2) ensuring correct functioning of the Iv; (3) using alternative medications; (4) re-establishing a supplementary airway; or (5) continuing with the chosen course of treatment, knowing that death was imminent.2 This reprimand now appears on Dr Morrison's formal discipline record held at the college.

Notes 1.

Barney Sneiderman, and Raymond Deutscher, "Dr. Nancy Morrison and Her Dying Patient: A Case of Medical Necessity" (2002), www.umanitoba.ca/centres/ethics/ articles/BarneyART3.pdf (accessed 11 December 2007).

Case 3

Tracy Latimer: Nonvoluntary Active Euthanasia of a Minor On 23 November 1980, T-acy Latimer of Wilkie, Saskatchewan, suffered from oxygen deprivation during her birth, which resulted in an extreme form of cerebral palsy, an incurable, permanent brain injury typically affecting muscle control and movement.' Tracy was a spastic quadriplegic, unable to walk, talk, or feed herself; indeed, she could not even

2.

College of Physicians and Surgeons of Nova Scotia, "Morrison Accepts College Reprimand" (1999) [online press release), www.cpsns.ns.ca/pr/morrison_3_30_99 .html (accessed 16 December 2007).

swallow—her throat had to be manually massaged to allow her to ingest food. Her full-brain impairment left Tracy with the mental age of a four- to fivemonth-old infant and a tendency toward seizures. Seizure medication reduced the number of seizures to approximately five per day but decreased Tracy's respiration and digestive systems, further burdening these critically over-taxed systems and diminishing Tracy's quality of life. By the age of four, increasing muscle tension and atrophy caused Tracy continuous pain, which was treated with her first of three surgeries. It was at this

Cases

time the family discovered that Tracy could not take analgesics stronger than regular Tylenol because of concerns about interactions with her anti-seizure and anti-convulsion medications. This would have significant implications for Tracy's quality of life as her condition deteriorated. A second surgery, in 1990, was meant to relieve the 10-year-old Tracy's muscle tension, redistribute strain, and alleviate pressure on her hips to minimize the potential for hip dislocation; however, the procedure was only minimally successful and relieved pressure for just a few months. She developed scoliosis—curvature of the spine—and her hip became dislocated as muscles were pulled away from the bones. Problems with eating and digestion, including vomiting, developed as the spine and vital organs were compressed by the scoliosis, and the frequency of her seizures increased.' In 1992, the scoliosis had developed to 75 per cent off perpendicular. Tracy required a third surgery, to have a steel rod inserted on either side of her spine, held secure by holes drilled into her pelvic bones. This surgery, like the previous one, relieved her symptoms for only a few months and introduced a whole new set of problems. Since the steel rods made Tracy's body rigid, there were few positions in which she could rest. She developed painful pressure sores and skin deterioration, had difficulty sleeping, and was losing weight. In 1993, as Tracy's condition deteriorated and the physicians recommended more surgeries, Robert Latimer, Tracy's father, became increasingly concerned about his daughter's quality of life. A proposed fourth operation would involve surgically removing Tracy's upper thigh bone, a procedure that the family was told would be excruciatingly painful to endure, with the pain continuing long afterward. Additional surgeries would be required to mitigate the pain Tracy felt in other joints. Tracy would have to endure all of this with inadequate pain management. The Latimers felt that further surgery would be futile and would, in

fact, constitute a mutilation of their daughter.3 In consideration of her situation, Mr Latimer decided that Tracy's death was a better alternative than constant surgery and unremitting pain. On Sunday, 24 October 1993, while his family was at church, Robert Latimer placed 12-year-old Tracy in the cab of his pickup truck and piped in carbon monoxide fumes from the truck's exhaust, killing his daughter by carbon monoxide poisoning. Mr Latimer was found guilty of the seconddegree murder of his daughter both at trial and on appeal. The jury recommended that he be eligible for parole after one year rather than have to serve the mandatory minimum sentence for second-degree murder-10 years before parole is considered. In December 1997, Justice Ted Noble granted Latimer a constitutional exemption from the minimum sentence, saying that imposing the minimum sentence on Latimer in this case of mercy killing, as opposed to cold-blooded killing, would constitute "cruel and unusual punishment." In November 1998, the Saskatchewan Court of Appeal overturned Noble's ruling and imposed the mandatory sentence: 25 years in prison with no chance of parole for 10 years. In December 2007, after having served 10 years in prison, Latimer was refused day parole. The parole board reported that they were "struck by Latimer's lack of insight into the crime he had committed. . . . Latimer should stay in prison and receive more counselling."4 Latimer had told the parole board that he felt no guilt for killing his daughter and had stated repeatedly during the hearing that it was the right thing to do. When asked if he had the moral authority to take someone's life, Latimer replied that the laws were less important than the welfare of his daughter, who was in a lot of pain. In February 2008, after an appeal spearheaded by the BC Civil Liberties Association, a review board reversed the parole board's decision and granted Robert Latimer day parole in 2010; he can now apply to travel outside Canada.

Notes 1. 2. 3.

Ontario Foundation for Cerebral Palsy "What Is Cerebral Palsy?" (2007), www.ofcp.on.ca/aboutcp.html##A. Robert Latimer, "Tracy's Illness" (2007), www.robertlatimer.net/story/tracysillness.htm. Supreme Court of Canada, 2001, R. v Latimer, 12001] 1 SCR 3, 2001 5CC 1, http://scc.lexum.umontrealca/ en/2001/2001scc1/2001sccl.html.

4.

CBC News, "Parole Board Denies Latimer's Bid for Partial

Freedom" (2007), www.cbc.ca/canada/story/2007/12/05/ Latimer-parole-mtg.html.

343

II All

1111:111

6 End-of-Life Decision-Making

Case 4

Mr McCullough: Recommending Voluntary Passive Euthanasia In September 2003, 85-year-old Burnaby, BC, resident Carl McCullough went to his local newspaper recounting his concern that health cave providers were encouraging him to take his own life. Three times a week, for seven years, Mr McCullough had been having kidney dialysis treatments at a local hospital, treatments that hadn't always been easy for him to endure. On two separate occasions during dialysis, as Mr McCullough complained about how hard he was finding the treatments that day, two separate HCPs offered him substantially the same unsolicited advice: "If I was getting fed up with things, I could take this way out: just don't come to dialysis."' When Mr McCullough's son, Davin, discussed his father's concerns with hospital staff and administrators, he was told that it was important to

present patients with options, particularly those patients who have reached the point of despondency, to allow patients to make well-informed choices about their future. Both McCulloughs saw it as an issue of planting seeds of fear in elderly patients: "How many aged patients over the past year have been slipped this information?"' In an environment where the need for medical resources exceeds supply, one way to reduce the drain on resources is to exclude some categories of patients, such as the elderly or addicts, from receiving medical treatment—even if doing so hastens death. HcPs may see the disclosure of the option to withhold or withdraw treatment in order to hasten death and end suffering as providing information necessary to a patient's informed consent. On the other hand, however, it opens the possibility that patients may perceive, rightly or wrongly, that they are being pressured to "let nature take its course" so they'll die sooner and cease to be a burden. This appeared to be the concern for the McCulloughs.

Note 1.

Julie MacLellan, "A Question o f Life o r Death?" i n Burnaby: NOW 27 September 2003, 1, •4.

Case 5

Elizabeth and Eric MacDonald: Assisted Suicide On 8 June 2007, 38-year-old Elizabeth MacDonald of Windsor, Nova Scotia, died at .3 clinic in Zurich, Switzerland, after drinking a fatal dose of barbiturates she received with the assistance of the Swiss suicide group Dignitas. Staff at the clinic provided Ms MacDonald with the glass of barbiturates and told her that if she drank it, she would die. She said, "I understand that" and drank it without hesitation. She died in her husband's arms shortly afterward. Ms MacDonald had been suffering from multiple sclerosis, an autoimmune disease affecting the central nervous system, which can result in a multitude of mobility problems, spasticity, and tremors, as well as cognitive and emotional issues, including depression. Ms MacDonald had a severe form of the disease and had been wheelchair-bound, unable to move. Recently, her throat had begun to paralyze.' Her condition was so distressing to her that she had already attempted to commit suicide a year earlier. Her husband believed

that if she'd been able to commit suicide with assistance in Canada, she would have lived longer: "she would have held on a bit longer because she was terrified of being trapped in her own body and being unable to travel."2 Ms MacDonald's feelings were made clear in a letter she left in which she expressed her frustration with Canadian law: "It is intolerable and unacceptable that I cannot be assisted to die here in Canada, in my own home, in my own bed, surrounded by those I love."2 When Ms MacDonald's obituary later publicly thanked the staff at the Dignitas clinic, Alex Schadenberg, a member of the Canadian-based Euthanasia Prevention Coalition (EPc) contacted the police, requesting an investigation into whether Ms MacDonald had been counselled to commit suicide while on Canadian soil. At that time, Canadian law punished assisting suicide with up to 14 years in prison, whereas Swiss law allowed assisted suicide if it was done for unselfish reasons. (Assisted suicide is now called PAD in Canada and became legal in 2015.) Attention turned to her husband, Eric MacDonald, a retired Anglican minister who had accompanied her to Switzerland and was present at her death. However, Canadian prosecutors

Cases

concluded that accompanying someone to the place of their suicide and being with them during the act is not the same thing as "aiding and abetting" a suicide; with no evidence that Eric counselled Elizabeth to commit suicide, no charges were laid against him. Schadenberg responded to the case by reminding us that Canadian laws are designed to protect the vulnerable and should remain unchanged and enforced

when evidence of wrongdoing occurs. He criticized Mr MacDonald for accompanying his wife to her suicide: "The woman was not terminally ill. She had MS. She had a disability. He [MacDonald] might consider it to be a loving act, but really what she didn't need was death. I consider the act of her husband to be an abandonment of her needs in this situation. This is not a supportive act."2

Notes 1.

News, "RCMP t o Question Man Who Took Ill Wife to Commit Suicide Overseas," 27 June 2007, www.cbc.ca/ canada/story/2007/06/27/suicide-assisted.html.

CBC

2.

John Jalsevac, CBC News, "No Charges Laid in Nova Scotia Assisted Suicide Case," CBC News 7 July 2007, wwwlifesite.net/ldn/2007/jul/07070511.htm.

6.6 Study Questions 1. What argument can be given to show that physicians who WLST do not kill their patients? What reply can be made to this? 2. How does the DDE allow physicians to perform terminal sedation but still oppose PAD? Does the DDE succeed in reconciling the prohibition of PAD and permissibility of terminal sedation? Is there a good reason to hold that physicians must never intentionally kill their patients? 3. What is the main philosophical argument for legalizing PAD? What is the Constitutional argument? What are the main arguments against legalizing PAD? Do you think PAD should be legalized? 4. What reasons can be given to extend Bill C-14 to allow for advance directives and be available to mature minors and (through substitute decision-makers) incapable children and adults? What reasons can be given for requiring that death be foreseen? Do you think those reasons are sufficient? Why or why not? 5. What are the two approaches to special safeguards articulated by the SCC? What are the arguments for and against each? Which do you prefer, and why? 6. What obligations, if any, do health care providers who have conscientious objection to participating in PAD have to help patients who want PAD to access it? Should institutions that are publicly funded and have conscientious objection to PAD be required to allow it on their premises, or should they be allowed to transfer patients to other hospitals to have it carried out? 7. What is cardiocirculatory death, whole-brain death, and higher-brain death? Will any of these serve as a uniform definition of death for our society? Should Canada have a uniform definition of death, and if so, what? Or should the definition of death be individualized, and if so, how? What do you think of DCD?

6.7 Suggested Further Reading WLST and Refusal ofTreatment Beauchamp, Tom L., and Robert Veatch, eds. 1996. Ethical Issues in Death and Dying. Upper Saddle River, NJ: Prentice-Hall. Bernat, James L., Bernard Gert, and R. Peter Mogielnicki. 1993. "Patient Refusal of Hydration and Nutrition: An Alternative to Physician-Assisted Suicide or Voluntary Active Euthanasia." Archives of Internal Medicine 153 (27 December): 2723-8.

11 .111

• El

111111

6 End-of-Life Decision-Making

DeGra.zia, David. 1992. "On the Right of Nondangerous' Incompetent Patients to Leave Psychiatric Units against Medical Advice." Contemporary Philosophy 14 (September): 1-5. Gostin, Lawrence 0.1991. "Life and Death Choices after Cruzan." Law, Medicine & Health Care 19: 9-12. Ross, Lainie Friedman. 1998. Children, Families, and Health Care Decision Making. New York: Oxford University Press. Sullivan, Mark D., and Stuart J. Youngner. 1994. "Depression, Competence, and the Right to Refuse Livesaving Medical Treatment." American Journal of Psychiatry 151 (July): 971-8. Weir, Robert F, and Charles Peters. 1997. "Affirming the Decisions Adolescents Make about Life and Death." Hastings Center Report 27 (November/December): 29-40. Terminal Sedation and Palliative Care Bennett, J.F. 1981. Morality and Consequences." In S.M. McMurrin, ed., The Tanner Lectures on Human Values 1981 II. Salt Lake City: University of Utah Press. Cantor, Norman L. 2001. "Glucksberg, the Putative Right to Adequate Pain Relief, and Death with Dignity." Journal of Health Law 34: 301-33. Kamm, Frances M. 1999. "Physician-Assisted Suicide, the Doctrine of Double Effect, and the Ground of Value." Ethics 109 (April): 586-605. Quill, Timothy E., et al. 2000. "Palliative Treatments of Last Resort: Choosing the Least Harmful Alternative." Annals of Internal Medicine 132: 488-93. Quill, Timothy E., and Ira R. Byock 2000. "Responding to Intractable Terminal Suffering: The Role of Terminal Sedation and Voluntary Refusal of Food and Fluids: Position Paper." Annals of Internal Medicine 132: 408-14. Truog, Robert, et al. "Barbiturates in the Care of the Terminally Ill." New England Journal of Medicine 327: 1678-82.

Physician-Assisted Death Battin, Margaret P, Rosamond Rhodes, and Anita Silvers, eds. 1998. Physician Assisted Suicide: Expanding the Debate. New York: Routledge. Beauchamp, Tom L., ed. 1996. Intending Death: The Ethics of Assisted Suicide and Euthanasia. Upper Saddle River, NJ: Prentice-Hall. Bickenbach, J. 1998. "Disability and Life-Ending Decisions." In M. Battin, R. Rhodes, and A. Silvers, eds., Physician-Assisted Suicide: Expanding the Debate, 123-32. London: Routledge. Dworkin, Gerald, Raymond G. Frey, and Sissela Bok. 1998. Euthanasia and Physician-Assisted Suicide: For and Against. New York: Cambridge University Press. Kamisar, Yale. 1958. "Some Nonreligious Views against Proposed `Mercy-Killing' Legislation." Minnesota Law Review 42 (6):324-32. Koop, C. Everett, and Edward R. Grant. 1986. "The 'Small Beginnings' of Euthanasia." Journal of Law, Ethics & Public Policy 2: 607-32. Quill, Timothy E. 1998. "The Debate over Physician-Assisted Suicide: Empirical Data and Convergent Views." Annals of Internal Medicine 128 (April): 488-93. Rachels, James. The End of Life: Euthanasia and Morality. 1986. Oxford University Press. Schaffner, Kenneth E 1988. "Recognizing the Tragic Choice: Food, Water, and the Right to Assisted Suicide." Critical Care Medicine 16: 1063-8. Sneddo a, Andrew 2006 "Equality, Justice, and Paternalism: Recentering the Debate about Physician-Assisted Suicide." Journal of Applied Philosophy 23 (4). Stingl, Michael, ed. 2010. The Price of Compassion: Assisted Suicide and Euthanasia. Broadview Press.

Suggested Further Reading

Sumner, L.W. 2011. Assisted Death: A Study in Ethics and Law. Oxford University Press. Thomson, Judith Jarvis. 1999. "Physician-Assisted Suicide: Two Moral Arguments." Ethics 109: 497-518. Velleman, J. David. 1992. "Against the Right to Die." Journal of Medicine and Philosophy 17: 665-81. Williams, Glanville. 1957. The Sanctity of Life and the Criminal Law, ch. 8. New York: Alfred A. Knopf. . 1958. -Mercy-Killing Legislation—A Rejoinder." Minnesota Law Review 43 (1): 1-12. Physician-Assisted Death in Canada: Government Documents (1) Before Carter

Law Reform Commission of Canada. 1982. Euthanasia, Aiding Suicide and Cessation of Treatment. Working Paper 28. . 1983. Euthanasia, Aiding Suicide and Cessation of Treatment. Final Report. National Assembly of Quebec. 2003. Bill 52: Act Respecting End-of-Life Care. Royal Society of Canada. 2011. End-of-Life Decision Making. Special Senate Committee on Euthanasia and Assisted Suicide. 1995. Of Life and Death— Final Report. (2) After Carter

Canadian Medical Association. July 2014. A Canadian Approach to Assisted Dying: CMA MemberDialogue:SummaryReport.https://www..cma.ca/Assets/assets-library/document/ en/advocacy/Canadian-Approach-Assisted-Dying-e.pdf. .End-of-LifeCare—ANationalDialogue: CMAMemberConsultationReport. July2014. haps://www..cma.ca/Assets/assets-library/document/en/advocacy/end-of-life-carereport-e.pdf. College of Family Physicians of Canada. September 2015. A Guide for Reflection on Ethical Issues Concerning Assisted Suicide and Voluntary Euthanasia. cfpc Task Force on End-of-Life Care. http://www.cfpc.ca/uploadedFiles/Health Policy/ PDFs/Guide for %20Euthanasia EN Final.pdf. House of Commons and Senate, Special Joint Committee on Physician-Assisted Dying. February 2016. Medical Assistance in Dying: A Patient-Centered Approach. httpi/ www.parl. gc.ca/HousePublications/Publication. aspx?Docid=8120006&Language=E& Mode=l&Par1=42&Ses=1. Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying. 30 November 2015. Final Report. http://www.health.gov.on.ca/en/news /bulletin/2015/docs/eagreport 20151214 en.pdf. Defining Death Bernat, James L 14 August 2008. "The Boundaries of Organ Donation after Circulatory Death." New England Journal of Medicine 359 (7): 669-75. Dworkin, Roger. 1973. "Death in Context." Indiana Law Journal 48: 623-46. Emanuel, Linda. 1995. "Reexamining Death: The Asymptotic Model and a Bounded Zone Definition." Hastings Center Report 25 (4): 27-35. Green, Michael, and Daniel Wikler. 1980. "Brain Death and Personal Identity." Philosophy and Public Affairs 9: 105-33. Institute of Medicine. 2000. Non-Heart-Beating Organ Transplantation. Washington, DC: National Academy Press: 2000.

6 End-of-Life Decision-Making

Law Reform Commission of Canada. 1979. Criteria for the Determination of Death. Working Paper 239. President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioural Research. 1981. Defining Death: Medical, Legal and Ethical Issues in the Definition of Death. Washington, DC: U.S. Government Printing Office. Shermie, S.D., Al Baker, G. Knoll, W Wall, G. Rocker, D. Howes, et al. 2006. National Recommendations for Donation after Cardiocirculatory Death in Canada. Canadian Medical Association Journal 175 (8 Suppl): S1—S24. Truog, Robert. 1997. "Is It Time to Abandon Brain Death?" Hastings Center Report 27 (1): 29-37. Van Norman, Gail A. 2003. "Another Matter of Life and Death: What Every Anesthesiologist Should Know about the Ethical, Legal, and Policy Implications of the Non-HeartBeating Cadaver Organ Donor." Anesthesiology 98: 763-73.

Delivery o Heakh Care and Resource AHocatIon 7.1 Introduction The topic of resource allocation is commonly divided into three areas. These are macroallocation, which concerns how much of society's resources should go to health care; mesoallocation, which concerns what areas of health care should receive what proportion of those resources; and microallocation, which addresses the question of which patients should receive health care when not all can. Together, these areas represent aspects of the problem of distributive justice or how to fairly allocate scarce resources in a society. We first describe macro, meso, and microallocation in more detail to give a clearer overview of the topic of distributive justice and then discuss related issues of fair allocation of health care resources. Macroallocation questions occur at a societal level and involve two basic questions. The first is about the relative value of health care. Health care is a scarce good that competes for resources with a number of other scarce goods such as education, social services, roads, and public safety. How much of the available resources should go to health care? We cannot answer this question without understanding the health care needs of a community and the costs of providing it. It follows that macroallocation decisions must be informed by answers to general questions about how much money should be available to fund specific areas of health care, including acute care, preventative care, public health, long-term care, rehabilitation, community care, payment of health care workers and administrators, and so on. Macroallocation decisions are, therefore, profoundly difficult to make with precision. The second macroallocation question is about the delivery of health care. Should there be a one-tier health care system with a public or government payer that tries to serve all necessary health care needs and does not allow individuals to purchase more? Or should there be a public tier available to everyone and an option to buy more privately if one can afford it? Or should the government stay out of the health care business and let health care be distributed by the free market? Mesoallocation questions occur at an institutional level. Once a health care region or public or private institution has its budget, how should it spend it? For instance, how much should a hospital spend on emergency care, the icu, surgery? And within those divisions further questions arise, such as what support is to be given to urology, cardiology, neurology etc. And since there is never enough to give every patient everything they want or need when they want or need it, rationing at a meso level is needed. Sometimes this involves eliminating classes of service or treatments—for example, not offering all

macroallocation the fair allocation of a society's resources for health care under conditions of scarcity. mesoallocation the fair allocation of scarce health care resources within a specific health care region or institution.

microallocation the fair allocation of scarce resources to specific individuals.

distributive justice the fair allocation of scarce resources.

rationing addressing scarcity of health care resources by eliminating services or treatments offered, or excluding classes of individuals from health services or treatments, or prioritizing access.

II

'MI

7 Delivery of Health Care and Resource Allocation

triage the assignment of degrees of urgency to wounds, injuries, or illnesses to decide the order of treatment of a large number of patients under conditions of scarcity of medical resources.

emergency services or leaving certain types of surgery to other institutions. Sometimes the remedy involves prioritizing the availability of care, as when hospitals impose guidelines or wait-lists to provide care in proportion to need. Such decisions merge substantially into the sphere of microallocation. Thus, one of the great health care allocation issues of our time— rationing (though the word is studiously avoided)—falls under the topic of mesoallocation M:.croallocation concerns making choices between particular patients (versus classes of patients as in mesoallocation). The topic arises because there are never enough resources to meet all the health care needs. Even if society forgot about all other goods such as roads and education and devoted all of its resources to health care, there would still be a gap between supply and demand. And since more and more can be continually done to provide for better health care, this gap seems permanent. Thus, hard choices between patients are inevitable when health care providers have to practise triage and decide which patients will receive organ transplants or ventilators in an influenza epidemic, as well as make many more, less dramatic decisions. The central ethical question provoked here is how to make these decisions. Should they be made on purely medical criteria, or should they also take into account, where possible, social criteria (such as being a teacher versus a street person or single versus married with children), age, and responsibility for the medical condition?

Macroallocation The two leading questions under this topic are (to repeat) how much of society's resources should go to health care and how those resources should be delivered. How we answer these questions depends entirely on whether we want to live in an egalitarian, libertarian, or liberal society. The problems of resource allocation thus quickly lead us to the basic question of social philosophy—how should we live together?—and it is not easy to choose .netween the alternatives. Indeed, it is not clear that there is a single right answer to the question. However, once we answer that question, we will have answers to the main questions of macroallocation. As we will see, an egalitarian society will spend generously on health care to cover all its citizens with a high level of care and is usually assumed to favour a one-tier public delivery system. A libertarian society will leave it to a free market to decide how much (or little) should be spent on health care. We now turn to examine these alternatives more closely. Egalitarian egalitarian theory of justice every person is equally valuable, and society has an obligation to provide them with the basic resources for a good life.

Canada has elements of an egalitarian social philosophy. The moral basis of the egalitarian theory of justice is the view that all people are equally valuable. Thus, since health is necessary for a good life, and health care for health, all people have an equal right to health care. This means that society has an obligation to provide it or see that it is provided. This does not mean that society has an obligation to provide its citizens with equal health, for delivering equal health is not within the domain of social control. Nor does it mean that there is an obligation on the part of society to provide citizens with all the health care they need to achieve a basic level of health, for conditions of scarcity may make this impossible. Rather, it means that a modern, developed society has an obligation to secure enough resources to provide its citizens with a decent minimum of health care. In "The Right to a Decent Minimum of Health Care" (see page 367), Allen E. Buchanan explains further what this obligation does and does not involve. The basic elements of the egalitarian approach to delivery of health care are articulated in the main articles of the Canada Health Act, which can be summarized as follows: 1.

Public administration: All medically necessary services offered by doctors and hospitals are insured by a single public payer.

Macroallocation

2. Universality: Everyone who is covered by the health care system is covered in the same manner and under the same terms. 3. Accessibility: There are no financial barriers (e.g., user fees, extra billing) to accessing health care services. Articles 2 and 3 amount to the egalitarian commitment that essential health care services are available on the basis of need and need alone; article 1 is advanced as the most efficient means to give that commitment practical effect. Libertarian

The mirror image of the egalitarian system is the libertarian. While the egalitarian seeks to maximize equality, the libertarian seeks to maximize freedom. This is a view made famous (though not invented) by Robert Nozick in Anarchy, State, and Utopia' and is based on two natural rights—i.e., rights that all human beings have as human beings and which are impervious to political interference. The first is a right to anything one justly acquires. One can justly acquire something by either working for it or inheriting it from someone who has justly acquired it. The second is a right to enter into just agreements. If I have something you want (e.g., health care) and you have something I want, (e.g., money), we can agree to exchange those things and thus buy and sell health care. Libertarians hold that any system of health care delivery (like the Canadian system) that does not permit these rights is unjust. They also hold that it is unjust to help those who cannot afford health care through public taxation (as, again, the Canadian system does), for the public does not owe anyone in need anything and thus cannot be forced to help. Application of the libertarian philosophy would almost certainly result in unequal access to health care, but it does not follow that it is therefore unjust. For (so the argument goes) any pattern of distribution that is reached by just steps must be just. Indeed, in the libertarian view, any attempt to interfere with a pattern of distribution based on libertarian principles will necessarily be unjust, for that will violate either the right to keep what one has justly acquired or the right to enter into just agreements. Libertarians will thus leave it to the free market to allocate health care. Pure libertarians will leave it entirely to the market and rely on the compassion of individuals and groups (versus the government) to care for those who cannot care for themselves, but some libertarians accept a basic governmentsupplied safety net.

libertarian theory of justice every person has a right to private property and to enter into any agreements they want to. Government can interfere only to prevent force or fraud.

natural rights rights that all human beings have just h virtue of being human and which are not subject to political control.

Liberal

Both the egalitarian and libertarian view can claim the backing of a theory of distributive justice (albeit strikingly different ones), and both have a point. A liberal view seeks to combine the two views by advocating a delivery system that provides for a public tier of health care as good as communal resources can afford and a private tier that will allow those who have resources to buy enhanced services and avoid a wait-time. This is the system in place in many European and developed countries, and some aspects of it are reflected in the US health care system as amended under President Barack Obama. The liberal system, it is argued, is the fair compromise between irreconcilable views that are both attractive. It is also sometimes argued that everyone will benefit from two-tier health care. Those who can afford enhanced care or quick access to health care will benefit. Those who cannot afford to buy health care in the private tier will also benefit because the private tier will shorten wait lists in the public tier. Thus, everyone will benefit, albeit not benefit equally. Egalitarians often reply by saying that the private tier will cause the public tier to deteriorate, because allowing some to buy care in a private tier reduces pressure to maintain and improve quality in the public tier. It is also often argued that even if everyone got better care in a two-tier

liberal theory of justice a social view that combines egalitarian and libertarian elements. In the context of resource allocation, it means two-tier health care.

MI RI

II

11

II

1111.1J

7 Delivery of Health Care and Resource Allocation

system, two-tier health care should still be resisted because having one health care for the rich and another for the poor will undermine social solidarity.

The Romanow Report The reading by Roy Romanow, "Sustaining Medicare: The Commission on the Future of Health Care in Canada" (see page 357) is part of a royal commission report, Building on Values, published in 2002.2 The commission's mandate was to review the Canadian public health care system and make recommendations to improve its quality and sustainability without compromising the principle that an individual's financial resources should not determine access to services. Romanow does not argue for the superiority of the Canadian system over its rivals. His main aim is to show that the system can be sustained, where "sustainability means ensuring that sufficient resources are available over the long term to provide timely access to quality services and address Canadians' evolving health needs" (see page 357). Romanow thinks that the Canadian health care system should continue to be financed by progressive taxation and higher taxes if necessary. He rejects a number of suggestions about alternative ways of paying for health care services (e.g., user fees, extra billing, medical savings accounts) on the ground that they will deter individuals from seeking health care and make access dependent on ability to pay rather than on need alone. The Romanow reading is included in this chapter because it expresses thoughtful support for the Canadian health care system and offers a plan that remains the current blueprint for maintaining and sustaining it. It remains an open question whether the Canadian health care system is sustainable and whether a more liberal two-tier system, such as those in other modern industrial democracies outside of North America, do a better job of providing decent minimum care for all their residents. Many of these communities with two-tier systems consistently rank ahead of Canada in provision of health care.3

Mesoallocation How Much for Health Care? The Romanow report and egalitarian and liberal approaches to health care all broadly agree on the appropriateness of providing at least a decent minimum of health care for all members of a community. But what is lacking from these discussions so far is an actual model for determining how much of society's resources should be spent on health care, how much should go to specific areas of health care, and how to make the hard rationing decisions that run throughout health care. In short, we need a decision procedure for answering these macro- and mesoallocation questions in practical terms and putting the idea of a decent minimum of health care into effect. Ronald Dworkin's "Justice and the High Cost of Health" (see page 372) attempts to address this issue. Although Dworkin is addressing a US audience, his arguments are general and relevant to all modern developed countries. Dworkin begins by noting that no society or individual could ever afford to fund "the rescue principle": the idea that society is obligated to allocate whatever resources are needed to save a life whatever the cost. If that is correct, a decision process is needed to figure out how to weigh competing claims on scarce societal resources and to make difficult decisions about not only which lives to save when all cannot be but also how to allocate resources to health care generally. The procedure Dworkin proposes is the "prudent insurance" test. This test attempts to combine the most compelling elements of the egalitarian, liberal, and libertarian approaches to address the problem of macroallocation. The test asks us to consider what prudent persons (called "prudent insurers") would decide to spend on their own health care if, as individuals, they were buying insurance under fair free-market conditions, did not know what special medical needs they were

Mesoallocation

likely to have (nor did insurers know this), knew all that is currently known about possible medical conditions and the efficacy of treatments for them, and had significant but limited financial resources to spend. Dworkin contends that how much money prudent insurers would spend on health care insurance provides the model for how many resources society should allocate for health care. He also contends that prudent insurers would not choose to buy "rescue principle" insurance because its expense would be prohibitive, requiring that other important personal needs and wants could not be served (for example, education, food, shelter, and entertainment). Dworkin proposes that how much money should be allocated toward a basic minimum of health care and how it should be spent could be determined by establishing an agency (or perhaps a sort of citizens' assembly) where average citizens and others with relevant expertise could attempt to apply the prudent insurance test. Dworkin's prudent insurance test is thus an all-purpose test of how society should allocate resources. How much of society's total resources should go to health care, how those resources should be allocated to specific areas of health care, and how rationing decisions should be made are all determined by estimating what prudent insurers would spend to insure their health care.

How Do We Measure and Prioritize Health Benefits? One problem that is not discussed in Dworkin's hypothetical insurance scheme is how to measure and prioritize health outcomes so that whatever money is allocated for health care is used efficiently to preserve and promote health. Ethicists and health care economists have come up with a variety of models that would be of use to the members of Dworkin's citizens' assembly. Bjarne Robberstad's "QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference Do They Make for Health Care Priority Setting?" (see page 382) sets out the main models and critically discusses them. All these models, however, have their limitations. Quality and Adjusted Life Years (QALYs), Disability Adjusted Life Years (DALYs), and Life Years Saved (LYs) are all types of cost-benefit analyses. The classic approach to cost-benefit analysis states that resources should be used in ways that produce the greatest economic benefit. This is widely rejected for health care as not being sufficiently egalitarian. It would mean health resources would go first to individuals with the greatest capacity to produce wealth in a society, typically the richest and most powerful persons in society. But there are also specific complaints of unfairness to be made to each of the other metrics. If we allocated resources to produce the most LYs, this would tend to overlook the claims of persons with disabilities and those with chronic illnesses. Using QALYs and DALYs would address these deficiencies to some degree because they attempt to measure quality of life and the burdens of disability but at the expense of LYs and other considerations, including what to do when maximizing QALYs and minimizing DALYs conflict. The fact is that there is no acknowledged metric for fairly assessing the benefits of different health outcomes. This is perhaps an argument for having a transparent democratic process for making macro and mesoallocation decisions like the one suggested by Dworkin, since a fair decision-making process establishes some legitimacy for decisions that everyone will have to acknowledge at the outset are bound to be imperfect.

The Rescue Principle Rejected Again Tony Hope's "Why Undervaluing 'Statistical' People Costs Lives" (see page 377) is a deeper investigation into the problems with using the rescue principle (he calls it "the rule of rescue") as a basis for allocating scarce resources. Hope's article describes how the rescue principle is reflected in many health care decisions for allocating scarce resources, especially in emergency care, and how it can undermine fair and just treatment of individuals. In particular, Hope describes how large amounts of scarce resources tend to be allocated to actual individuals who are in dire straits at the expense of using those resources to save the lives of other, and often a greater number of, "statistical" individuals (that is, individuals

7 Delivery of Health Care and Resource Allocation

who are not currently being treated but who may be affected by decisions to allocate large amounts of scarce resources to rescue others). Ethicists are widely agreed that arguments like Hope's against the rescue principle are compelling, However, accepting the weight of those arguments has radical implications for the way health care resources are allocated, particularly at the level of mesoallocation. A lesser role for the rescue principle would mean fewer resources for emergency and acute care and more for preventative care, since this would mean more life years saved and higher quality of life generally. Hope notes that despite the apparently compelling nature of these argummts from distributive justice, there is no country in the world that has attempted to put them into effect. See the Hope article for a discussion of why this has been the case. It is interesting to consider whether, contrary to appearance, the rescue principle represents a failure of empathy or compassion, in particular for those "statistical" persons whose needs are jus: as real, though not immediately seen or felt, as those needing immediate rescue.

Microallocation Even if questions about macroallocation and mesoallocation could be fully and fairly addressed, there would still not be enough resources to meet everyone's medical needs. Such predictable shortages mean that decisions would still have to be made at an individual level about who will or will not receive a scarce health care resource. Microallocation is concerned with making these decisions fairly. More bluntly, microallocation is concerned with choosing between particular patients when there are not sufficient resources for each one's medical needs to be met. The central question is who should get treatment when not everyone can. These decisions can involve life and death questions, such as who should receive vital organ transplants or ventilation in an influenza epidemic, as well as less dramatic but very important interventions, such as access to speech language pathology, psychiatry, and rehabilitation services. Sometimes the exclusion is permanent, sometimes it is a matter of prioritizing patients into a waiting list. We can also note that microallocation questions typically arise only where health care is a public good—i.e., a good to which everyone in society has an equal claim. Where health is not such a good, as in private systems, it can (except perhaps in special conditions of scarcity) be distributed in any way the owner of health care wants, such as for money or to friends and relatives.

Complex Criteria There are three main approaches to the question o f microallocation. The first is a "complex criteria .system." All complex criteria systems first exclude patients who will not benefit from the treatment at all. Then they apply a number of criteria to select the candidate or candidates who will receive the treatment. Complex criteria systems vary in the criteria they use, but a standard list will include (1) how much the patient will benefit in terms of quantity and quality of life; (2) whether the patient has dependents; (3) the patient's past contributions to society, (4) the patient's expected future contributions to society, (5) the patienth age; and (6) whether the patient's behaviour is responsible for his or her illness. If there are still more patients than can be provided with resources, random selection ("first come first served" or a lottery) is used to make the final selection. This is the kind of system proposed (with some wrinkles of his own and a scoring system for comparing candidates) by Nicholas P Rescher in "The Allocation of Exotic Life-Saving Therapy" (see page 391).

Random Selection The second method of microallocation is by a "random selection system." Random selection systems also begin by excluding patients who are not expected to benefit from

Microallocation

the treatment but then studiously turn a blind eye to other factors used in complex criteria systems. The main virtue of such a system is that it preserves the inherent dignity and worth of all human beings by giving all an equal opportunity to live. Need and need alone determines access to health care. Random selection also has the advantage of simplifying decision-making by eliminating the need to assemble a committee to evaluate patients on the basis of difficult-to-apply and potentially discriminatory criteria.

Modified Random Selection There is a further question of whether random selection is otherwise satisfactory as it stands. It is sometimes argued that fairness requires that choosing between patients take into account the age of the patient (age-based rationing) and whether his or her behaviour is responsible for the illness (responsibility for condition). Thus, we get a third method of microallocation that lands between complex criteria and random selection methods. Age-Based Rationing

The problem that provokes the proposal to ration care on the basis of age is that as a population ages, and as medicine can increasingly help the elderly live longer and better lives, it is impossible to give both the elderly and non-elderly the best possible health care. Thus, we must either ration care for the elderly or disadvantage the young. In "Just Caring: In Defense of Limited Age-Based Healthcare Rationing" (see page 399), Leonard M. Fleck canvasses the main arguments for and against age-based rationing, and we will look at age-based rationing and holding individuals responsible for their condition in turn. Many bioethicists hold that age-based rationing is straightforward discrimination, no different in principle from treating people differently on the basis of race or gender. If an elderly person and a non-elderly person each have essentially the same medical problem requiring the same life-preserving treatment, then (so the argument goes) they must have an equal right to receive that treatment. Elderly Canadians have contributed to universal health care for most or all of their working lives, and so they are now entitled to the health care they need. Many other bioethicists, however, argue that fairness requires age-based rationing, and Fleck identifies two different ways in which this has been argued. Sometimes the judgment of fairness rests on the fact that the old have had a chance to experience life, while the young have not had the same chance. The argument is that this difference in opportunity should be equalized by giving the young priority in receiving life-preserving care. This argument is known in the literature as the "fair innings" argument. On this view, everyone is entitled to a fair and equal shot at life, and a just health care system will try to make it so. A second argument for age-based rationing begins with a thought-experiment. The central character in this experiment is the "prudent insurer" described earlier. (This, you will recall is a person who wants to buy insurance, is knowledgeable about what illnesses may befall human beings and what can medically be done about them, but has limited funds and does not know his state of health.) We are now asked to suppose that the prudent insurer is given a choice between a policy that will provide life-preserving care in his middle years and one that will provide for that care in his old age. There can be (as Dworkin claims) no doubt that he would choose the one that will protect him in his middle years. But if so, the prudent insurer must be in favour of age-based rationing, because to choose to have life-preserving treatment in his middle years rather than old age is just another way of choosing to have age-based rationing. Since everyone planning what health care should be delivered to him or her over a lifetime can also be expected to prefer life-preserving treatment in the middle years rather than old age, everyone could

age-based rationing rationing based on age and age alone. As used in the literature, it means that if an elderly person and a non-elderly person each has essentially the same medical problem, needs life-preserving treatment, and has the same prognosis, the treatment can be denied the elderly person on the basis of age.

II

URI

II HUI

7 Delivery of Health Care and Resource Allocation

be expected to want age-based rationing. But if everyone wants age-based rationing, this rationing must be fair. Responsibility for Condition luck egalitarianism individuals are responsible for consequences if and only if they are the result of their voluntary choices.

determinism every event is brought about and necessitated by prior causes that occur according to the laws of nature.

The second deviation from random selection—responsibility for one's condition—is based on luck egalitarianism. Applied to health care, this is the view that individuals are not responsible for health problems that are a result of brute bad luck, and so they should be treated by the common pool of resources that make up the health care system. However, if they brought the condition on themselves by behaviour they could reasonably have been expected to know would put them at risk (e.g., smoking, drinking, overeating, or extreme sports), it is only fair, in conditions of scarcity, that they should either not be treated or have a lower priority. Alvin H Moss and Mark Siegler in "Should Alcoholics Compete Equally for Liver Transplantation?", Morten Ebbe Juul Nielsen and Martin Marchman Andersen in "Should We Hold the Obese Responsible? Some Key Issues," and Katherine Browne in "Voluntary Sterilisation and Access to IVF in Quebec" consider how luck egalitarianism fares in limiting treatment in these cases. (See the Suggested Readings for the references.) The main problem with appeals to luck egalitarianism is how to be able to say when choices are voluntary. This problem surfaces in two ways. First, there is a question of whether anyone can ever be held responsible for his or her actions. This rests on the possibility that determinism is true. Determinism is the view that every event has a cause. On this view, all our behaviour can be explained in terms of causes: our behavior is caused by our choices, our choices by our decisions, our decisions by our desires, our desires by our charac:er, and our character by our genetics and environment. Given that our genetics and environment lie beyond our control, the unbroken causal chain stretching from them to our behaviour entails that, for anything we do, we could not have done otherwise. But if so, individuals cannot be held responsible for their actions. One need not, however, resort to metaphysical speculations to be reluctant to hold individuals such as alcoholics responsible for their condition. There are a number of more mundane reasons that we can appeal to. Social determinants of behaviour such as parents, upbringing, poverty, peer pressure, educational and vocational opportunities, coupled perhaps with a genetic predisposition to alcoholism cast doubt on how big a role voluntary choice plays in becoming an alcoholic. It is thus problematic to invoke luck egalitarianism to give alcoholics second-rate treatment because they are responsible for their condition. The same can be said for other patients such as smokers, substance abusers, the obese, and those who have brought their condition on themselves by their choices All these types of patients may require special handling, but it is hard to see how they can, without considerable further argument, be punished or otherwise disadvantaged because they are responsible for their behaviour.

Increasing Resources One way to address problems of scarce resources in health care is to attempt to increase those resources. This can be done in a variety of ways that include increasing taxes or growing an economy so that more money is available to fund health care. Sometimes, however, non-monetary resources are what is scarce—for example, blood or organs. In Canada. it is illegal to sell or buy these resources, and so the health care system relies fully on the altruism of fellow citizens to provide them. The question arises, then, as to whether it should be possible to purchase them from individuals to increase the availability of these resources and to help sick individuals get access to potentially life-saving surgeries in more timely ways. In "Human Organs, Scarcities, and Sale: Morality Revisited" (see page 407), R.R. Kishore focuses mainly on whether individuals should be permitted to sell one of their kidneys to an individual who needs it. Kishore argues that concerns that sale of human

Increasing Resources

organs violates human dignity are mistaken and that concerns about abuse, exploitation, and difficulties obtaining genuinely informed consent can be addressed by regulation.

Notes 1. 2. 3.

Robert Nozick, Anarchy, State, and Utopia (Cambridge, MA: Harvard University Press, 1974). Roy Romanow, Building on Values: The Future of Health Care in Canada—Final Report (Ottawa: Royal Commission on the Future of Health Care in Canada, 2002). Conference Board of Canada, 2012, "International Ranking: Health" (2012) http://www.conferenceboard.ca/hcp/details/health.aspx.

7.2 Access to Health Care in Canada

Sustaining Medicare: The Commission on the Future of Health Care in Canada Roy Romanow

The heart of the Commission's mandate was to make recommendations "to ensure the long-term sustainability of a universally accessible, publicly funded health system." The rationale behind this mandate was quite simple. For a number of years now, Canadians have been told by some of their governments and a number of health policy experts that the system popularly known as medicare is no longer "sustainable." At the same time, the Commission's extensive consultations with Canadians and its comprehensive research program clearly indicate that Canadians want the system to be sustainable, not only for themselves but for future generations of Canadians. They want it to change, and to change in some very fundamental and important ways. But they also want it to endure and, indeed, to thrive. Is it possible to reconcile these two perspectives? The place to start is with a clear understanding of what makes a system sustainable and what needs to be done to ensure that Canada's health care system is sustainable in the future. What Is Sustainability?

In some ways, the word "sustainability" both illuminates and obscures the debate. It is a word that is immediately understandable and yet open to multiple

interpretations and misinterpretations. Moreover, much of the recent debate on health care has focused on one aspect only—namely costs. People conclude that the system is not sustainable because it costs too much money it takes too large a proportion of governments' budgets, or it is an impediment to lowering taxes. There are others who argue that the problem with the system is the way it is organized and the inefficiencies that result. Reorganize the system, they argue, and there is more than enough money to meet our needs. Still other voices have argued that the only problem with the system is the lack of money provided in recent years. Restore and increase the financial resources, they argue, and all will be well. In the Commission's view, this narrow focus on money is inadequate and does not help inform the debates or enable an overall assessment of whether or not Canada's health care system is sustainable. Instead, the Commission takes the view that: Sustainability means ensuring that sufficient resources are available over the long term to provide timely access to quality services that address Canadians' evolving health needs. For many years, health policy experts have focused on three essential dimensions that are each key to sustaining the health care system: •

services—A more comprehensive range of necessary health care services must be available to meet Canadians' health needs. The services must be of

Source: From Building on Values: The Future of Health Care in Canada (Health Canada, 2002). Reproduced with the permission of the Minister of Public Works and Government Services Canada, 2008.

MI NI

358

II

11

II

all

1111411

7 Delivery of Health Care and Resource Allocation

•

a high quality and accessible on a timely basis. This aspect of sustainability involves looking at the changing ways health care services are delivered, whether they are accessible for Canadians, and whether they are efficiently and effectively delivered. needs—The health care system must meet Canadians' needs and produce positive outcomes not only for individual Canadians but also for the population as a whole. This dimension examines how Canada's health care outcomes measure up to other countries, identifying disparities in the health of different Canadians and looking at trends in health. resources—This includes not only financial resources but also the required health care providers and the physical resources (facilities, equipment, technology, research, and data) that are needed to provide the range of services offered.

There is no "invisible hand" that silently and unobtrusively keeps these elements in balance. Decisions about providing adequate resources imply that there is political support by governments and by Canadians to continue supporting the system through public funds and public oversight. Maintaining the balance is, in fact, a deliberate act of will on the part of society and, thus, it is the overall governance of the system at all levels that ultimately decides how these elements are balanced. . . . Health and Health Care Services

Canada health care system provides a range of services, some of which are covered by the Canada Health Act and the well-known five principles, some that are covered by provinces and territories, and some that are provided through the private sector. In large part, provinces and territories are responsible for organizing and delivering health care services to people across the country Since medicare was first established, there have been considerable changes in both the scope of health care services provided in Canada and the different ways they are organized and delivered from relatively large regional health authorities to small clinics or doctors' offices. Services offered in our health care system can be differentiated by their complexity and intensity: the more or less specialized nature of interventions to maintain or restore health and the number of qualified health personnel needed to see the interventions through. At one end of the spectrum are a wide variety of services that are covered by the public health care

system: public health programs aimed at the prevention of illness such as the immunization of children; visits to family physicians, pediatricians, or gynecologists; diagnostic tests; and day surgery. Moving across the spectrum, we find the complex and intense care that requires the increasing use of advanced technology as well as highly trained specialists and large support teams. In addition, long-term or continuing care is typically provided in nursing homes or other specialized residential settings for people who require ongoing medical attention and support but who do not need to be treated in hospitals. Palliative care is provided to people who are dying and is available in hospitals, hospices, and, to a growing extent, at home. Home care is an increasingly important component of health care that can allow people to avoid hospitalization or recover at home following a shorter hospital stay. At any point along the spectrum, people can and frequently do receive prescription drugs. The key question in terms of sustainability is whether this vast continuum of services provided in Canada's health care system meets the needs of Canadians, is accessible, and can be adapted in the future to meet the changing needs of Canadians. . . . Private For-Profit Service Delivery: The Debate

One of the most contentious issues facing Canadians is the extent to which the private sector should be involved in delivering health care services. Currently, provincial and territorial governments provide coverage for a range of services, and those services can be delivered in any number of ways. Almost all Canadian hospitals are not-for-profit institutions and, in most provinces, are operated by regional health authorities. Most physician services are delivered by what are effectively owner-operated small businesses ranging from single-physician practices to multi-provider clinics that may include a range of health care providers. Large forprofit corporations deliver a narrower range of services including laboratory services and continuing and longterm care. In the face of continuing pressures on the health care system, some argue that more private for-profit service delivery ought to be introduced in order to bring more resources, choice, and competition into the Canadian health care system and to improve its efficiency and effectiveness. Others argue as strongly that the private sector should be completely excluded from health care delivery, suggesting that private for-profit delivery runs counter to Canadians' values, is inequitable, and is less cost-effective than public delivery in the long run.

Roma now: Sustaining Medicare:The Commission on the Future of Health Care in Canada

To try to make sense of this debate, it is important to distinguish between two types of services: direct health care services such as medical, diagnostic, and surgical care; and ancillary services such as food preparation, cleaning, and maintenance. An increasing proportion of ancillary services provided in Canada's not-for-profit hospitals are now contracted out to forprofit corporations. Canadians seem to find this role for private sector companies acceptable and some studies suggest that these enterprises achieve economies of scale (McFarlane and Prado 2002). Ancillary services are relatively easy to judge in terms of quality—the laundry is either clean or it is not, the cafeteria food is either good or it is not. Consequently, it is relatively easy to judge whether the company is providing the service as promised. Also, there is a greater likelihood that there are competitors in the same business to whom hospitals can turn for laundry or food services if their current contractor is unsatisfactory. In terms of direct health care services, the precise number of for-profit facilities delivering direct health care services is unknown. One estimate in 1998 (Deber et al.) suggested that there were 300 private for-profit clinics in Canada delivering many diagnostic and therapeutic services formerly provided in hospitals, including abortions, endoscopies, physiotherapy, new reproductive technologies, and laser eye surgeries. In addition, there are a growing number of small private for-profit hospitals or stand-alone clinics in some provinces providing more complex surgeries, some requiring overnight stays. These facilities vary considerably in terms of the number of services they offer and their ownership structure. Furthermore, some provinces have expressed an interest in contracting out an increasing number of surgical services to private for-profit hospitals and clinics in the hope of realizing efficiencies. Unlike ancillary services, direct health care services are very complex and it is difficult to assess their quality without considerable expertise. Indeed, the effects of poorly provided service may not be apparent until some time after the service has been delivered, as in the event of a post-operative complication. This is what most clearly distinguishes direct health care services from ancillary services—a poorly prepared cafeteria meal may be unpleasant, but poor quality surgery is another matter altogether. It is also unlikely that there would be a significant number of competitors able to offer health care services if a given for-profit provider is unsatisfactory. There simply is not a significant surplus of health care administrators or providers waiting in the wings to take over service delivery in a hospital. Thus, if services are of poor quality it is going to be

much harder to find a replacement once public facilities have stopped providing the services—the capacity that existed in the public system will have been lost. Some suggest that private for-profit delivery is more efficient than not-for-profit delivery (Gratzer 1999 and 2002). Given that most of the private facilities currently operating and being planned focus only on providing a limited range of services, there are some important concerns that must be addressed in terms of how these facilities interact with the more comprehensive public system. In effect, these facilities "cream-off' those services that can be easily and more inexpensively provided on a volume basis, such as cataract surgery or hernia repair. This leaves the public system to provide the more complicated and expensive services from which it is more difficult to control cost per case. But if something goes wrong with a patient after discharge from a private facility—as a result, for example, of a post-operative infection or medical error—then the patient will likely have to be returned to a public hospital for treatment insofar as private facilities generally do not have the capacity to treat individuals on an intensive care basis. Thus, the public system becomes liable for the care triggered by a poor-quality outcome within a private facility, yet under current arrangements there is no way for the public system to recover those costs from the private facility. In other words, the public system is required to provide a "back-up" to the private facilities to ensure quality care. Proponents of for-profit care may insist that the quality of care is not an issue, but there is evidence from the United States to suggest that the non-profit sector tends to have better-quality outcomes than the for-profit sector in such things as nursing home care (Harrington 2001; Marmor et al. 1987) and managed care organizations and hospitals (Kleinke 2001; Gray 1999). More recently, a comprehensive analysis of the various studies that compare not-for-profit and forprofit delivery of services concluded that for-profit hospitals had a significant increase in the risk of death and also tended to employ less highly skilled individuals than did non-profit facilities (Devereaux et al. 2002). For those reasons, the Commission believes a line should be drawn between ancillary and direct health care services, and that direct health care services should be delivered in public and not-for-profit health care facilities. There are, however, several grey areas around the issue of private for-profit delivery. First, diagnostic services have expanded considerably in the past few years and, in many cases, these services are provided in private facilities under contracts with regional health authorities or provincial governments. Much

I I 111

111111

7 Delivery of Health Care and Resource Allocation

of this involves relatively routine procedures such as laboratory tests and x-rays triat can be done with little delay or wait on the part of the patient. But there appears to be a growing reliance on the private provision of more advanced and expensive diagnostics such as MRIs (magnetic resonance imaging), for which the waiting times in the public: system can be frustratingly long because of what appears to be an underinvestment in such technology within the public system. The growth of private advanced diagnostic facilities has permitted individuals to purchase faster service by paying for these services out of their own pocket and using the test results to "jump the queue" back into the public system for treatment. While this is not currently a common occurrence, Canadians made it clear to the Commission that they are deeply concerned about the prospect of this becoming routine (Commission 2002a). Medicare rests on the principle that an individual's financial resources should not determine access to services. In the Commission's view, governments have a responsibility to guarantee that the public system has sufficient resources to ensure appropriate access to advanced technology. Increased investment within the public system for new diagnostic technology can remove the temptation to "game" the system by individuals and health care providers through the private purchase of diagnostic tests :hat could allow them to jump the queue. The second grey area is services provided to workers' compensation clients wim job-related injuries and illnesses. Because of the belief that it is important to get these people back to work quickly, these clients get preferential treatment in accessing diagnostic and other health care services over those whose illness or injury is not work-related or who may not be formally employed. . . . This current exception under the Canada Health Act should be reconsidered. The third grey area is contracting out of surgical services. In some cases, regional health authorities have contracted with private for-profit facilities that provide specific surgeries such as cataract and some day surgeries. Again, there is no clear evidence that this practice is more efficient or less costly than providing the services in an adequately resourced not-for-profit facility. Services and Sustainability

Services are the first element in our definition of sustainability The previous information suggests that there are complex, and sometimes confusing, relationships between the federal, provincial, and territorial governments. Much has changed since medicare was first introduced. The range of services is growing and

changing with new advances in medicine and, as a result, the biggest growth in services is outside of hospital and physician services. Subsequent chapters will show that there is tremendous growth in home care and that prescription drugs have become the fastest growing part of the health care system. Canadians also are only too well aware of the fact that services are not always available on a timely basis. In areas like diagnostic services and some surgeries, people sometimes wait too long for access to the services they need. People in rural and remote communities also have problems in accessing services. In spite of what appears to be almost overwhelming support for primary health care, only limited progress has been made in extending primary health care across the country. All of these issues apply in every province and territory. The conclusion, then, on services and sustainability is that more needs to be done to ensure timely access to quality services. The answer, however, is not to look to the private sector for solutions. Instead, governments should seek the best solutions within the public system and ensure that adequate resources are available and services are accessible to all. The Commission is strongly of the view that a properly funded public system can continue to provide the high-quality services to which Canadians have become accustomed. Rather than subsidize private facilities with public dollars, governments should choose to ensure that the public system has sufficient capacity and is universally accessible. In addition, . . . any decisions about expanding private for-profit delivery could have implications under international trade agreements that need to be considered in advance. Needs and Sustainability

The second key dimension of assessing sustainability is needs, namely: does the health care system adequately meet Canadians' needs? The answer is a qualified yes. Canada's health outcomes compare favourably with other countries and evidence suggests that we are doing a good job of addressing factors that affect the overall health of Canadians. There are, however, areas where there is room for improvement. And there are serious disparities in both access to health care and health outcomes in some parts of Canada. Clearly, more needs to be done to reduce these disparities and also to address a number of factors that affect Canadians' health such as tobacco use, obesity, and inactivity . . . The other conclusion is that aging is not the ominous threat to future sustainability of our system that some would suggest. Aging will challenge and add costs to our health care system, but those costs can be managed, particularly if we begin

Roma now: Sustaining Medicare: The Commission on the Future of Health Care in Canada

to prepare and make adjustments to anticipate the impact of an aging population. Resources in the System: The Case of Funding

As was noted at the outset of this chapter, the third major component of the definition of sustainability relates to the availability of necessary resources. The health care system needs a variety of resources in order to deliver services and meet the health care needs of the population. That includes not only financial resources but also human and physical resources such as equipment, facilities, and technology . . . However, the primary focus of much of the debate about sustainability has been about money. Questions about the increasing costs of health care, who pays for what aspects of the health care system, and whether we will be able to afford the health care system in the future have played a significant part in the debates about medicare's sustainability. The debate has centred on whether there is too little public money in the system, whether there should be different ways of raising those public funds, and whether the system as we know it is "affordable" any longer. Because other chapters do not deal with these issues in detail, the remainder of this chapter addresses the fiscal questions directly, beginning with how Canada's funding for health care compares with other countries, whether other options for funding should be considered, and the relative shares paid by different governments. Canada's Reliance on Taxes

Canadians pay directly or indirectly, for every aspect of our health care system through a combination of taxes, payments to government, private insurance premiums, and direct out-of-pocket fees of varying types and amounts. Some have suggested that Canada relies too heavily on taxation to support its health care system. . . . Seven per cent of the total funding for Canada's health care comes from taxation In countries such as Germany Japan, France, and the Netherlands, the majority of funding for health care comes from social insurance premiums in the form of employment payroll taxes. In most developed countries (other than those that rely heavily on social insurance), between 70 and 80 per cent of total health care is funded through the taxation system (Mossialos et al. 2002). . . . [II t is hard to conclude that Canada depends too heavily on taxes to support health care. Use of Private Insurance and Out-of-pocket Payments

One area where Canada differs from most OECD countries is in co-payments and user fees. While Canada

relies almost entirely on taxes to fund hospital and physician services, co-payments and user fees for these services are common in most OECD countries. At the same time, Canada relies more heavily on private insurance and out-of-pocket payments for health care services that are not covered by the Canada Health Act. . . . Dental services, for example, are almost entirely funded (94 per cent) through private insurance and direct fees in Canada but are often part of public coverage in many western European countries. In comparison with selected countries, only Japan and Australia have higher levels of out-of-pocket expenditures than Canada while in the United Kingdom, Sweden, the Netherlands, Germany, and France all have substantially lower levels of out-of-pocket payments. This is because the fees charged in those countries are low and represent a relatively small proportion of the real cost of the services provided. Canadians, however, pay relatively high co-payments and deductibles for prescription drugs and health services outside the CHA and this results in Canada having a higher percentage of out-of-pocket payments than other countries. Even though the co-payments and deductibles are high, the percentage of out-of-pocket payments in Canada accounts for a relatively small percentage of the total costs of health care services and is lower than the OECD average. Canada, like most of the wealthier OECD countries including the United States, relies primarily on funding provided through governments or through insurers. In high-income countries, what we call "third-party" payments (i.e., payments made by governments or insurers) make up between 80 and 90 per cent of health expenditures (OECD 2002). In less wealthy OECD countries, however, there tends to be a much higher reliance on out-of-pocket payments. The Balance between Public and Private Funding of Health Care

There is some debate in Canada about the appropriate balance between public and private funding for health care. Recently, a number of Canadian providers, scholars, and journalists argued in favour of a greater private role in funding Canadian health care on the assumption that Canadian health care spending is overly weighted to the public side (Gratzer 2002). However, a comparison with other industrialized countries shows that Canada is hardly an exception in terms of the public share of total health expenditures. The United Kingdom, Sweden, Germany, France, Japan, and Australia all have larger public health care sectors than

I I ill

IIILLI

7 Delivery of Health Care and Resource Allocation

Canada, while the Netherlands' public share is slightly lower than Canada's. . . What is truly noteworthy is the extent to which these countries' public health care expenditures resemble each other. While most wealthy countries rely heavily on public funding for health care, private insurance plays a significant role in funding health care in the United States. Private insurance in the United States is supported by tax breaks known as "tax expenditure subsidies." These subsidies exist, but to a much lesser extent, in all the comparison countries. Since these subsidies are not generally included when public health care expenditures are tallied, they are difficult to trace and are therefore referred to as "covert" expenditures (Mossialos and Dixon 2002). In fact, tax subsidies play an enormous role in providing health care coverage in the United States. When these tax breaks are taken into account, the public share of health care spending in the United States increases to nearly 60 per cent of its total health care spending (Woolhandler and Himmelstein 2002). This changes the common perception tha: the United States has a predominantly private system of health care. Even without including tax subsidies, the extraordinarily high level of total health care spending in the United States translates into far more spending per capita than in Canada and the other OECD countries. This has been described as tantamount to paying for national health insurance and, in return, getting a fragmented system with significant gaps in coverage—the worst of both worlds. While the United States' "health care system is usually portrayed as largely private," a more apt description is "[plublic money, private control" (Woolhandler and Himmelstein 2002: 22). Indeed, the larger the public share of health care financing beyond tax expenditure subsidies, the more total health expenditures are capable of being controlled. In contrast, the larger the pr.vate share of health care financing, the more difficult it is to control health care expenditures (Majnoni d'Intignano 2001). Alternative Funding Sources

In recent years, a number c f suggestions have been made that Canada should consider alternative ways of paying for health care services. These proposals may be a reaction to the fact that people see costs increasing, are worried about sustainability, and question whether we should change the current funding system to look for additional sources of revenue. Undeniably, each of these proposals has some potential to raise additional money to fund the health care system. But some pose problems in terms of the impact they would have on

access and equity. A number of the most common proposals are critically examined below. User Fees and Out-of-Pocket Co-payments User fees are definitely a "hot button" issue for many Canadians. While many are opposed to user fees because they discourage poorer people from accessing health care services, others see user fees as a necessary way of either raising additional funds for health care or curbing abuse of the health care system. Interestingly, during the Citizens' Dialogue sessions held by the Commission, the interest in user fees was not aimed at raising more revenue for the system but at curbing what some participants felt was abuse and unnecessary use of the system (Commission 2002a). There is overwhelming evidence that direct charges such as user fees put the heaviest burden on the poor and impede their access to necessary health care. This is the case even when low-income exemptions are in place. The result may be higher costs in the long run because people delay treatment until their condition gets worse. In addition, user fees and payments also involve significant administrative costs that directly reduce the modest amount of revenue generated from the fees (Evans 2002a; Evans et al. 1993; Barer et al. 1993, 1979; as 2001). One of the key features of the Canada Health Act was its effective ban on user fees for hospital and physician services. Given what we know about the impact of even relatively low user fees, the Commission feels that this was the right decision then and remains the right decision today. Medical Savings Accounts Perhaps no recent suggestion for raising additional revenue has attracted as much attention as medical savings accounts, in part because they seem to address some of the criticisms of user fees. Medical savings accounts (MSAs) can be designed in a number of different ways but the fundamental concept is that individuals are allotted a yearly health care allowance and they can use it to "purchase" health care services (Gratzer 1999, 2002; Migue 2002; Ramsay 2002). If they have funds left in their MSA allowance at the end of the year, depending on how the plan is designed, they may be able to keep the funds or save them for future years when their health care costs may be higher. MSAs are intended to provide patients with more control and to inject market forces into the organization and delivery of health care services. They provide patients with an incentive to "shop" for the best services and best prices, and to avoid unnecessary

Roma now: Sustaining Medicare: The Commission on the Future of Health Care in Canada

particularly if they get to keep any surplus in their account at the end of the year. If the costs of health care services people use in a year are higher than their yearly allowance, they would be required to pay all or a portion of the additional costs, depending on how the plan was designed. Most MSA proposals discussed in Canada involve a so-called corridor where people pay some of the cost of health care expenses above their annual allowance up to a certain point before catastrophic coverage funded entirely by government would cover any remaining costs (Mazankowski 2001). Because medical savings account approaches are relatively new, we know very little about their effects and the literature to date is contradictory . . . The limited evidence available suggests that medical savings accounts have a number of shortcomings that have been understated or ignored by their proponents (Maynard and Dixon 2002; Shortt 2002; Hurley 2000, 2002; Barr 2001). Overall, mSAs are based on the assumption that the use of necessary health care services is highly discretionary, when this is almost invariably not the case. mSAs are unlikely to effectively control overall spending on health care (Forget et al. 2002). Most health care costs are incurred by a small proportion of people who have very high health care needs and they will continue to spend a lot regardless of whether or not they have an MSA... MSAs may compromise equity in access to health care services. If individuals are required to pay once they have used all of their MSA allowance, it could cause hardships for people with lower incomes or higher health care needs due to chronic or lifethreatening conditions. This is precisely the reason why Canada's medicare system was introduced—to avoid a situation where wealthy people could get access to all the health care services they needed and poor people could not. treatments,

Tax-Based Co-payments, Tax Credits, and Deductibles A number of recent articles have focused on the use of the tax system as a way of increasing private payment in the health care system (Aba et al. 2002; Aba and Mintz 2002; Reuber and Poschmann 2002). The simplest way of doing this would be to include publicly provided health care services as a taxable benefit on individuals' annual income tax returns (Kent 2000). People would get something like a T4-H showing the cost of the health services they received in a year. This amount would be added to their taxable income and they would pay additional taxes to

cover a portion of the cost of the health services they received. On the positive side, this approach would raise additional revenues. People would know the costs of the services they received, and any additional taxes would be based on their ability to pay. On the other hand, the approach could potentially bankrupt people who had chronic health conditions or who suffered a catastrophic illness or injury. To address this concern, the amount of the co-payment or additional taxes a person paid could be capped at a certain percentage of his or her income and very low-income people could be exempt (Aba and Mintz 2002). Even with these conditions, there are concerns with this approach. Fundamentally, it means that if people are sick or injured, they will be taxed more and pay more for health care. This is counter to the basic premise in Canada's health care system that access should be determined only by need and not by ability to pay. As in the case of mSAs or user fees, it may result in people not using needed health care services, a phenomenon that has been seen in a number of European systems (cES 2001). It also raises the question of whether middle- and higherincome earners, who currently pay the bulk of the costs of a universal health care system, will eventually become dissatisfied when they also have to pay even more at tax time based on their use of the health care system. Public—Private Partnerships While different options like user fees, taxable benefits, or medical savings accounts are designed to provide more private payments for health services, other approaches such as public—private partnerships (P3s) are being considered as a way of supporting capital projects. P3s involve a number of different options including long-term outsourcing contracts, joint ventures, strategic partnerships, or private financing models. In the United Kingdom, under private financing initiatives (PH), private-sector firms are awarded long-term contracts to design, build, finance, and operate hospitals. While P3s may be a useful means of bringing the innovation of the private sector to bear, they are not without their critics. In many cases, governments find P3s attractive because the private-sector company assumes the heavy capital costs of a project and governments are only required to pay "rental fees" over the longer term. Unfortunately, while P3s may cost governments and taxpayers less in the short term, these arrangements often cost more in the longer term

II

111111

7 Delivery of Health Care and Resource Allocation

(Sussex 2001). The rental costs charged to governments must be high enough to allow the private-sector partner to recoup its costs and make a profit for its shareholders. The cost of borrowing is often higher for the private sector than for governments. And P3s often have higher administration costs. Critics also suggest that the quality of private, for-profit-run facilities can be lower than publicly run facilities and that, in some cases, these arrangements have resulted in beds being closed and staff being reduced (Pollack et al. 2001). This is not to say that P3s are without a place (for example, in the case of health information systems), but they are no panacea, and their use and value need to be carefully considered. Should Canada Consider Alternative Funding Schemes?

Each of the alternative options outlined above would raise more money for the health care system or free up money for governments to spend on other priorities such as lowering taxes or paying down debt. However, many of the options also compromise the principles and values on which Canadians built the health care system. Some of the options would simply shift the burden of health expenditures from the public purse to individuals and would ultimately undermine the equity that currently exists in both funding and access to needed health care services. Through the Commission's consultations, Canadians indicated that they were willing to pay more in taxes to sustain the health care system, but only if changes are made to improve the current system. Consistent with this view, some have suggested a dedicated tax for health care. This could take a number of different forms. At one end of the spectrum is what public finance experts call a hypothecated tax—a single-purpose tax that is formally separated from all other revenue streams in a special fund similar to the Canada or the Quebec Pension Plans. At the other end of the spectrum, a health tax or premium could be established, but the money flows into the general revenue funds of governments. Both may satisfy the public's desire to ensure some degree of transparency and accountability but they provide less than perfect solutions in other respects. . . . Based on evidence both in Canada and internationally, progressive taxation continues to be the most effective way to fund health care in Canada. From what the Commission heard from Canadians through the Citizens' Dialogue and other consultations, the large majority of Canadians do not want to see any change in the single-payer insurance principle for core

hospital and physician services. There also continues to be a strong consensus among Canadians that "ability to pay" should not be the predominant factor in how we fund key aspects of our health care system. Canadians want necessary hospital and physician services to be fully funded through our taxes. This may be because our tax-funded, universal health care system provides a kind of "double solidarity" It provides equity of funding between the "haves" and the "have-nots" in our society and it also provides equity between the healthy and the sick. . . . Health as a Major Contributor to the Economy

Discussions about health care are most often focused on costs while, in fact, health care is also a major contributor to Canada's economy and economies around the world. According to American economist William Nordhaus, the "medical revolution over the last century appears to qualify, at least from an economic point of view, for Samuel Johnson's accolade as 'the greatest benefit to mankind— (2002: 38). This increase in economic value comes from numerous directions including improvements in: •

• • • •

basic knowledge from the germ theory of disease at the beginning of the twentieth century to the more recent DNA revolution; public health capital and infrastructure; diagnostic tools and processes; logistics in terms of obtaining critical care (e.g., emergency response); and treatment technologies and protocols including pharmaceuticals.

In the early 1990s, rising health care costs were seen in many countries as an obstade to balancing budgets and cutting taxes. This created the view that health care costs were a threat to future national competitiveness. But based on Nordhaus's calculations, it appears that health care spending contributed at least as much to the American economy as spending on all other consumption expenditures combined. Canadian economist Tom Courchene (2001) has made a similar argument about viewing health care expenditures as a dynamic investment in the economy rather than simply as consumption. Health care investments not only lead to longer and more productive working lives on an individual basis; properly targeted public health care investments can also provide countries with a competitive advantage. According to the Canadian Council of Chief

Romanow: Sustaining Medicare:The Commission on the Future of Health Care in Canada

Executives' submission to the Commission (2002: 2), "Canada's business leaders have been strong supporters of Canada's universally accessible public health care system" because it provides a "significant advantage in attracting the people and investment that companies need to stay competitive." Indeed, the "big three" automakers (Ford, General Motors, and Daimler-Chrysler) recently signed joint letters with their largest union, the Canadian Auto Workers, expressing support for Canada's publicly funded health care system and noting that it provides an important competitive advantage to the Canadian auto and auto-parts industries relative to their American counterparts. In short, it is more economical for the employers to pay taxes in support of medicare than to be forced to buy private health insurance for their workers. It is also true that health care is what economists call a superior good in that, as individuals, we tend to spend progressively more on health care than other goods and services as our incomes go up. Based on a series of international studies summarized by Gerdtham and Jonsson (2000), higher income is the single most important factor determining higher levels of health spending in all countries. Indeed, the more economically developed the country, the more pronounced the effect (Scheiber and Maeda 1997). According to Reinhardt et al. (2002: 171), per capita GDP is without doubt "the most powerful explanatory variable for international differences in health spending."

Resources and Sustainability

What conclusions can we draw about resources and sustainability? Canada's spending on health care is comparable with other OECD countries although we spend considerably less per capita than the United States. All OECD countries are facing increasing health care costs and experience suggests that the wealthier the country, the more it spends on health care. Some suggest that Canada relies too heavily on taxation, and yet, comparisons show that we are not much different from other countries. A look at various alternative ways of funding health care shows that each option raises a number of problems and many would simply shift the burden of funding from governments to individual Canadians. At the same time, there are serious problems in the balance between federal and provincial—territorial funding for health care, and health care is taking up an increasing proportion of provincial budgets. Later chapters of this report address specific ways in which steps can be taken to control rising costs, especially for prescription drugs. But the reality is that health care costs are likely to continue to increase and choices have to be made about how those costs will be managed. Overwhelmingly, Canadians told the Commission that they are prepared to pay more for health care to ensure the system's sustainability, provided the system is prepared to change to meet their needs and expectations.

References Aba, S., W.D. Goodman, and J.M. Mintz. 2002. Fund-

ing Public Provision of Private Health: The Case for a Copayment Contribution through the Tax System. C.D. Howe Institute Commentary 163. C.D. Howe Institute: Toronto. Aba, S., and J. Mintz. 2002. "Should Public Health Care Benefits Be Included as Part of Taxable Income?" Paper presented at the Roundtable on Financing Options for the Commission on the Future of Health Care in Canada. 24 May, C.D. Howe Institute, Toronto. Barer, M., V. Bhatia, G.L. Stoddart, and R.G. Evans. 1993.

The Remarkable Tenacity of User Charges: A Concise History of the Participation, Positions, and Rationales of Canadian Interest Groups in the Debate over "Direct Patient Participation" in Health Care Financing. Centre for Health Services and Policy Research, University of British Columbia: Vancouver. Barr, M.S. 2001. "Medical Savings Accounts in Singapore: A Critical Inquiry," in Journal of Health Politics, Policy and Law 26 (4): 709-26.

Canadian Council of Chief Executives. 2002. Written submission to the Commission on the Future of Health Care in Canada, entitled "Shared Enterprise: Sustaining and Improving Health Care for Canadians." as (College des Economistes de la Sante). 2001. "Utilisation Fees Imposed to Public Health Care System Users in Europe." Proceedings of the Workshop on 29 November organised for the Commission on the Future of Health Care in Canada College des Economistes de la Sante: Paris. Commission. 2002a. Report on Citizens' Dialogue on the Future of Health Care in Canada. Prepared for the Commission on the Future of Health Care in Canada by J. Maxwell, K. Jackson, and B. Legowski (Canadian Policy Research Networks), S. Rosell and D. Yankelovich (Viewpoint Learning), in collaboration with P.-G. Forest and L Lozowchuk (Commission on the Future of Health Care in Canada). Commission on the Future of Health Care in Canada: Saskatoon. Courchene, T. 2001. A State of Minds: Toward a Human Capital Future for Canadians. Institute for Research on Public Policy: Montreal.

O 111

366

II

11

II ill

111111

7 Delivery of Health Care and Resource Allocation

Deber, R. 2002. "Delivering F.:ealth Care Services: Public, Not-for-Profit, or Private?" Discussion Paper prepared for the Commission on the Future of Health Care in Canada. Devereaux, P.J., P.T.L. Choi, C. Lacchetti, B. Weaver, H.J. Schunemann, T. Haines, J.N. Lavis, B.J.B. Grant, D.R.S. Haslam, M. Bhandari, T. Sullivan, D.J. Cook, S.D. Walter, M. Meade, H. Khan, N. Ehatnagar, and G.H. Guyatt. 2002. "A Systematic Review and Meta-Analysis of Studies Comparing Mortality Rates of Private For-Profit Hospitals and Private Not-For-Profit Hospitals," in ow 166 (11): 1399-1406. Evans, R.G. 2002a. "Financing Health Care: Taxation and the Alternatives," in Funding Health Care: Options for Europe, ed. E Mossialos, A. Dixon, and J. Figueras. Open University Press: Buckingham, 31-58. - . 2002b. "Raising the Money: Options, Consequences and Objectives for Financing Health Care in Canada." Discussion Paper prepared for the Commission on the Future of Health Care in Canada. Evans, R.G., M.L. Barer, G.I. Stoddart, and V Bhatia. 1993. It's Not for the Money, It's the Principle: Why User Charges for Some Services and Not Others? Centre for Health Services and Policy Research, University of British Columbia: Vancouver. Forget, E.L., R. Deber, and L.L. Roos. 2002. "Medical Savings Accounts: Will They Reduce Costs?," in CMAJ 167 (2): 143-7. Gratzer, D., ed. 2002. Better Medicine: Reforming Canadian Health Care. ECW Press: Toronto. Gratzer, D. 1999. Code Blue: Reviving Canada's Health Care System. ECW Press: Toronto. Gray, B.H. 1999. The Empirical Literature Comparing For-

Profit and Nonprofit Hospitals, Managed Care Organizations and Nursing Homes: Updating the Institute of Medicine Study. Coalition for Nonprofit Healthcare: Washington, DC. Harrington, C. 2001. "Residential Nursing Facilities in the United States," in smj 323: 507-10. Hurley, J. 2002. "Medical Savings Accounts Will Not Advance Canadian Health Care Objectives," in ow 167 (2): 152-3. Kent, T. 2000. What Should Be Done about Medicare? Caledon Institute of Public Policy: Ottawa. Kleinke, J.D. 2001. Oxymorons: The Myth of a U.S. Health Care System. Jossey-Bass: San Francisco. McFarlane, L., and C. Prado. 2002. The Best-Laid Plans: Health Care's Problems and Prospects. McGill-Queen's University Press: Montreal. Majnoni d"Intignano, B. 2001. Sante et economie en Europe. Presses Universitaires de France: Paris. Marmor, T.R. M. Schlesinger, and R.W. Smithey. 1987. "Nonprofit Organizations and Healthcare," in The Nonprofit Sector: A Research Handbook, ed. W.W. Powell. Yale University Press: New Haven, 1-35.

Maynard, A., and A. Dixon. 2002. "Private Health Insurance and Medical Savings Accounts: Theory and Experience," in Funding Health Care: Options for Europe, ed. E. Mossialos, A. Dixon, J. Figueras, and J. Kutzin. Open University Press: Buckingham, 109-27. Mazankowski, D. 2001. [Alberta 2001.] A Framework for

Reform. Report of the Premier's Advisory Council on Health. D. Mazankowski, Chair. Premier's Advisory Council on Health: Edmonton. Migue, Jean-Luc. 2002. "Funding and Production of Health Services: Outlook and Potential Solutions." Discussion Paper prepared for the Commission on the Future of Health Care in Canada. Mossialos, E., and A. Dixon. 2002. "Funding Health Care: An Introduction," in Funding Health Care: Options for Europe ed. E Mossialos, A. Dixon, J. Figueras, and J. Kutzin. Open University Press: Buckingham, 1-30. Mossialos, E., A. Dixon, J. Figueras, and J. Kutzin, eds. 2002. Funding Health Care: Options for Europe. Open University Press: Buckingham. OECD (Organisation for Economic Co-operation and Development). 2002. OECD Health Data 2002b. A Comparative Analysis of 30 Countries. Organisation for Economic Co-operation and Development: Paris. Pollack, A., J. Shaoul, D. Rowland, and S. Player. 2001. "Public Services and the Private Sector: A Response t o the IPPR." Working paper. Catalyst: London. Ramsay, C. 2002. "A Framework for Determining the Extent of Public Financing of Programs and Services." Discussion Paper prepared for the Commission on the Future of Health Care in Canada. Reinhardt, U.E., P.S. Hussey, and G. F. Anderson. 2002. "Cross-National Comparisons of Health Systems Using OECD Data, 1999," in Health Affairs 21 (3): 169-81. Reuber, G.L., and F. Poschmann. 2002. "Increasing Patient Incentives to Improve the Financial Stability of the Health Care System." Paper presented 24 May at the Roundtable on Financing Options for the Commission on the Future of Health Care in Canada. C.D. Howe Institute. Scheiber, G., and A. Maeda 1997. "A Curmudgeon's Guide to Financing Health Care in Developing Countries," in "Innovation in Health Care Financing," ed. G.J. Schieber. World Bank Discussion Paper No. 365. World Bank Washington, DC, 1-38. Shortt, S.E.D. 2002. "Medical Savings Accounts in Publicly Funded Health Care Systems: Enthusiasm Versus Evidence," in ami 167 (2), 159-62. Sussex, J. 2001. The Economics of the Private Finance Initiative in the NHS. The Office of Health Economics: London. Woolhandler, S., and D.U. Himmelstein. 2002. "Paying for National Health Insurance-and Not Getting It," in Health Affairs 21 (4): 88-98.

Buchanan: The Right to a Decent Minimum of Health Care

73 The Right to Health Care, Macroallocation, and Setting Priorities

The Right to a Decent Minimum of Health Care Allen E. Buchanan The Assumption That There Is a Right to a Decent Minimum

A consensus that there is (at least) a right to a decent minimum of health care pervades recent policy debates and much of the philosophical literature on health care. Disagreement centres on two issues. Is there a more extensive right than the right to a decent minimum of health care? What is included in the decent minimum to which there is a right? Preliminary Clarification of the Concept

Different theories of distributive justice may yield different answers both to the question "Is there a right to a decent minimum?" and to the question "What comprises the decent minimum?" The justification a particular theory provides for the claim that there is a right to a decent minimum must at least cohere with the justifications it provides for other right-claims. Moreover, the character of this justification will determine, at least in part, the way in which the decent minimum is specified, since it will include an account of the nature and significance of health care needs. To the extent that the concept of a decent minimum is theorydependent, then, it would be naive to assume that a mere analysis of the concept of a decent minimum would tell us whether there is such a right and what its content is. Nonetheless, before we proceed to an examination of various theoretical attempts to ground and specify a right to a decent minimum, a preliminary analysis will be helpful. Sometimes the notion of a decent minimum is applied not to health care but to health itself, the claim being that everyone is entitled to some minimal level, or welfare floor, of health. I shall not explore this variant of the decent minimum idea because I think its implausibility is obvious. The main difficulty is that assuring any significant level of health for all is simply not within the domain of social control. If the alleged

right is understood instead as the right to everything which can be done to achieve some significant level of health for all, then the claim that there is such a right becomes implausible simply because it ignores the fact that in circumstances of scarcity the total social expenditure on health must be constrained by the need to allocate resources for other goods. Though the concept of a right is complex and controversial, for our purposes a partial sketch will do. To say that person A has a right to something, x, is first of all to say that A is entitled to x, that xis due to him or her. This is not equivalent to saying that if A were granted x it would be a good thing, even a morally good thing, or that x is desired by or desirable for A. Second, it is usually held that valid right-claims, at least in the case of basic rights, may be backed by sanctions, including coercion if necessary (unless doing so would produce extremely great disutility or grave moral evil), and that (except in such highly exceptional circumstances) failure of an appropriate authority to apply the needed sanctions is itself an injustice. Recent rights-theorists have also emphasized a third feature of rights, or at least of basic rights or rights in the strict sense: valid right-claims "trump" appeals to what would maximize utility, whether it be the utility of the right-holder, or social utility. In other words, if A has a right to x, then the mere fact that infringing A's right would maximize overall utility or even A's utility is not itself a sufficient reason for infringing it.' Finally, a universal (or general) right is one which applies to all persons, not just to certain individuals or classes because of their involvement in special actions, relationships, or agreements. The second feature—enforceability—is of crucial importance for those who assume or argue that there is a universal right to a decent minimum of health care. For, once it is granted that there is such a right and that such a right may be enforced (absent any extremely weighty reason against enforcement), the claim that there is a universal right provides the moral basis for using the coercive power of the state to assure a decent minimum for all. Indeed, the surprising absence of attempts to justify a coercively backed decent minimum policy by arguments that do not aim at establishing a universal right suggests the following hypothesis: advocates of a coercively backed decent minimum have operated on

Source: From President's Commission: Securing Access to Health Care, vol. 2 (Washington, DC: US Government Printing Office, 1983). Reprinted by permission of the author.

11 311

368

MILLI

7 Delivery of Health Care and Resource Allocation

the assumption that such a policy must be based on a universal right to a decent minimum. The chief aim of this article is to show that this assumption is false. I think it is fair to say that many who confidently assume there is a (universal) right to a decent minimum of health care have failed to appreciate the significance of the first feature of our sketch of the concept of a right. It is crucial to observe that the claim that there is a right to a decent minimum is much stronger than the claim that everyone ought to have access to such a minimum, or that if they did :Lt would be a good thing, or that any society which is capable, without great sacrifice, of providing a decent minimum but fails to do so is deeply morally defective. None of the latter assertions implies the existence of a right, if this is understood as a moral entitlement which ought to be established by the coercive power of the state if necessary . . . The Attractions of the Idea of a Decent Minimum

There are at least three features widely associated with the idea of a right to a decent minimum which, together with the facile consensus that vagueness promotes, help explain its popularity over competing conceptions of the right to health care. First, it is usually, and quite reasonably, assumed that the idea of a decent minimum is to be understood in a society-relative sense. Surely it is plausible to assume that, as with other rights to goods or services, the content of the right must depend upon the resources available in a given society and perhaps also upon a certain consensus of expectations among its members. So the first advantage of the idea of a decent minimum, as it is usually understood, is that it allows us to adjust the level of services to be provided as a matter of right to relevant social conditions and also allows for the possibility that as a society becomes more affluent the floor provided by the decent minimum should be raised. Second, the idea of a decent minimum avoids the excesses of what has been called the strong equal access principle, while still acknowledging a substantive universal right. According to the strong equal access principle, everyone has an equal right to the best health care services available. Aside from the weakness of the justifications offered in support of it, the most implausible feature of the strong equal access principle is that it forces us to choose between two unpalatable alternatives. We can either set the publicly guaranteed level of health care lower than the level that is technically possible or we can set it as high as is technically possible. In the former case, we shall be committed to the uncomfortable conclusion that no matter how many resources

have been expended to guarantee equal access to that level, individuals are forbidden to spend any of their resources for services not available to all. Granted that individuals are allowed to spend their after-tax incomes on more frivolous items, why shouldn't they be allowed to spend it on health? If the answer is that they should be so allowed, as long as this does not interfere with the provision of an adequate package of health care services for everyone, then we have retreated from the strong equal access principle to something very like the principle of a decent minimum. If, on the other hand, we set the level of services guaranteed for all so high as to eliminate the problem of persons seeking extra care beyond this level, this would produce a huge drain on total resources, foreclosing opportunities for producing important goods other than health care. So both the recognition that health care must compete with other goods and the conviction that beyond some less than maximal level of publicly guaranteed services individuals should be free to purchase additional services point toward a more limited right than the strong access principle asserts. Thus, the endorsement of a right to a decent minimum may be more of a recognition of the implausibility of the stronger right to equal access than a sign of any definite position on the content of the right to health care. A third attraction of the idea of a decent minimum is that since the right to health care must be limited in scope (to avoid the consequences of a strong equal access right), it should be limited to the "most basic" services, those normally "adequate" for health, or for a "decent" or "tolerable" life. However, although this aspect of the idea of a decent minimum is useful because it calls attention to the fact that health care needs are heterogeneous and must be assigned some order of priority, it does not itself provide any basis for determining which are most important. The Need for a Supporting Theory

In spite of these attractions, the concept of a right to a decent minimum of health care is inadequate as a moral basis for a coercively backed decent minimum policy in the absence of a coherent and defensible theory of justice. Indeed, when taken together they do not even imply that there is a right to a decent minimum. Rather, they only support the weaker conditional claim that if there is a right to health care, then it is one that is more limited than a right of strong equal access, and is one whose content depends upon available resources and some scheme of priorities which shows certain health services to be more basic than others. It appears, then, that a theoretical grounding for the right to a decent minimum of health care is indispensable. . . .

Buchanan: The Right to a Decent Minimum of Health Care

My suggestion is that the combined weight of arguments from special (as opposed to universal) rights to health care, harm prevention, prudential arguments of the sort used to justify public health measures, and two arguments that show that effective charity shares features of public goods (in the technical sense) is sufficient to do the work of an alleged universal right to a decent minimum of health care. Arguments from Special Rights

The right-claim we have been examining (and find unsupported) has been a universal right-claim: one that attributes the same right to all persons. Special right-claims, in contrast, restrict the right in question to certain individuals or groups. There are at least three types of arguments that can be given for special rights to health care. First, there are arguments from the requirements of rectifying past or present institutional injustices. It can be argued, for example, that American blacks and Native Americans are entitled to a certain core set of health care services owing to their history of unjust treatment by government or other social institutions, on the grounds that these injustices have directly or indirectly had detrimental effects on the health of the groups in question. Second, there are arguments from the requirements of compensation to those who have suffered unjust harm or who have been unjustly exposed to health risks by the assignable actions of private individuals or corporations—for instance, those who have suffered neurological damage from the effects of chemical pollutants. Third, a strong moral case can be made for special rights to health care for those who have undergone exceptional sacrifices for the good of society as a whole— in particular those whose health has been adversely affected through military service. The most obvious candidates for such compensatory special rights are soldiers wounded in combat. Arguments from the Prevention of Harm

The content of the right to a decent minimum is typically understood as being more extensive than those traditional public health services that are usually justified on the grounds that they are required to protect the citizenry from certain harms arising from the interactions of persons living together in large numbers. Yet such services have been a major factor—if not the major factor—in reducing morbidity and mortality rates. Examples include sanitation and immunization The moral justification of such measures, which constitute an important element in a decent minimum of health care, rests upon the widely accepted Harm (Prevention) Principle, not upon a right to health care.

The Harm Prevention argument for traditional public health services, however, may be elaborated in a way that brings them closer to arguments for a universal right to health care. With some plausibility one might contend that once the case has been made for expending public resources on public health measures, there is a moral (and perhaps Constitutional) obligation to achieve some standard of equal protection from the harms these measures are designed to prevent. Such an argument, if it could be made out, would imply that the availability of basic public health services should not vary greatly across different racial, ethnic, or geographic groups within the country. Prudential Arguments

Prudent arguments for health care services typically emphasize benefits rather than the prevention of harm. It has often been argued, in particular, that the availability of certain basic forms of health care make for a more productive labour force or improve the fitness of the citizenry for national defence. This type of argument, too, does not assume that individuals have moral rights (whether special or universal) to the services in question. It seems very likely that the combined scope of the various special health care rights discussed above, when taken together with harm prevention and prudential arguments for basic health services and an argument from equal protection through public health measures, would do a great deal toward satisfying the health care needs which those who advocate a universal right to a decent minimum are most concerned about. In other words, once the strength of a more pluralistic approach is appreciated, we may come to question the popular dogma that policy initiatives designed to achieve a decent minimum of health care for all must be grounded in a universal moral right to a decent minimum. This suggestion is worth considering because it again brings home the importance of the methodological difficulty encountered earlier. Even if, for instance, there is wide consensus on the considered judgment that the lower health prospects of inner-city blacks are not only morally unacceptable but an injustice, it does not follow that this injustice consists of the infringement of a universal right to a decent minimum of health care. Instead, the injustice might lie in the failure to rectify past injustices or in the failure to achieve public health arrangements that meet a reasonable standard of equal protection for all. Two Arguments for Enforced Beneficence

The pluralistic moral case for a legal entitlement to a decent minimum of health care (in the absence of a

■ IN I

370

II

I I III

111111

7 Delivery of Health Care and Resource Allocation

universal moral right) may be strengthened further by non-rights-based arguments from the principle of beneficence.' The possibility of making out such arguments depends upon the assumption that some principles may be justifiably enforced even if they are not principles specifying valid right-claims. There is at least one widely recognized class of such principles requiring contribution to the production of "public goods" in the technical sense (for example, tax laws requiring contribution to national defence). It is characteristic of public goods that each individual has an incentive to withhold his contribution to the collective goal even though the net result is that the goal will not be achieved. Enforcement of a principle requiring all individuals to contribute to the goal is necessary to overcome the individud.'s incentive to withhold contribution by imposing penalties for his own failure to contribute and by assuring him that others will contribute. There is a spec:al subclass of principles whose enforcement is justified not only by the need to overcome the individua:.'s incentive to withhold compliance with the principle but also to ensure that individuals' efforts are appropriately coordinated. For example, enforcing the rule of the road to drive only on the right not only ensures a joint effort toward the goal of safe driving but also coordinates individuals' efforts so as to make the attainment of that goal possible. Indeed, in the case of the "rule of the road" a certain kind of coordinated joint effort is the public good whose attainment justifies enforcement. But regardless of whether the production of a public good requires the solution of a coordination problem or not, there may be no right that is the correlative of the coercively backed obligation specified by the principle. There are two arguments for enforced beneficence, and they each depend upon both the idea of coordination and on certain aspects of the concept of a public good. Both arguments begin w.th an assumption reasonable libertarians accept: there is a basic moral obligation of charity or beneficence to those in need. In a society that has the resources and technical knowledge to improve health or at least to ameliorate important health defects, the application of this requirement of beneficence includes the prcvision of resources for at least certain forms of health care. If we are sincere, we will be concerned with the efficacy of our charitable or beneficent impulses. It is all well and good for the libertarian to say that voluntary giving can replace the existing array of government entitlement programs, but this possibility will be cold comfort to the needy if, for any of several reasons, voluntary giving falters.

Social critics on the left often argue that in a highly competitive acquisitive society such as ours it is naive to think that the sense of beneficence will win out over the urgent promptings of self-interest. One need not argue, however, that voluntary giving fails from weakness of the will. Instead one can argue that even if each individual recognizes a moral duty to contribute to the aid of others and is motivationally capable of acting on that duty, some important forms of beneficence will not be forthcoming because each individual will rationally conclude that he should not contribute. Many important forms of health care, especially those involving large-scale capital investment for technology, cannot be provided except through the contributions of large numbers of persons. This is also true of the most important forms of medical research. But if so, then the beneficent individual will not be able to act effectively, in isolation. What is needed is a coordinated joint effort. First Argument There are many ways in which I might help others in need. Granted the importance of health, providing a decent minimum of health care for all, through large-scale collective efforts, will be a more important form of beneficence than the various charitable acts A, B, and C, which I might perform independently, that is, whose success does not depend upon the contributions of others. Nonetheless, if I am rationally beneficent I will reason as follows: either enough others will contribute to the decent minimum project to achieve this goal, even if I do not contribute to it; or not enough others will contribute to achieve a decent minimum, even if I do contribute. In either case, my contribution will be wasted. In other words, granted the scale of the investment required and the virtually negligible size of my own contribution, I can disregard the minute possibility that my contribution might make the difference between success and failure. But if so, then the rationally beneficent thing for me to do is not to waste my contribution on the project of ensuring a decent minimum but instead to undertake an independent act of beneficence, A, B, or c —where I know my efforts will be needed and efficacious. But if everyone, or even many people, reason in this way, then what we each recognize as the most effective form of beneficence will not come about. Enforcement of a principle requiring contributions to ensuring a decent minimum is needed. The first argument is of the same form as standard public goods arguments for enforced contributions to national defence, energy conservation, and many other

Buchanan: The Right to a Decent Minimum of Health Care

goods, with this exception. In standard public goods arguments, it is usually assumed that the individual's incentive for not contributing is self-interest and that it is in his interest not to contribute because he will be able to partake of the good, if it is produced, even if he does not contribute. In the case at hand, however, the individual's incentive for not contributing to the joint effort is not self-interest, but rather his desire to maximize the good he can do for others with a given amount of his resources. Thus if he contributes but the goal of achieving a decent minimum for all would have been achieved without his contribution, then he has still failed to use his resources in a maximally beneficent way relative to the options of either contributing or not to the joint project, even though the goal of achieving a decent minimum is attained. The rationally beneficent thing to do, then, is not to contribute, even though the result of everyone's acting in a rationally beneficent way will be a relatively ineffective patchwork of small-scale individual acts of beneficence rather than a large-scale, coordinated effort. Second Argument I believe that ensuring a decent minimum of health care for all is more important than projects A, B, or C, and I am willing to contribute to the decent minimum project, but only if I have assurance that enough others will contribute to achieve the threshold of investment necessary for success. Unless I have this assurance, I will conclude that it is less than rational—and perhaps even morally irresponsible—to contribute my resources to the decent minimum project. For my contribution will be wasted if not enough others contribute. If I lack assurance of sufficient contributions by others, the rationally beneficent thing for me to do is to expend my "beneficence budget" on some less-than-optimal project A, B, or C, whose success does not depend on the contribution of others. But without enforcement, I cannot be assured that enough others will contribute, and if others reason as I do, then what we all believe to be the most effective form of beneficence will not be forthcoming. Others may fail to contribute either because the promptings of self-interest overpower their

sense of beneficence, or because they reason as I did in the First Argument, or for some other reason. Both arguments conclude that an enforced decent minimum principle is needed to achieve coordinated joint effort. However, there is this difference. The Second Argument focuses on the assurance problem, while the first does not. In the Second Argument all that is needed is the assumption that rational beneficence requires assurance that enough others will contribute. In the First Argument the individual's reason for not contributing is not that he lacks assurance that enough others will contribute, but rather that it is better for him not to contribute regardless of whether others do or not. Neither argument depends on an assumption of conflict between the individual's moral motivation of beneficence and his inclination of self-interest. Instead the difficulty is that in the absence of enforcement, individuals who strive to make their beneficence most effective will thereby fail to benefit the needy as much as they might. A standard response to those paradoxes of rationality known as public goods problems is to introduce a coercive mechanism which attaches penalties to noncontribution and thereby provides each individual with the assurance that enough others will reciprocate so that his contribution will not be wasted and an effective incentive for him to contribute even if he has reason to believe that enough others will contribute to achieve the goal without his contribution. My suggestion is that the same type of argument that is widely accepted as a justification for enforced principles requiring contributions toward familiar public goods provides support for a coercively backed principle specifying a certain list of health programs for the needy and requiring those who possess the needed resources to contribute to the establishment of such programs, even if the needy have no right to the services those programs provide. Such an arrangement would serve a dual function: it would coordinate charitable efforts by focusing them on one set of services among the indefinitely large constellation of possible expressions of beneficence, and it would ensure that the decision to allocate resources to these services will become effective.

Notes 1. 2.

Ronald Dworkin, (1977), Taking Rights Seriously (Harvard University Press: Cambridge, MA), 184-205. For an exploration of various arguments for a duty of beneficence and an examination of the relationship between justice and beneficence, in general and in

health care specifically, see Allen E. Buchanan, (1982), "Philosophical Foundations of Beneficence," in Beneficence and Health Care, ed. Earl E. Shelp (Reidel Publishing Co.: Dordrecht, Holland).

II ill

111111

7 Delivery of Health Care and Resource Allocation

Justice and the High Cost of Health Ronald Dworkin

1. . .1 II For millennia doctors have paid lip service, at least, to an ideal of justice in medi:ine which I shall call the rescue principle. It has two connected parts. The first holds that life and health are, as Rene Descartes put it, chief among all goods: everything else is of lesser importance and must be sacrificed for them. The second insists that health care must be distributed on grounds of equality: that even in a society in which wealth is very unequal and equality is otherwise scorned, no one must be denied the medical care :le needs just because he is too poor to afford it. These are understandable, even noble, ideals. They are grounded in a shared human understanding of the horror of pain, and, beyond that, of the indispensability of life and health to everything else we do. The rescue principle is so ancient, so intuitively attractive, and so widely supported in political rhetoric that it might easily be thought to supply the right standard for answering questions about rationing. In fact, however, the rescue principle is almost wholly useless for that purpose, and the assumption that it sets the proper standard for health-care reform has done more harm than good. The principle does offer an answer to the question of how mucl-. America should spend on health care overall: it says we should spend all we can until the next dollar would buy no gain in health or life expectancy at all. No sane society would try to meet that standard, any more than a sane person would organize his life on that principle. In past centuries, however, there was not so huge a gap between the rhetoric of the rescue principle and what it was medically possible for a community to do. But now that science has created so many vastly expensive forms of medical care, it is preposterous that a community should treat longer life as a good that it must provide at any cost—even one that would make the longer lives of its people lives barely worth living. So the rescue principleb answer to the question of how much a society should spend on health care overall must be rejected as incredible. Once that answer is rejected, the principle has no second-best or fallback level of advice: it simply is silent. That is worse than unhelpful, because it encouragers the idea that justice has nothing to say about how much a society should spend on health care as against other goods, like education or controlling crime or material prosperity or the arts.

The rescue principle does have something helpful, though negative, to say about the other question of justice, which is how health care should be distributed. It says that if rationing is necessary, it should not be done, as it now largely is in the United States, on the basis of money. But we need more positive advice: What should the basis of rationing be? The egalitarian impulse of the principle suggests that medical care should be distributed according to need. But what does that mean— how is need to be measured? Does someone "need" an operation that might save his life but is highly unlikely to do so? Is someone's need for life-saving treatment affected by the quality his life would have if the treatment were successful? Does the age of the patient matter—does someone need or deserve treatment less at seventy than at a younger age? Why? How should we balance the need of many people for relief from pain or incapacity against the need of fewer people for lifesaving care? At one point the procedures of an Oregon commission appointed to establish medical priorities ranked tooth-capping ahead of appendectomy, because so many teeth can be capped for the price of one operation. Why was that so clearly a mistake? We need a different, more helpful statement of ideal justice in health care, and we should start by noticing one problem that seems to make reform mandatory. Why does America spend so much—so much more than other nations—on medicine? In large part because individual decisions about how much health care to buy are made by patient and doctor but paid for by a third party, the insurance company, so that those who make the decisions have no direct incentive to save money. Insurance premiums are tax-deductible, moreover, and an employer's contribution is not treated as part of the employee's taxable income. So health insurance makes patients insensitive to cost at the moment of decision, and the real price of that insurance is subsidized by the nation. People would probably spend less on their own or their family's care if they had to pay the actual cost themselves, at the expense of other goods and opportunities they might also want or want their families to have. Of course, in the long run most people do pay the true costs of their health care, but they do so indirectly and unwisely, because employer contributions and tax funds could be used to buy what they would choose to have if they made the choice themselves: better schools for their children, for example, or economic investments and programs that would improve America's

Source: From Sovereign Virtue: The Theory and Practice of Equality, Chapter 8, pp. 309-18 (Harvard University Press, 2002). Reprinted with permission.

Dworkin: Justice and the High Cost of Health

competitiveness and give them greater job security. Our medical expenditures are therefore irrational: the system makes choices for people that they would not make for themselves, and the result is that our collective expenditures are too high—measured, as they should be, by how much care we really want, taken together, at the price we really want to pay. Conservative economists seize on this fact: they say we should create a free market in health care by removing all tax benefits and subsidies so that people can have only the care they can afford. While that is, of course, an unacceptable solution, it is important to see why. It is unacceptable for three reasons. First, wealth is so unfairly distributed in America that many people would be unable to buy any substantial health insurance at market rates. Second, most people have very inadequate information about health risks and medical technology; they do not know what the risk of breast cancer is before the age of fifty, for example, or how many years having routine mammography before that age would add to their life expectancy. Third, in an unregulated market, insurance companies would charge some people higher premium rates because they were greater health risks (as, indeed, many insurance companies now do) so that people with a poor health history, or who were members of ethnic groups particularly susceptible to certain diseases, or who lived in areas where the risk of violent injury was greater would be charged prohibitive rates. This analysis points to a more satisfactory ideal of justice in health care—the "prudent insurance" ideal. We should allocate resources between health and other social needs, and between different patients who need treatment, by trying to imagine what health care would be like if it were left to a free and unsubsidized market and if the three deficiencies I have just described were somehow corrected. So try to imagine that America is transformed in three ways. Suppose, first, that the distribution of wealth and income is as fair as it possibly can be. In my view, that means that the resources people can initially command, in making their decisions about education, work, and investment, are as nearly equal as possible; but you should imagine an economic distribution that is fair according to your own views, whatever these are. (I shall assume, however, that on your views, as on mine, the wealth of everyone in a fair society would be much closer to the average than is true in America now: the great extremes between rich and poor that mark our economic life now would have largely disappeared.) Second, imagine that America has also changed so that all the information that might be called state-ofthe-art knowledge about the value and cost and side

effects of particular medical procedures—everything, in other words, that good doctors know—is generally known by the public at large as well. Third, imagine that no one—including insurance companies—has any information available about how likely any particular person is to contact any particular disease or to suffer any particular kind of accident. No one would be in a position to say, of himself or anyone else, that that person is more or less likely to contract sickle-cell anemia or diabetes, or to be the victim of violence in the street, than anyone else.' The changes I am asking readers to imagine are very great, but they are not, I think, beyond the reach of the imagination. Now suppose that health-care decisions in this transformed community are left simply to individual market decisions in as free a market as we can imagine so that doctors and hospitals and drug companies are free to charge whatever they wish. Medical care is not provided by the government for anyone, nor are medical expenses or health-insurance premiums tax-deductible. There is no need to subsidize medical care in any such way, because people have enough resources to buy for themselves the medical care they decide is appropriate. What kind of health-care institutions would actually develop in such a community? Would most people join health maintenance organizations that provided care by staff doctors at a relatively inexpensive rate? Would any substantial number choose more expensive insurance arrangements that allowed more freedom of choice in doctors or hospitals? Would the average plan or policy provide coverage for routine medical examinations or diagnostic screenings? What kind, how often, and at what age? How many plans or policies would provide, at appropriately high rates or premiums, experimental or very expensive or high-risk or low-expected-benefit procedures of different kinds? How much of its aggregate resources would the community devote to medical care through these various individual decisions? It is impossible to answer these questions with any precision.2 But we can nevertheless make two crucial claims about justice. First, whatever that transformed community actually spends on health care in the aggregate is the morally appropriate amount for it to spend: it could not be criticized, on grounds of justice, for spending either too much or too little. Second, however health care is distributed in that society is just for that society: justice would not require providing health care for anyone that he or his family had not purchased. These claims follow directly from an extremely appealing assumption: that a just distribution is one that wellinformed people create for themselves by individual choices, provided that the economic system and the

II III

111111

7 Delivery of Health Care and Resource Allocation

distribution of wealth in the community in which these choices are made are themselves just.3 These important conclusions help us to decide what health care we should aim to provide for everyone in our own, imperfect, and unjust community. We can speculate about what kind of medical care and insurance it would be prudent for most Americans to buy for themselves if the changes I :Imagined had really taken place, and we can use those speculations as guidelines in deciding what justice requires now—in deciding, for example, which medical tests and procedures the National Health Board should decide are "necessary and appropriate" if the Health Security Act is passed. Consider a twenty-five-year-old with average wealth and prospects and state-of-the-art knowledge of medicine. Suppose he can choose from a wide variety of possible arrangements to provide for the health care he might want, under various contingencies, over the course of his life. What arrar gements would it be prudent for him to make?4 He might be tempted, initially, to buy insurance providing every form of treatment or care that might conceivably be beneficial for him under any circumstance. But he would soon realize that the cost of such wildly ambitious insurance would be prohibitive—he would have nothing left for anything else—and decide that prudence required a much less comprehensive insurance program. Of course, what is prudent for someone depends on that person's own individual needs, tastes, personality, and preferences, but we can nevertheless make some judgments with confidence that they would fit the needs and preferences of most contemporary Americans. We can be confident, for example, about what medical insurance it would not be prudent for most people to buy, because some insurance would be a mistake no matter what happened in the future, including the worst outcome. It woule be irrational for almost any twenty-five-year-old to insure himself as to provide for life-sustaining treatment if he falls into a persistent vegetative state, for example. The substantial sum he would have to spend in insurance premiums, year by year, to provide that coverage would be much better used in other ways to enhance his actual, conscious life. Even someone who lived only a few months after purchasing the insurance before he fell into a vegetative state would have made, in retrospect, a mistake, giving up resources that could have made his short remaining conscious life better in order to buy a longer unconscious state. We can enlarge this claim to include dementia as well as unconsciousness: it would not be prudent, for almost anyone, to purchase insurance providing for expensive medical intervention, even of a life-saving

character, after he entered the late stages of Alzheimer's disease or other form of irreversible dementia. The money spent on premiums for such insurance would have been better spent, no matter what happens, in making life before dementia more worthwhile. Of course, most prudent people would want to buy insurance to provide custodial care, in conditions of dignity and adequate comfort, if they became demented; providing for such care would be much less expensive than providing for life-saving treatment—for example, renal dialysis or an organ transplant—if it were needed.5 Now consider a somewhat more controversial suggestion. In most developed countries, a major fraction of medical expense—over a quarter of Medicare payments in the United States, for example—is spent on people in the last six months of their lives. Of course, doctors do not always know whether a particular patient will die within a few months no matter how much is spent on his care. But in many cases, sadly, they do know that he will. Most young people on reflection would not think it prudent to buy insurance that could keep them alive, by expensive medical intervention, for four or five more months at the most if they had already lived into old age. They would think it wiser to spend what that insurance would cost on better health care earlier, or on education or training or investment that would provide greater benefit or more important security. Of course, most people would want to live those additional months if they did fall ill: most people want to remain alive as long as possible, provided they remain conscious and alert and the pain is not too great. But prudent people would nevertheless not want to guarantee those additional months at the cost of sacrifices in their earlier, vigorous life, although, once again, they would certainly want insurance to provide the much less expensive care that would keep them as comfortable and as free of pain as possible. We can use these assumptions about what most people would think prudent for themselves, under fairer conditions than those we now have, as guides to the health care that justice demands everyone have now. If most prudent people would buy a certain level of medical coverage in a free market if they had average means—if nearly everyone would buy insurance covering ordinary medical care, hospitalization when necessary, prenatal and pediatric care, and regular checkups and other preventative medicine, for example—then the unfairness of our society is almost certainly the reason some people do not have such coverage now. A universal health-care system should make sure, in all justice, that everyone does have it. On the other hand, if even under fair conditions very few prudent people would want to insure

Dworkin: Justice and the High Cost of Health

themselves to a much higher level of coverage—if, as I said, very few people would insure to provide lifesaving care when demented or heroic and expensive treatment that could prolong their lives only by a few months, for instance—then it is a disservice to justice to force everyone to have such insurance through a mandatory scheme. Of course, any judgment about what most prudent people would do is subject to exceptions: some people have special preferences and would make very different decisions from those that many other people would. Some people might think, even on reflection, that guaranteeing a few extra months of life at the end was worth great sacrifice earlier, for example.° But it seems fair to construct a mandatory coverage scheme on the basis of assumptions about what all but a small number of people would think appropriate, allowing those few who would be willing to spend more on special care to do so, if they can afford it, through supplemental insurance. If we substituted the prudent insurance approach for the rescue principle as our abstract ideal of justice in health care, we would therefore accept certain limits on universal coverage, and we would accept these not as compromises with justice but as required by it. Expensive treatment for unconscious or demented or terminally ill patients would be relatively easy cases to decide if we adopted that approach. Other decisions would be more difficult to make, including, for example, heart-wrenching decisions about the care of babies born so deformed or diseased that they are unlikely to live more than a few weeks even with the most heroic and expensive medical intervention. A few years ago doctors in Philadelphia separated newborn Siamese twins who shared a single heart, though the operation would certainly kill one baby and give the other only a one-in-a-hundred chance of surviving for long, and though the total cost was estimated to be a million dollars. (The twins' parents had no medical insurance, but Indiana, where they lived, paid $1,000 a day toward the cost, and the Philadelphia hospital absorbed the rest.) The chief surgeon justified the procedure by appealing to the rescue principle: "There has been a unanimous consensus," he said, "that if it is possible to save one life, then it is worth doing this." But the different standard I am defending would probably have recommended against the operation. Suppose people of average wealth, when they marry, are offered the opportunity to buy one of two insurance policies: the first provides that if any of their children is born with a life-threatening defect, neonatal treatment will be covered only if it offers a reasonable (say, 25 percent) chance of success, and the second—much more expensive—provides that such treatment will be

guaranteed even if it offers only the barest hope. Most potential parents would decide, I believe, that it would be better for them and their families to buy the first policy and to use the premiums they would save each year to benefit their healthy children in other ways—to provide better routine medical care, or better housing, or better education, for example—even though they would be giving up the chance for a desperate gamble to save a defective child if they ever had one. Any public body charged with overseeing the distribution of health care would have to decide what medical procedures are "necessary and appropriate" and thus should be part of a comprehensive package of benefits everyone is guaranteed. Some of these decisions would be particularly difficult: deciding when very expensive diagnostic techniques or experimental organ transplants with a low chance of success are appropriate, for example. Such decisions must of course be based on the best and latest medical evidence and must constantly be reviewed as that evidence changes. But they, too, should be guided by the standard of individual prudence: Would it make sense for someone to insure himself when young to guarantee a vastly expensive blood test which would improve the diagnosis of a heart attack by a very small percentage of accuracy if he should ever have doubtful symptoms of cardiac disease? Or to provide a risky, expensive, and probably ineffective bowel and liver transplant if doctors decided it would give him a small chance to live? The rescue principle insists that society provide such treatment whenever there is any chance, however remote, that it will save a life. The prudent insurance principle balances the anticipated value of medical treatment against other goods and risks: it supposes that people might think they lead better lives overall when they invest less in doubtful medicine and more in making life successful or enjoyable or in protecting themselves against other risks, including economic ones, that might also blight their lives. An agency might well decide that while prudent people would provide their family with the prenatal and well-child care that so many Americans lack and would insure against serious medical risks at all stages of their lives by providing tested and reasonably effective treatment should they need it, they would forgo heroic treatment of improbable value if they needed it in return for more certain benefits like education, housing, and economic security. If so, then justice demands that a universal health scheme not provide such treatment. In summary, the prudent insurance test helps to answer both questions of justice I mentioned at the beginning How much should America spend overall on its health care, and how should that health care

I 1 311

M MI

lIlli1

7 Delivery of Health Care and Resource Allocation

be distributed among its citizens? The test asks what people would decide to spend on their own medical care, as individuals, if they were buying insurance under fair free-market conditions, and it insists, first, that we as a nation should spend what individuals would spend, collectively, under those conditions; and, second, that we should use that aggregate expenditure to make sure that all have now, as individuals, what they would have then. Of course some of the decisions I have been discussing would be made differently by different people trying to apply the prudent insurance test. It is very important that any agency charged with those decisions should be made up of representatives of different groups that might be expected to make such judgments differently; it should have doctors and health-care specialists, of course, but it should also have ordinary people of various ages drawn from different parts of the country and, if possible, different ways of life. Such an agency could draw on the experience of countries with "single-payer" governmentrun health services which have had systematically to ration health care. In Britain, for example, doctors in the national health system have been forced to allocate scarce resources like renal dialysis machines and organs for transplant, and they have worked out informal guidelines that take into account a potential recipient's age, general health, life quality, and prospects, as well as prospects for adequate care by family or friends. Though this supposed cost-benefit test is different from the prudent insurance test, the decisions doctors have made under the former presumably reflect their judgments, guided by experience, about the relative value of different kinds of treatment at different ages and in

different circumstances, and these are also judgments that a prudent insurer would be required to make. The prudent insurance test also makes plain why it is so important to consult public opinion before rationing decisions are made. Since rationing should reflect not just technical cost-benefit calculations but also the public's sense of priorities, consultation is essential. When Oregon sought to establish priorities in health care under Medicaid, it organized a series of "town meetings" and a "parliament" to discuss the matter, and though the meetings were criticized by some because they were attended by very few of the poor whose health .care was being debated, the meetings were nevertheless valuable sources of information about what those who did participate thought would be prudent insurance decisions. Still, no matter how much information an agency seeking to apply the prudent insurance test is able to gather, its results must be provisional, open to revision on the basis of further evidence of public preference as well as of medical technology and experience. Clinton's failed health-care plan would have allowed people covered by the plan to purchase supplemental health insurance at market prices, with no tax deduction or subsidy. That provision would have fitted the prudent insurance approach particularly well.' If, after an agency has established a basic coverage package, a very substantial number of people of average income buy supplemental insurance, in spite of its expense, the basic package should be expanded. If most men of average wealth bought supplemental coverage providing yearly prostate examinations beginning at a younger age than the basic package specified, for example, the prudent insurer test would require that the age specified in the basic package be lowered. F. . .1

Notes

E. . .1 1. 1 am ignoring a n important issue that my argument raises but that I will not pursue here. Is it right, in the hypothetical exercise I am ccnstructing, to exclude information relating risk of disease to voluntarily chosen behavior? Should insurance companies be in a position to charge cigarette smokers cr mountain climbers higher premiums, for example? That seems reasonable. But if so, what counts as voluntary behavior? Should sexual behavior of a particular kind be treated as voluntary for this purpose? It would seem wrong for insurance companies to charge active male homosexuals higher premiums because they are considered more likely to contract AIDS. Is this because sexual preference is less under people's control than nicotine addiction? Or because

2.

3.

the sacrifice in giving up sex is so much greater than that in giving up smoking? It does seem likely that even though the members of the imagined community would begin by making individual insurance decisions, they would soon develop, through these individual decisions, collective institutions and arrangements like cooperative insurance purchasing agencies or pools, because these would provide economic advantages in a free market among people of roughly equal wealth. The result of the process might very well be something functionally very close to that proposed in Clinton's plan. My claim needs minor qualification. Even in the imagined community, some paternalistic interference might be necessary to protect people from imprudent insurance

Dworkin: Justice and the High Cost of Health

4.

5. 6.

decisions, particularly when they are young. And some constraints might be necessary to provide adequate resources for later generations. That is different from asking what any particular twentyfive-year-old would in fact do, because many people, particularly when young, do not make prudent decisions. They do not, that is, make the decisions that best serve the plans, convictions, tastes, and preferences they would find, on reflection, that they already have. It is prudent, of course; to provide for change—any prudent long-term insurance policy is written in general clauses rather than precise details of treatment and is open to revision year by year. See Ronald Dworkin, life's Dominion (New York Alfred A. Knopf, 1993), chap. 9. People with certain religious convictions might make that choice, for example. But it is worth noticing that even those Catholics and others who think that it is always wrong to refuse available life-prolonging treatment do not necessarily think that all possible sacrifices must be made, in advance, to insure that very expensive life-prolonging treatment is available. Someone who thought it would

Why Undervaluing "Statistical" People Costs Lives Tony Hope

Whether happiness be or be not the end to which morality should be referred—that it should be referred to an end of some sort, and not left in the dominion of vague feeling or inexplicable internal conviction, that it be made a matter of reason and calculation, and not merely of sentiment, is essential to the very idea of moral philosophy . . . J.S. Mill, London and Westminster Review, 1838 The Cash Value of Life

In January 1997 Tony Bullimore was attempting to sail round the world in the Vendee Globe race. He had reached the dangerous and cold waters of the Southern Ocean, over 1,500 miles south of the Australian coast, when his boat was capsized by hurricane force winds and enormous waves. He spent four days trapped under its hull before he was rescued in the largest and most expensive such operation ever undertaken by the Australian defence forces. How much money should a civilized society be prepared to spend in order to save

7.

be wrong, as a matter of religious principle, to decline an expensive and arduous operation that he could afford in order to extend his life a few months might nevertheless, with perfect consistency, think it prudent not to pay for the insurance, over his lifetime, that would enable him to afford it. He might think it more sensible to use the money that such insurance would cost to provide better medical treatment for himself or his family earlier, or better education, or some other good or opportunities his convictions also deemed important. If so, then the prudent insurance test offers no reason why a national health scheme should make available for him what he would not have insured to provide for himself. The rescue principle might hold it unjust for anyone to buy better medical care than everyone can have and that medical care outside a system of universal coverage should therefore be abolished or, as it is in Canada, seriously discouraged. But the prudent insurance approach begins in a different idea: that no one can complain, on grounds of justice, that he has less of something than someone else does so long as he has all he would have if society were overall just.

a life? Is the answer "whatever it takes," or should there be a limit? When is the chance of success too low even to attempt a costly rescue operation? Let me pose a more general question. What is the cash value of a human life? This question is a disturbing one to ask but, paradoxically, there are situations where avoiding the question may cost lives, and allocating scarce medical resources is one of them. There is no health care system in the world that has sufficient money to provide the best possible treatment for all patients in all situations, not even those that spend relatively large sums on health care. New and better treatments are being developed all the time. On average, in the UK, about three new medicines are licensed each month. Almost all have some benefit over existing treatments and some will extend people's lives. Many of these new medicines are expensive. When is the extra benefit worth the extra cost? This question must be asked by all health care systems, whether private systems, such as "managed care" in the US, or publicly funded systems, such as the British National Health Service. If the best treatment cannot always be provided, then choices have to be made. The general question of how our limited health care resources should be distributed is one of the most important in medical ethics. The quality and quantity of thousands of people's lives will be affected by the answers that we give.

Source From Medical Ethics: A Very Short Introduction, pp. 26-41 (Oxford University Press, 2004). Reprinted with permission.

II AI

1111.1i

7 Delivery of Health Care and Resource Allocation

Quality of Life Some medical treatments have little or no effect on lifespan but improve quality of life: hip replacement for osteoarthritis is an example. One rather deep problem that faces us in thinking about the right way to distribute health resources is how we compare and evaluate the relative importance of improving quality of life vis-à-vis extending it. I am not going to tackle this issue, nor the problems associated with the measurements of quality of life in the first place. I will focus exclusively on lifeextending treatments since there are more than enough problems in thinking about allocating resources to these treatments alone. There are many examples of lifeextending treatments. Surgery for appendicitis extends life because without such surgery most people would die. Breast cancer screening can extend life because early detection and treatment can increase life-span. High blood pressure increases the risk of death from heart attack and stroke. Treatment that lowers blood pressure reduces, although it does not eliminate, this risk. Renal dialysis keeps those people alive whose kidneys no longer function adequately. Each year of dialysis is a year more life.

In Control of a Budget Imagine that you are in charge of a health service for a particular population. You have a limited budget—you cannot afford the best treatment for all of the people all of the time. You have decided how to spend most of your budget, and you have a few hundred thousand pounds left uncommitted. You sit down with your advisers to consider the best way of spending this last remaining tranche of money. There are three possibilities, and you must choose one of them. The possibilities are: 1) a new treatment for bowel cancer that gives the relevant patients a small but significant chance of increased life-expectancy; 2) a new drug that lowers the chance of death from heart attack in people with genetically induced raised blood cholesterol; 3) a new piece of surgical kit that ensures a lower mortality from a particularly difficult kind of brain surgery. On what basis do you choose between these possibilities? One approach that has a lot going for it is to say. there is no good reason to prefer one person's year of life over another person's, or to give any priority to people who would benefit from the bowel cancer treatment over people who have the genetically induced high blood

cholesterol or to people with the brain tumour. In each case people stand to die prematurely and in each case the treatment increases the chance that they will live for longer. What we should do, therefore, is to spend the money so that we can "buy" as many life years as possible. By doing this we are treating everyone fairly we are valuing one year of life equally, regardless of whose life it is.

The Distribution Problem Even amongst people (like me) who are attracted by this approach, there is an issue that needs to be faced: the "distribution problem." Take a look at the three interventions described in Table 1. Suppose that all these interventions cost the same and that we can only afford one of them. Suppose further that the distributions are as follows. The two people who are benefited by intervention 3 will enjoy 8 more years of life each. Of the ten people who are benefited by intervention 1, the average benefit is 3.5 years, and the range is 2-4 years. Of the fifteen people who are benefited by intervention 2, the average benefit is 2 years, and the range is 1 to 3 years. Which of the three interventions should we go for? If we think that what we should do is to "buy" the maximum number of life years that we can (the maximization view) then we should put our money into intervention 1 because we buy 35 life years, which is more than we will get if we spend the money on either of the other two interventions. Some might argue that intervention 2 is preferable because we help more people (15 as opposed to 10) although each person gains fewer extra years of life. Still others might argue that intervention 3 is the best option because the two people who are helped receive a really significant gain (eight years of life) whereas no one gains more than four years of life with either of the other two options. The question of whether it is only the total number of life years that matters, or whether the way in which those years are distributed between people is important, is known as "the distribution problem." Those who reject the maximization view have to specify how they balance the value in helping more people, but each gaining relatively less, against the value in helping fewer people, but each gaining relatively more. Except at extremes I am generally happy to go with maximizing the total number of life years and not worry too much about their distribution.

Table 1 Choosing between Three Interventions Intervention 1 benefits 10 people total life years gained: 35 Intervention 2 benefits 15 people total life yeas gained: 30 Intervention 3 benefits 2 people total life years gained: 16

Hope: Why Undervaluing "Statistical" People Costs Lives

In being generally happy with using resources to maximize total number of life years I am in a minority—and no health care system in the world behaves remotely in this way. One problem with my position (the maximization view) takes us right back to Tony Bullimore and his attempt to sail round the world. My position gives no moral weight to what has been called "The Rule of Rescue"—and yet this rule seems, intuitively to be right. The Rule of Rescue

The "rule of rescue" is relevant to a situation where there is an identified person whose life is at high risk. There exists an intervention ("rescue") which has a good chance of saving the person's life. The value that is at the heart of "the rule of rescue" is this: that it is normally justified to spend more per life year gained in this situation than in situations where we cannot identify who has been helped. Consider two hypothetical, but realistic, situations in health care. Intervention A (Saves Anonymous "Statistical" Lives)

A is a drug which will change the chance of death by a small amount in a large number of people. For example, out of every 2,000 people in the relevant group, if A is not given, then 100 people will die over the next few years. If A is given, then only 98 will die. Although we know that drug A will prevent deaths, we do not know which specific lives will be saved. Drug A is cheap—the cost per life year gained is £20,000. One example of a medical treatment like this is treatment that lowers moderately raised blood pressure. Another example is a class of medicines known as statins that lower blood cholesterol. Lowering blood pressure, and lowering cholesterol, reduce risk of heart attack, stroke, and death. Intervention B (Rescues an Identified Person)

B is the only effective treatment for an otherwise lifethreatening condition. Those with the condition face a greater than 90 per cent chance of death over the next year if not given B. If given B, then there is a good chance of cure—say 90 per cent. B is expensive. The cost per life year gained is £50,000. Renal (kidney) dialysis is an example of this type. There are three, potentially relevant, differences between intervention A and intervention B. The first is that B saves lives within the next year, whereas the benefits of A are not realized for many years. This difference has some moral relevance. A few of those who might benefit from intervention A will die from

some quite independent cause before any benefit from A could be gained. There are also problems in calculating the cost per life year gained when at least some of the costs of the intervention are borne years before the benefits are seen. This is because of monetary inflation. Both these effects can be allowed for in the calculation of cost per life year gained. Having made such allowances, there seems no good reason to value the saving of life years in the future any less than saving life years now. The second difference between the interventions is that B will almost certainly save the lives of the relevant patients, but A only has a low probability of doing so. Thus B might be seen as giving greater benefit to individuals than A. I will argue, in a moment, that this is false. The third difference is that intervention B benefits identifiable people. Intervention A benefits a proportion of patients within a group (e.g., those with raised blood pressure), but we cannot know who within the group will benefit (although we may know the likely proportion that will benefit). According to the rule of rescue it may be right for a health care system to fund intervention B but not intervention A, even though B is more expensive in terms of life years gained. For example, the rule of rescue would provide justification for spending more per life year gained on treatments such as renal replacement therapy than on treatments like statins. In practice this is exactly what health care systems do. The British National Health Service provides renal dialysis at costs over £50,000 per life year gained, whilst paying for statins only for those with very high cholesterol levels. This is despite the fact that treatment with statins for those with moderately raised cholesterol levels would cost only about £10,000 per life year gained. In other words, if the money spent on some people for renal dialysis were, instead, spent on some people with moderately raised cholesterol, five times as many life years could be gained. But we don't do it—because we would feel that we had condemned the person needing dialysis to death, whereas all we would be doing in the case of statins is slightly lowering an already quite small chance of death. The most powerful reason in support of the rule of rescue is that in the typical case, the identified person, like Tony Bullimore, stands to gain a significant increase in chance of life, whereas in the typical case of saving anonymous "statistical" lives no one stands to gain more than a small decrease in probability of death. I will put this argument in favour of the rule of rescue as strongly as I can. I will then say why I do not agree with it.

•

II

'SI

1111.13

7 Delivery of Health Care and Resource Allocation

The Strongest Argument in Favour of the Rule of Rescue

Premature death is, normally, a very significant harm indeed. But a very small chance of premature death is by no means a great harm—and we cannot claim that we need something which reduces by a very small amount the chance of premature death. All of us in our lives trade small increases ir. the chance of premature death against really quite small benefits. Consider the Sunday morning cyclist.

(because we had moderately raised cholesterol levels) it would be reasonable, and not extraordinarily altruistic, for us to prefer that the money go not to provide us with statins but towards the cost of renal dialysis for someone who would otherwise die. From the point of view of those who have to decide how limited health care resources should be distributed, it certainly seems better to keep a few people alive (who would otherwise certainly die) than to reduce only slightly the chance of death of a large number of people, particularly when the risk of premature death is fairly low anyway.

The Sunday Morning Cyclist

Back to the Distribution Problem

On Sunday mornings I cycle along the busy Banbury Road in my home town of Oxford to buy a newspaper. In doing this I am putting myself at (what I hope is) only a small extra risk of premature death. I am trading this extra risk against the pleasure and value of reading the Sunday morning paper. In balancing these two I find that the pleasure of the paper—a really rather small pleasure in my life—outweighs the extra risk of premature death. There seems nothing irrational in this. A very small chance of a terrible harm is itself only a small negative weight easily outweighed by other benefits. Most of us will take these small risks not only for our own benefit but for the benefit of others. Consider the friend's job application.

The rule of rescue seems to be a particular example of the distribution problem. Most people reject maximizing life years gained (which would favour paying for statin treatment). Essentially, the intuitive appeal is as follows: it is better to provide a great benefit (continuing life in people who would otherwise certainly die) to a few people than a trivial benefit (a small reduction in chance of premature death) in a large number of people.

The Friend's Job Application

Suppose that a friend is applying for a job which he is keen to get. To meet the application deadline it has to be in the postbox today. Owing to a severe bout of influenza, my friend cannot post it himself. To help him I cycle to his house to collect the application and post it. Again, this action increases by a very small amount my chance of premature death. This is easily outweighed by the value of helping my friend. With these considerations in mind I will propose an argument in favour of a health care system paying for a "rescue" intervention of type B (at, for example, a cost of £50,000 per life year gained) whilst refusing to pay for an anonymous "statistical" intervention of type A (at, for example, a cost of only £20,000 per life year gained). I will make the argument using the cholesterol-lowering drugs (statins) as an example of the anonymous "statistical" intervention and renal dialysis as an example of the rescue intervention. Those who would benefit from treatment with statins gain very little—a very small reduction in the risk of premature death. The "friend's job application" shows that we readily risk small changes in the chance of premature death, even for benefits to other people. If we ourselves stood to gain from the statins treatment

Why I Disagree with the Rule of Rescue

Despite the strong intuitive appeal of the rule of rescue, and the arguments that I have outlined in favour of it, I stick by my preference for maximizing benefit. I will argue for my position by considering a counter-example to this conclusion: the case of the trapped miner. The Case of the Trapped Miner

Consider the case of the trapped miner (see box). Suppose that the facts are these (perhaps not entirely realistic). There is a small risk of death to those in the rescue party, and this risk varies according to the size of the rescue party. If there were 100 rescuers, there would be a 1:1,000 chance for each rescuer of death. If there were 1,000 rescuers, each would face a 1:2,000 chance of death. If 10,000 rescuers, then each would face a 1:5,000 chance of death. If 100,000 rescuers (an extraordinarily large rescue party—but this is a "thought experiment" to test a theoretical point)—then each would face a 1:10,000 risk. Thus, the larger the size of the rescue party, the smaller the risk of death faced by each individual rescuer. It is also the case, however, that the larger the size of the rescue party, the more people are likely to die in the rescue attempt. With a rescue party of 100,000, each member of the rescue party faces a very small risk of death—well within the risks that we normally take for much less important gains than saving a life. However, with such a rescue party, about ten people are likely to die in order to save the life of the one trapped miner.

Hope: Why Undervaluing 'Statistical" People Costs Lives

The Case of the Trapped Miner A miner lies trapped following an accident. Without rescue he will die. Given a sufficiently large rescue party, the miner can be saved. Take a moment to consider the following questions: 1. Do you think you should join the rescue party if you faced a 1:10,000 risk of death in so doing? 2. Is there any further key information you need to know before you can answer the first question?

will be lost in rescuing the one miner. I would be much happier (from the moral point of view) volunteering for the first rescue party than the second.

If we assume that most people are altruistic at least to a small extent, and most people will accept a very small level of risk of personal death in order to save another's life, and if we assume, further, that most people, given the choice, would like to face as low a personal risk of death as possible, then respecting the wishes of each potential member of the rescue party would have the following result. The wishes of potential members of the rescue party would be most respected by putting together an enormous rescue party in order to save the trapped miner—at the expense of many lives. Thus, if the issue of rescue is seen simply as a question of balancing individual risks for each rescuer against the benefit to the individual of being rescued, then it would seem right to pursue a policy which overall was very costly in terms of lives lost. Suppose that a senior army officer will lead the rescue. If that army officer were to coordinate the rescue, with the foreseeable result that more people would die in the attempt to rescue than would be saved by the rescue, then the army officer might reasonably be criticized, even if the rescue party were made up entirely of volunteers who knew and accepted the risk to themselves. He would have been responsible for a rescue operation that caused, and had been expected to cause, more deaths amongst the rescuers than the number of people who were rescued. Leading such a rescue even with fully informed volunteers is highly problematic from a moral point of view.

Let us reconsider statins and renal dialysis. It is not clear that those who could benefit from the anonymous "statistical" intervention (e.g., statins) have voluntarily agreed to forgo their treatment in order for identifiable patients to receive expensive life-extending treatment. A health care system that spends more per year of life gained on rescue treatments (such as renal dialysis) than on "statistical" treatments is effectively volunteering those who would benefit from the preventive treatment to take part in a "rescue party" for those requiring the rescue treatment. Because of limited resources, any health care system, in making decisions about treatments which extend people's lives, has to extend some people's lives at the expense of other people's lives. In the absence of a clear mandate from the group of people who stand to lose by a particular decision, it seems to me that the core principle must be that those decisions should be taken which overall maximize the number of life years gained. And even if there were such a clear mandate (which there is not) it remains questionable, as with the army officer leading the rescue operation with fully informed volunteers, whether it would be right for a health care system to let more die to save fewer.

Further Key Information

A Counter-intuitive Conclusion

Let me return to the second question I asked about the case of the trapped miner is there any further key information you need before answering the first question? I think you should know not only your personal risk in joining the rescue party but also the size of the rescue party. Because if the rescue party needs only ten people and each member has a risk of 1:10,000 of dying, then the miner's life will be saved with (almost certainly) no loss of life. But if the rescue party needs to be 100,000 strong, then almost certainly many lives

But can we accept this conclusion? Let's go back to Tony Bullimore and the dramatic and successful rescue undertaken by the Australian defence forces. Only a stone-hearted theorist could read Bullimore's account and conclude that it was wrong to mount such a rescue. The Australian defence forces were right to spend millions of tax-payers' dollars. In the same way, it is right for a society to spend £50,000 a year to keep a patient alive on renal dialysis. How could we stand by and say to a patient: we could keep you alive for many years,

Back to Health Care

II

5111

111111

7 Delivery of Health Care and Resource Allocation

but we will not provide the necessary money—we have other priorities. And how could we say this to the relatives who would be bereaved? This seems very different from the situation of the patient with moderately raised cholesterol. Without treatment the chances are that the person will not have a heart attack and die. By refusing the treatment we are not condemning him to death as we are the person who needs renal dialysis. But the logic of the case of the trapped miner refutes this. It is true that if we do not provide treatment for the raised cholesterol we will not know which specific people die as a result of lack of treatment, nor which relatives have been bereaved. But we do know that there are such people. Enlarging Our Moral Imagination

So how do we square the circle? What do we learn from our empathy with Tony Bullimore or a person with renal

QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference Do They Make for Health Care Priority Setting? Bjarne Robberstad

Introduction

The total population of Africa is some 655 million people, who experience an annual loss of about 358 million disability adjusted life years (DALYs). This is almost half a DALY per capita per year, with HIV/ADS, childhood diarrhoea, measles, malaria, and respiratory infections among the most important causes of disease burden.' The life expectancy in the region is generally short, with 47 years in Tanzania and 37 years in Malawi as examples.' Effective treatment or prevention is available for most of the above conditions,' but health care budgets that are extremely scarce represent effective barriers against improved population health. Generally for the region, health care budgets are in the range 3-20 USD per capita per year, while for e.g., Tanzania the health budget in 2002 was 7 USD per capita3 To put this into perspective, the public health care expenditures in Norway were 3366 USD per capita last year,3 almost 500 times higher.

failure? The answer, I think, is not that we should become stone-hearted logicians and refuse to attempt the rescue of Bullimore or to provide renal dialysis. It is right that our moral imagination and our human sympathy are awakened. What we should learn from the logic of the case of the trapped miner is that our moral imagination must also be awake to the sadness of lives cut short, and relatives bereaved, because we did not provide treatment for moderately raised cholesterol. Deaths are not less significant because we cannot put a face or a name to the person whose life could have been saved. Health care is good value for money. The lesson we should learn from our empathy for those in need of rescue is to widen our moral imaginations. We rightly respond to the person in distress by being prepared to spend money to save a life. We should respond in the same way to prevent "statistical" deaths, for such deaths are real people and the friends and relatives who are left behind mourn in just the same way.

In this situation, with health challenges that are immense and budgets that are extremely scarce, giving priority to the "wrong" programs may have large implications for health system performance. The opportunity cost of wasting 10 USD in a country like Tanzania is more than one year of human life. An example is immunisation against hepatitis B, which is not routinely done in a majority of sub-Saharan African countries.4 This intervention has been demonstrated to cost as little as 8 USD per life year saved.5. Putting weight on cost-effectiveness seems like a good idea in this situation, because when resources are very scarce it seems reasonable to make the most of it. Economic evaluation, like cost-effectiveness analysis (CEA), has therefore been assigned a key role since the World Development Report 1993.6 For CEA to serve a meaningful role in comparing the relative value of different health interventions, it is necessary to apply a commonly agreed measure of population health. The objectives of this paper are therefore to present the major differences between three commonly applied health measures: life years gained Os gained), quality adjusted life years (QALYs), and disability adjusted life years (DALYs). Furthermore, it will be explored what difference these alternative measures may have for health care priority setting in sub-Saharan Africa.

Source: From Norsk Epidemiologi 1` -•, 2 (2005), 183-191. Reprinted with permission.

Robberstad: QALYs vs DALYs vs LYs Gained

Life Years Gained

Traditionally, the impact of health care has been measured in terms of its effect on mortality,' for example, deaths averted. A potential drawback of using deaths averted to measure health effects is that the method doesn't distinguish between children, adults and the elderly. This is not unproblematic since most people, if forced to make a choice, would prefer to save e.g., a 5-year-old infant rather than a person aged 95. Life years gained is a modified mortality measure where remaining life expectancy is taken into account. This method accrues more weight to young target populations, because saving the life of an infant yields more life years than saving the life of an old person. Life years are calculated as the remaining life expectancy at the point of each averted death. Life expectancies may be taken from life tables that are specific for each setting or standardized across larger regions. The choice of life table is not uncontroversial, as a life table with high life expectancies will yield more life years and render interventions more attractive than life tables with shorter life expectancies. Discounting of Health Effects

LYs gained can be discounted, to adjust for time differences, or presented undiscounted. Common practice is to discount health outcomes at a rate between 3 and 5%." Discounting accrues less weight to future than to immediate health outcomes, which is illustrated in Figure 1. With a 3% discount rate, for example, half the weight is accrued to outcomes 23 years into the future compared to immediate outcomes, while a 5% discount rate reduces the weight to half in about 14 years. There is a relatively extensive amount of literature discussing whether or not the practice of discounting future health effects is appropriate. For example, discounting has been criticized for discriminating future generations and for being immoral.1° Others claim that for the sake of consistency, discounting of health as well as costs is necessary."2 This debate will not be pursued further in this paper, because discounting is not what makes LYs, QALYs and DALYs different. Nevertheless, the choice of discount rate is important because it may determine the mix between typical curative interventions, like improved case management of malaria, and typical preventive interventions, like hepatitis B vaccination. Advantages and Disadvantages of LY Gained

LYs saved is as indicated a relatively easy and transparent method for measuring population health, and there

are few value choices involved. The perhaps most important disadvantage is that the method ignores the obvious fact that health is more than merely staying alive. The method will fail to acknowledge health improvements such as improved physical ability, reduced neuropsychological stress, and reduced chronic pain. This makes it difficult to compare outcomes of health interventions that target e.g., chronic conditions with interventions targeting fatal conditions. Furthermore, if only the mortality aspect is quantified and included in the denominator of the cost-effectiveness ratio, LYs gained represent an intrinsic bias against conditions that are largely non-fatal like for example poliomyelitis, onchocercosis (river blindness) and schizophrenia. Taking Morbidity into Account It is non-controversial that it is necessary to say some-

thing about the majority of people who do not suffer premature death.° Albert Schweitzer once said that "Pain is a more terrible lord of mankind than even death itself."14 It is therefore, at one level or another, necessary to take people's morbidity into account. This recognition has led to the development of measures that simultaneously seek to capture gains from reduced morbidity and reduced mortality The two dominating such summary measures are the Quality Adjusted Life Years (QALYs) and the Disability Adjusted Life Years (DALYs). QALYs and DALYs represent an implicit tradeoff between quantity for quality of well-being. Such trade-offs are well known from many aspects of life, but many are unfamiliar with the concept of trading off quality for quantity of years of life and health. William Somerset Maugham, in the novel "As a Lily in the Field," describes a young banker in London (Thomas Wilson) who found the prospective of doing the same kind of work, day by day, for the rest of his life rather depressing. Thomas Wilson once went on a holiday to Capri in Italy, and later he kept daydreaming about the swimming, the wine yards, the hillsides in moonlight, the sea and piazza in the evening. At the age of 35 he quits his job, sells his house and uses all his assets to invest in a life pension. The life pension will provide him with sufficient means to rent a house at Capri, employ a servant to tend for him, enough for tobacco, good food, and a book from time to time for a period of 25 years. Rather than living, in his view, a miserable life as a banker for the remaining life time, he decides to live like "a lily in the field" for 25 years, till he is 60, and then put an end to it. In other words, Thomas Wilson makes a plan that enables him to trade-off a long and miserable life for a shorter but qualitatively superior life.

II MN

384

7 Delivery of Health Care and Resource Allocation

1.0

0.8

0%

a6

0.4

10

20

30

40

50

Years from now

Figure 1 The Effect of Discounting on the Weight of Future Health Effects

Quality Adjusted Life Years (QALYs) The story of Thomas Wilson brings us to the QALY, which was first introduced by Klarman et al.,15 although the concept is more commonly associated with an applied paper by Weinstein and Stason.16 In QALYs, premature death is combined with morbidity by attaching a weight to each health state such that value 0 represents death, while value 1 represents full health. The number of QALYs for a health profile is found by multiplying the health related quality of life weight (HRQof..) of the health state with the duration of the health state. Figure 2 illustrates a person who gets some disability at the age of 10, lives with the condition for 35 years, and suffers premature death at the age of 45. If the life expectancy is 60 years, and the health related quality of life weight associated with the condition is 0.75, the (undiscounted) lifetime QALYs of this person are 1.0 x 10 (QALYs before onset of disease) + 0.75 x 35 (QALYs during disease) = 36.25. The QALY loss would be 1.0 x 60 (QALYs in case of full health for full life expectancy) — 36.25 = 23.75. Like the LY measure, QALYs can be presented discounted or undiscounted, although the former is most common. Disability Adjusted Life Years (DALYs) The DALY concept was introduced by the World Bank in the World Health Report 1993.6 Two important propositions are underlying the DALY concept, namely

(i) that the burden calculated for like health outcomes should be the same, and (ii) that the non-health characteristics that should be considered should be restricted to age and sex.14 Other characteristics, like income, education, ethnicity, etc., should in other words not be taken into account. These propositions represent intentions of creating a methodology that treat people as equally as possible. However, some would say that more weight should be put on the diseases of the disadvantaged, and that failing to do so is inequitable. Anand and Hanson, for example, claim that disease weights should take account of the way in which individual and social resources can compensate for the level of disability experienced.17 For example, being blind in Niger is worse than being blind in the UK, because the structural interventions in the UK make the disability less severe than in Niger.18 Blindness, according to this view, should therefore be weighted more heavily in Niger than in the UK. Like the QALY, the DALY measure facilitates comparisons of all types of health outcomes by attaching disease weights where value 0 represents full health and value 1 represents death. Note that these disease weights are the opposite of the HRQ0L weights in the QALY. A DALY can therefore be seen as an inverse QALY. Figure 2 illustrates that while QALYs are a good, DALYs are something one seeks to avert. In the following section the statement that a DALY is an inverse QALY will be somewhat modified because of different weighting and elicitation techniques.

Robberstad: QALYs vs DALYs vs LYs Gained 0.0

Health-related qu

3461am as easia

DALYs

QALYs

0.0

Death 45 years

Onset of disability 10 years

1.0 LE 60 years

Age

Figure 2 Relationship between QALYs and DALYs When the DALYs Are Not Age Weighted

Weighting of Health Outcomes

In the above section it has been shown that the different health measures LYs, QALYs and DALY5 are weighted in different ways. They are all adjusted for the life expectancy of people affected, giving more weight to the young. Furthermore, they are all usually discounted, giving more weight to immediate over distant outcomes. However, only DALY5 and QALYs are weighted to account for disease severity in order to summarise mortality and morbidity The following section will partly be devoted to explaining how such weights can be elicited. Furthermore, DALY5 are so-called age weighted, and this also calls for some explanation. Finally, it has been proposed that such health measures should incorporate that extra weight or priority should be given to the most severely ill. This last concept has yet not been implemented in any of our outcome candidates in applied research, but the argument has some intuitive appeal, and at the end of this section some advantages and problems with such severity weighting will briefly be discussed. Elicitation of Weights

Several methods are being used to elicit health related quality of life or disability weights. The simplest approach is the so-called Visual Analogue Scale (VAS), where respondents are asked to rate health states on a scale. This scale often has the range from zero to one, where zero is the worst possible outcome (assumed to be death) and one is the best possible (perfect health). VAS is a simple method but has been strongly criticised for not being choice based, in other words, that it does not involve a trade-off between quantity for quality of years.

In the time trade-off (Tro) methodology people are asked to choose between two certain alternatives involving trade-off between quantity and quality of life. For example, they may be asked how many years of perfect health they would consider equally good as living 10 years with psoriasis. TTO questions are the most common technique for elicitation of HRQoL weights in QALYs, although VAS and standard gamble (see below) are also sometimes used. Person trade-off (PTo) is structurally similar to no, but here people are asked to trade off extending the lives of people with full health vs improving the health expectancy of people with some disability from suboptimal to perfect health. This method was used to elicit disease weights for the DALY methodology from a panel of health experts, while the HRQoL weights in QALYs are usually found by interviewing lay people and patients. In the standard gamble (SG) approach, respondents are asked to choose between the certainty of an intermediate health state and the uncertainty of a treatment with two possible outcomes, where one of the outcomes is more attractive than the certain outcome and the other is less attractive (e.g., death). The inclusion of uncertainty makes SG more consistent with standard economic utility theory than the other methods, but it is also probably the cognitively most demanding of the techniques for respondents to answer. Disease weights or HRQoL weights may easily influence priority decisions. It is therefore of interest whether or not the choice of elicitation technique is likely to influence the weights. From a theoretical point of view, one may expect that weights elicited with SG are higher than the VAS because of risk aversion. One may also expect TTO weights to be higher than VAS weights because of time preferences. These expectations were confirmed by Salomon and Murray, who in a sample of 69 individuals found that PTO weights were higher than SG, which again were higher than TTO, while weights produced by VAS were the lowest.19 In other words, VAS seemed to weight conditions as more severe than the other alternatives. While there was significant disagreement between the methods on the level of severity in this study, there was, however, considerable agreement in the rank orderings of the 12 conditions.19 Currently there is no agreement on which method should be preferred for eliciting disease or quality of life weights.19,2° Age Weighting

Age weighting is a concept unique to the DALY methodology and is intended to account for the fact that

I I JAI

Ai

386

mw

7 Delivery of Health Care and Resource Allocation

people are supported by others during infancy and at an advanced age but support others during adulthood. This notion is called welfare interdependence. Welfare interdependence does not imply that the time lived at different ages are more or less important to those individuals but that the social vi.ue is different.14 The age weighting of DALYs therefore accrue more weight to the life years of breast-feeders and caregivers and can be illustrated by a hump-shaped curve which starts at zero for newborns, peaks at 25, and gradually declines throughout adulthood without ever reaching zero. It can be argued that welfare interdependence is an arbitrary choice for an age-weighting function. In fact there exist several alternative ways to specify age weights depending on different perceptions on fairness. I will mention two alternatives to the DALY age weights. The first alternative is the so-called simple utilitarian ageism,21' 22 or horizontal age weights. Horizontal age weights accrue equal weight to all years and is what is being used in the traditional QALY measure and non-age-weighted DALYs. The last alternative is funded on the fair innings argument2' and is called egalitarian ageism.22,24 According to this argument everyone is entitled to a similarly long and healthy life, and the implication is that anyone failing to achieve this has been cheated, while anyone getting more is living on borrowed time. Z5 The elderly according to this argument, have already had their fair share of life and should be given gradually less weight in the distribution of scarce health care resources. These three alternative ageweighting functions are illustrated in Figure 3, but it should be noted that several sub-versions of the principles exist 21 Tsuchia has reviewed the empirical evidence on age weights and sums up that (i) there is no support for a horizontal age-weight profile (current QALY practice) and (ii) there is broad agreement that the profile declines after middle age, but (iii) there is some disagreement on whether middle age or childhood is the peak.22 Despite the agreement that the young should have some priority over the old, as Williams points out, the consensus is likely to evaporate when the quantitative issue is addressed as to how much priority the young should have.25 Equity Weighting

There is considerable empirical evidence suggesting that people in general are inequality averse in the sense that they hold the view that more weight should be allocated to those that are most severely ill. However, equity weighting is currently not standard procedure either in DALYs, or QALYs or LY1; gained. Introduction of

Simple Utilitarian Ageism

•_

Age Productivity Ageism

Age Egalitarian Ageism

Age

Figure 3 Three Alternative Age-Weighting Functions

equity-weighted QALYs has b y Alan Williams been proposed as ". . . the best way to integrate efficiency and equity in health care."25 And furthermore that equity weights ". . . rules out the giving of priority to things that do no good, which is the danger with equalization policies." Williams argued for a version of the fair innings, where the quality of a person's life is important as well as the length of it.25 There are two major barriers to implementing equity-weighted QALYs or DALYs. First, there exists no agreement as to over which notion of disease severity one should pursue equity. Is it, for example, acute illhealth, or is it lifetime ill-health?26 Secondly, empirical evidence including both estimation of equity weights and implementation of them in QALY maximization exercises is scarce. Bleichrodt and colleagues estimated equity weights in two Dutch populations. When they incorporated these weights in QALY calculations, they found that this in fact changed the ranking of 12 health interventions.27 These findings suggest that the issue of equity weighting needs further research, although it is likely that such an enterprise will be quite data hungry.25 What Difference Do They Make?

After having briefly presented some major differences between LYs gained, QALYs and DALYs, it is time to approach

Robberstad: QALYs vs DALYs vs LYs Gained

the second objective of this paper: What difference do they make for health care priority setting? I will first address the question by looking at a small selection of the literature in the area. For example, it has been argued that quality adjustment of life years makes little difference for priority setting. Others have argued that indeed it makes a difference, and moreover, that the difference is unfair or discriminating. Secondly, I will address the question by looking into two applied cases that I know well, namely, my own studies on the cost-effectiveness of zinc as adjunct therapy in childhood diarrhoea28 and on pharmaceutical preventive cardiology29

therefore increase the likelihood of implementation, all other things being equal. "DALY Weights Are Unfair"

There have been some concerns on whether or not the weighting of DALYs is fair. With the exception of the age weights, this critique also applies to QALYs. Anand and Hanson have claimed that "Using the DALY framework for resource allocation may lead to consequences that are at odds with principles of equity." They are worried that life-extension of disabled people avert fewer DALYs, and therefore count less, than life-extension of able-bodied people." Hypertension management "Quality Adjustments Make Little Difference" of people with polio would, for example, avert fewer Chapman and colleagues concluded that quality ad- DALYs than hypertension management of people withjustments make little difference for priority setting after out such a chronic condition because each life year having considered 63 CEAs reporting both costs per gained will be given less weight. The disability weight QALY and costs per LY gained. They therefore concluded of polio is 0.369.14 that in most cases findings can be reported as costs per This critique seems to be formally correct but inLY rather than the technically more challenging costs correctly applied. When DALYs, or QALYs, are used to per QAa.3° This conclusion is partly based on the find- estimate burden of disease, or in cost-effectiveness aning that quality adjustment of life years did not sub- alysis to inform resource allocation, no such distinction stantially change the estimated level cost-effectiveness. is made between disabled and able-bodied people. The More specifically, they found that switching from QALY cost-effectiveness of e.g., preventive cardiology is done to a pure mortality measure like as gained altered the assuming that people have no unrelated conditions, conclusion of whether or not to support implementa- like polio. The critique of Anand and Hanson therefore tion of health programs only in 18.5% of the cases.3° does not seem to apply to DALYs and QALYs the way This is of course interesting, but if one believes that they are used in economic evaluations. morbidity should be taken into account, it may also be Anand and Hanson are also critical of age weighting interpreted as a simple measure like LY gained leading and argue that there are no reasons for valuing time lived to the wrong conclusion in almost one-fifth of the cases. at different ages differently and that a principle of uniFurthermore, Chapman and colleagues investi- versalism would argue strongly for a common intrinsic gated the number of times the ratio difference (Costs/ valuation of human life, regardless of the age at which LY — Costs/QALY) was positive, zero or negative. They it is lived." Others are less worried by this kind of disfound that 60% of the ratio differences were positive.30 crimination, because of the fact that we may all experiA positive ratio difference is the expected finding ence each age level. The discriminatory effect of setting and implies that the cost-effectiveness ratios become unequal age weights is therefore ethically very differmore favourable when life years are quality adjusted. ent from the setting of weights according to ethnicity, Furthermore, they found that 3% of the ratio differen- sex, education or income." ces were zero, implying that morbidity is not an issue for these conditions, while 36% of the ratio differences "DALY Weights Are Discriminating" were negative.30 The latter of these observations means that quality adjustment has reduced the denominator It has been argued that disability weights are discriminin the cost-effectiveness ratio. This is conceivable only ating against developing countries. In the DALY manual, in two situations: (i) when a treatment has side effects disability weights for clinical conditions are reported affecting the quality of life sufficiently to out-crowd both before and after treatment, based on the general its life-extending effect (e.g., cancer treatment when idea that a disease is less burdensome if it is treated. prognosis is poor) or (ii) when the health improve- In a paper exploring whether DALY weights can influment of a treatment is out-crowded by shortened life ence global priority setting, Arnesen and Kapiriri claim expectancy. The main lesson, nevertheless, seems to that since few of the communicable diseases, which are be that quality adjustments increase the attractiveness typical developing country conditions, are reported of interventions in terms of cost-effectiveness and with reduced disease weights after treatment, the DALY

387

II

388

,911

II

7 Delivery of Health Care and Resource Allocation

methodology is discriminating against developing countries." Arnesen and Kapiriri are also critical of the general level of disease weights between communicable and non-communicable diseases, indicating that the non-communicable diseases generally should have been weighted more heavily.3. Furthermore, Arnesen ar d Kapiriri claim that disease weights discriminate against the young, because non-communicable diseases, which are a relatively higher burden in the elderly population, generally are assigned higher weights than the communicable diseases, which are relatively n- ore important in young populations. In particular, they question that a condition like developmental disability due to malnutrition, primarily affecting children, has a disease weight of only 0.024, while unipolar major depression, which is a typical adult condition, ;las a weight of 0.600." The young are also discriminated against, in their view, because of the age weights.' It is evidently correct that the disease weights of the DALY methodology need to be looked further into, with respect both to elicitation techniques, level of disease burden and change of weights after treatment. Several examples given by Arnesen and Kapiriri illustrate this, for example, that amputation of a thumb has been assigned higher disease weight than amputation of an arm. 14,31 Without further empirical evidence, there are few reasons either to support or to dismiss Arnesen and Kapiriri's general critique of the level of disability weights for different groups of diseases. Nevertheless, it is a main point that since health care resources are generally distributed within and not between countries, the claim that people with communicable diseases and developing countries are discriminated against due to disease-weighted DALYs remains undocumented. Arnesen and Kapiriri also conclude that age weights lead to the disease burden carried by the youngest age groups are given considerably less weight and, hence, less priority than it would without these choices.31 This conclusion is true, but only in the case of non-fatal and non-chronic conditions. For fatal or chronic conditions, on the other hand, it can be demonstrated that age weights favour people aged less than 35-45 years. The reason for this is that for fatal or chronic conditions, the DALY calculations of young people will include the peak of the hump shaped ageweighting function. This is in fact illustrated in Figure 1 in Arnesen and Kapiriri's own paper," where different combinations of discounted/undiscounted and ageweighted/non-age-weighted DALYs are presented. However, the authors fail to sufficiently pull forward this finding and base their conclusion on the joint effect of age weighting and discounting, which in fact pulls in

opposite directions with respect to the burden on the youngest. The aggregate effect of age weights on children, including chronic, recurrent, fatal and non-fatal conditions, remains uncertain. However, if we believe that mortality is more important for burden of disease than morbidity, age weights will lead to more resource being allocated to interventions targeting children. The Effect of Weighting in Two Applied Economic Evaluations From the above theoretical considerations, it is useful to turn our attention towards two applied economic evaluations and consider how disability weighting, age weighting and discounting influence the costeffectiveness and attractiveness in these cases. The scenarios of these two studies are different over several dimensions and therefore useful to illustrate what difference different outcome measures may have for health care priority setting. The first of these is an economic evaluation of zinc as adjunct therapy in diarrhoea,28 while the second is a study of the cost-effectiveness of pharmaceutical interventions to prevent cardiovascular disease in sub-Saharan Africa.29 Both of the two economic evaluations include a number of treatment alternatives. In Table 1, the costeffectiveness ratios (cERs) for two arbitrary example interventions are presented, zinc and oral rehydration solution in diarrhoea management and aspirin and a diuretica given to patients with very high cardiovascular risk Results are first presented as USD per LY gained, secondly as USD per non-age-weighted DALY, which roughly corresponds to QALYs, and finally as USD per standard age-weighted DALY. In all cases, both costs and health effects are discounted using a rate of 3%. The CER of zinc is 74 USD per LY gained, which improves slightly to 70 USD per non—age-weighted DALY (or QALY) by including a disease weight of 0.119

Table 1 The Impact on the Cost-Effectiveness Ratio of Using LY Gained, QALYs (Non-AgeWeighted DALYs) and Standard Age-Weighted DALYs in Interventions Targeting Childhood Diarrhoea and Preventive Cardiology Aspirin and Zinc in diuretica in childhood preventive diarrhoea cardiology USD per LY gained USD per DALY (no age weights) USD per DALY (age weighted)

74 70 66

55 46 56

Robberstad: QALYs vs DALYs vs LYs Gained

for diarrhoea. When the outcome in addition is age weighted to standard DALYs, the CER further improves to 66 USD per DALY (see Table 1). In the case of diarrhoea, it therefore turns out that both disease weighting and age weighting makes the intervention more costeffective and therefore more attractive for implementation. This result supports the view that age weighting is likely to favour interventions targeting children. While diarrhoea is a disease group typically affecting young children, preventive cardiology is foremost relevant for adults aged 40 and higher. In this case, the CER improved from 55 USD per LY to 46 USD per non-ageweighted DALY (or QALY) by including disease weights of 0.268 and 0.490 for stroke and coronary heart disease, respectively. This improvement is rather large because the disease weights for stroke and coronary heart disease are quite high and because people who survive such attacks can be expected to live with a disability for their remaining lives. Inclusion of the morbidity component therefore makes a larger difference in cardiology than in diarrhoea, where each case has an expected duration of only 7 days28 and the disease weight is relatively small. While age weighting made diarrhoea management more attractive, the opposite is the case for preventive cardiology. For this intervention, the positive effect of disability weighting is out-crowded by age weighting, resulting in a CER of 56 USD per DALY. This is because cardiovascular patients are relatively old and already have passed the "hump" of the age-weight curve and are facing declining age weights for the remaining life span. This supports the undisputed view that age weights are disfavouring the elderly. Concluding Remarks

In this paper some of the differences between the health measures life years gained, QALYs and DALYs are briefly presented. To wrap up, one main difference is that whereas LYs gained is a pure measure of mortality the two latter seek to incorporate morbidity by doing disability and quality of life weighting. A major difference between QALY and DALY is that the latter measure in addition incorporates age weights. It has also been discussed what difference these alternative measures of health improvement may have for priority setting. Claims that health related quality of life weighting has little impact on priority setting have been briefly presented and discussed. Furthermore, worries that DALY weights are unfair and discriminating have been briefly presented. Some of the critique is evidently reasonable; in particular there seems to be a need to further improve and justify the DALY disease weights. Other parts of the critique are in my opinion

undocumented, in particular the claim that age weights discriminate against children. It is undoubtedly true that age weighting and disease weighting have ethical implications that are challenging, as critics frequently have pointed out. We have for example seen that the method used to elicit disease weights may influence the morbidity estimates, with visual analogue scales (vAS) seemingly weighting conditions as more severe than the other methods. If comparison is made across diseases where different elicitation techniques for disease severity have been used, this may lead to erroneous priority setting decisions. The direction of this error is, however, dependent on the circumstances in each case. The important question is not whether these weights and value choices may influence resource allocation. Because dearly, making a difference is what they are intended to do. A weighting principle that doesn't create winners, as well as losers, is inconceivable. One must therefore ask whether the weights brought about by QALYs and DALYs are contributing to improved population health and reduced health inequalities in sub-Saharan Africa. All the measures—LYs, QALY and DALY—clearly lead to more weight being accrued to the young over the elderly compared to a simple measure of mortality like deaths averted. This is in accordance with both an objective of improved population health and common notions of equality. The effects on population health and health inequalities of moving from LYs gained to QALY or DALY are perhaps less obvious. It might be argued that in settings with extreme resource scarcity with many conditions that can be treated cheaply and effectively, crude estimates of population health are sufficient to inform priority setting. It may further be argued that more subtle analysis, such as the QALY and DALY, will not give more real information than such crude measures. Many doubt that it at all is possible to translate fatal and nonfatal health outcomes into one numerical unit. How, asks Hilda Bastian, do you decide which problems and limitations are worse than others, and give them a numerical weight?32 Arnesen and Kapiriri worry that valuable epidemiological information may lose importance if it is mixed with value choices of unknown validity.31 Taken together, such concerns cast doubt on the value of summary health measures like the QALY and DALY in improving population health. On the other hand, the number of even basic health services that cannot be financed in settings with extreme resource scarcity is very large. This, I believe, increases the importance of developing and applying methodology that is as objective and consistent as possible. Obviously, there are challenging value choices embedded in these methods, but trade-offs between

I

II :III

MILLI

7 Delivery of Health Care and Resource Allocation

quantity for quality of health need to be made in public health systems, they are in fact being made, and they will continue to be made, with implicit or explicit value judgements. The adva :nage of health outcomes combining mortality and morbidity, like the DALY, is that they can be used to inform such trade-offs in a fairly transparent and consistent way. Many feel that only through quantification of morbidity, as well as

mortality, it is possible to have a fair amount of resources allocated to rehabilitation of chronic conditions and marginal patient groups.33 Irrespective of these views, in order for the policy mix to be perceived fair (equitable) as well as effective in improving population health within limited budgets, a continuous public scrutiny and evaluation of the applied weights and value choices needs to be made.'

References 1. WHO. World Health Report 2002. Geneva: World Health Organization, 2002. 2. Lopez AD, Ahmad OB, Guillot M, Inoue M, Ferguson BD, Salomon JA. Life Tables for 191 countries for 2000: Data, Methods, Results. Geneva: Evidence and Information for Policy (EIP). World Health Organization, 2001. GPE Discussion Paper No. 90. 3. WHO. World Health Report 2005. Geneva World Health Organization, 2005. 4. who.int IHomepage on the internet). World Health Organization, 2004. Vaccines, Immunization and Biologicals. (Accessed 09.11.2004, at www.who.int/ vaccines-surveillance/StatsAndGraphs.htm). 5. Miller MA, McCann L. Policy analysis of the use of hepatitis B, haemophilus influelzae type B, streptococcus pneumoniae-conjugate and rotavirus vaccines in national immunization schedules. Health Econ 2000; 9: 19-35. 6. World Bank. World Development Report 1993: Investing in Health. New York Oxford University Press, 1993. 7. McKeown T. The direction of medical research. Lancet 1979; 2 (8155): 1281-4. 8. Smith DH, Gravelle H. The practice of discounting in economic evaluation of health care interventions. Int J Technol Assess Health Care 2001; 17: 236-43. 9. Walker D, Kumaranayake L. Allowing for differential timing in cost analyses: discounting and annualization. Health Policy Plan 2002; 17 (1): 112-18. 10. Parfit D. Reasons and Persons. Oxford: Clarendon Press, 1984. 11. Viscusi WK. Discounting health effects for medical decisions. In: Sloan FA, ed. Valuing Health Care Costs, Benefits, and Effectiveness of Pharmaceuticals and Other Medical Technologies. Cambridge: Cambridge University Press, 1995: 125-47. 12. Weinstein MC, Stason WB. Foundations of costeffectiveness analysis for health and medical practices. N Engl J Med 1977; 296: 716-21. 13. Dolan P. The measurement of health-related quality of life for use in resource allocation decisions in health care. In: Culyer AJ, Newhouse JP, eds. Handbook of Health Economics. Amsterdam: Elsevier, 2000: 1723-60. 14. Murray C, Lopez A. The Global Burden of Disease. A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020. The Harvard School of Public Health, 1996.

15. Klarman H, Francis J, Rosenthal G. Cost-effectiveness analysis applied to the treatment of chronic renal disease. Med Care 1968; 6 (1): 48-54. 16. Weinstein MC, Stason WB. Hypertension: A Policy Perspective. Cambridge, MA: Harvard University Press, 1976. 17. Anand 5, Hanson K. Disability-adjusted life years: a critical review. J Health Econ 1997; 16 (6): 685-702. 18. Reidpath D, Allotey PA, Kouame A, Cummins RA. Measuring health in a vacuum: examining the disabilityweight of the DALY. Health Policy Plan 2003; 18 (4): 351-6. 19. Salomon JA, Murray CJL. A multi-method approach to measuring health-state valuations. Health Econ 2004; 13: 281-90. 20. Nord E. Cost-Value Analysis in Health Care. Cambridge: Cambridge University Press, 1999. 21. Tsuchiya A. QALYs and ageism: Philosophical theories and age weighting. Health Econ 2000; 9: 57-68. 22. Tsuchiya A. Age-related preferences and age weighting health benefits. Soc Sci Med 1999; 48: 267-76. 23. Harris J. QALYfying the value of life. J Med Ethics 1987; 13: 119-20. 24. Nord E, Street A, Richardson J, Kuhse H, Singer P. The significance of age and duration of effect in social evaluation of health care. Health Care Anal 1996; 4 (2): 103-11. 25. Williams A. Intergenerational equity: An exploration of the "fair innings" argument. Health Econ 1997; 6: 117-32. 26. Cookson R, Dolan P. Principles of justice in health care rationing. J Med Ethics 2000; 26: 323-9. 27. Bleichrodt H, Doctor J, Stolk E A nonparametric elicitation of the equity-efficiency trade-off in cost-utility analysis. J Health Econ 2005; 24 (4): 655-78. 28. Robberstad B, Strand T, Black RE, Sommerfelt H. Costeffectiveness of zinc as adjunct therapy for acute childhood diarrhea in developing countries. Bull WHO 2004; 82 (7): 523-31. 29. Robberstad B, Hemed Y, Norheim OE Cost-effectiveness of medical interventions to prevent cardiovascular disease in a sub-Saharan African country-the case of Tanzania. In: Robberstad B. Economic evaluation of health interventions in sub-Saharan Africa Applied economic evaluations and studies on time preferences for health in Tanzania Bergen: Dissertation for the PhD degree, University of Bergen, 2005. 30. Chapman RH, Berger M, Weinstein MC, Weeks JC, Goldie 5, Neumann PJ. When does quality-adjusting life-years matter in cost-effectiveness analysis? Health Econ 2004; 13: 429-36.

Robberstad: QALYs vs DALYs vs LYs Gained 31. Amesen T, Kapiriri L Can value choices in DALYs influence global priority setting? Health Policy 2004; 70: 137-49. 32. Bastian H. A consumer trip into the world of the DALY calculations: An Alice in Wonderland experience. Reprod Health Matters 2000; 8 (15): 113-15.

33. Jelsma J, Weerdt WD, Cock PD. Disability Adjusted Life Years (DALYs) and rehabilitation. Disabil Rehab 2002; 24 (7): 378-82. 34. Sen A. Development as Freedom. Oxford: Oxford University Press, 1999.

7.4 Microallocation: Decisions at the Bedside

The Allocation of Exotic Medical Life-Saving Therapy Nicholas P. Rescher L The Problem Technological progress has in recent years transformed the limits of the possible in medical therapy. However, the elevated state of sophistication of modern medical technology has brought the economists' classic problem of scarcity in its wake as an unfortunate side product. The enormously sophisticated and complex equipment and the highly trained teams of experts requisite for its utilization are scarce resources in relation to potential demand. The administrators of the great medical institutions that preside over these scarce resources thus come to be faced increasingly with the awesome choice: Whose life to save? A (somewhat hypothetical) paradigm example of this problem may be sketched within the following set of definitive assumptions: we suppose that persons in some particular medically morbid condition are "mortally afflicted"; it is virtually certain that they will die within a short time period (say, 90 days). We assume that some very complex course of treatment (e.g., a heart transplant) represents a substantial probability of life prolongation for persons in this mortally afflicted condition. We assume that the facilities available in terms of human resources, mechanical instrumentalities, and requisite materials (e.g., hearts in the case of a heart transplant) make it possible to give a certain treatment—this "exotic (medical) lifesaving therapy," or ELT for short—to a certain, relatively small number of people. And finally we assume that a substantially greater pool of people in the mortally afflicted condition

is at hand. The problem then may be formulated as follows: how is one to select within the pool of afflicted patients the ones to be given the Err treatment in question; how to select those "whose lives are to be saved"? Faced with many candidates for an ELT process that can be made available to only a few, doctors and medical administrators confront the decision of who is to be Oven a chance at survival and who is, in effect, to be condemned to die. As has already been implied, the "heroic" variety of spare-part surgery can pretty well be assimilated to this paradigm. One can foresee the time when heart transplantation, for example, will have become pretty much a routine medical procedure, albeit on a very limited basis, since a cardiac surgeon with the technical competence to transplant hearts can operate at best a rather small number of times each week and the elaborate facilities for such operations will most probably exist on a modest scale. Moreover, in "spare-part" surgery there is always the problem of availability of the "spare parts" themselves. A report in one British newspaper gives the following picture: "Of the 150,000 who die of heart disease each year [in the UK], Mr Donald Longmore, research surgeon at the National Heart Hospital [in London] estimates that 22,000 might be eligible for heart surgery. Another 30,000 would need heart and lung transplants. But there are probably only between 7,000 and 14,000 potential donors a year."' Envisaging this situation in which at the very most something like one in four heart-malfunction victims can be saved, we clearly confront a problem in ELT allocation. A perhaps even more drastic case in point is afforded by long-term hemodialysis, an ongoing process by which a complex device—an "artificial kidney machine"—is used periodically in cases of chronic renal failure to substitute for a non-functional kidney

Source: From Ethics 79,3 (1969), 173-86. Reprinted with permission.

II

1111:111

7 Delivery of Health Care and Resource Allocation

in "cleaning" potential poisons from the blood. Only a few major institutions have chronic hemodialysis units, whose complex operation is an extremely expensive proposition. For the present and the foreseeable future the situation is that "the number of places available for chronic hemodialysis is hopelessly inadequate."2 The traditional medical ethos has insulated the physician against facing the very existence of this problem. When swearing the Hippocratic Oath, he commits himself to work for the benefit of the sick in "whatsover house I enter."3 In taking this stance, the physician substantially renounces the explicit choice of saving certain lives rather than others. Of course, doctors have always in fact nad to face such choices on the battlefield or in times of disaster, but there the issue had to be resolved hurriedly, under pressure, and in circumstances in which the very nature of the case effectively precluded calm deliberation by the decisionmaker as well as criticism by others. In sharp contrast, however, cases of the type we have postulated in the present discussion arise predictably, and represent choices to be made deliberately and "in cold blood." It is, to begin with, appropriate to remark that this problem is not fundamentally a medical problem. For when there are sufficiently many afflicted candidates for ELT then—so we may assume—there will also be more than enough for whom the purely medical grounds for ELT allocation are decisively strong in any individual case, and just about equally strong throughout the group. But in this circumstance a selection of some afflicted patients over and against others cannot ex hypothesi be made on the basis of purely medical considerations. The selection problem, as we have said, is in substantial measure not a medical one. It is a problem for medical men, which must somehow be solved by them, but that does not make it a medical issue—any more than the problem of hospital building is a medical issue. As a problem it belongs to the category of philosophical problems—specifically a problem of moral philosophy or ethics. Structurally, it bears a substantial kinship with those issues in this field that revolve about the notorious whom-to-save-on-the-lifeboat and whom-to-throw-tothe-wolves-pursuing-the-sled questions. But whereas questions of this just-indicated sort are artificial, hypothetical, and far-fetched, the ELT issue poses a genuine policy question for the responsible administrators in medical institutions, indeed a question that threatens to become commonplace in the fo:1-eseeable future. Now what the medical administrator needs to have, and what the philosopher is presumably ex officio in a position to help in providing, is a body of rational guidelines for making choices in these literally

life-or-death situations. This is an issue in which many interested parties have a substantial stake, including the responsible decision-maker who wants to satisfy his conscience that he is acting in a reasonable way Moreover, the family and associates of the man who is turned away—to say nothing of the man himself—have the right to an acceptable explanation. And indeed even the general public wants to know that what is being done is fitting and proper. All of these interested parties are entitled to insist that a reasonable code of operating principles provides a defensible rationale for making the life-and-death choices involved in ELT. II.The Two Types of Criteria

Two distinguishable types of criteria are bound up in the issue of making ELT choices. We shall call these Criteria of Inclusion and Criteria of Comparison, respectively. The distinction at issue here requires some explanation. We can think of the selection as being made by a two-stage process: (1) the selection from among all possible candidates (by a suitable screening process) of a group to be taken under serious consideration as candidates for therapy, and then (2) the actual singling out, within this group, of the particular individuals to whom therapy is to be given. Thus the first process narrows down the range of comparative choice by eliminating en bloc whole categories of potential candidates. The second process calls for a more refined, case-bycase comparison of those candidates that remain. By means of the first set of criteria one forms a selection group; by means of the second set, an actual selection is made within this group. Thus what we shall call a "selection system" for the choice of patients to receive therapy of the ELT type will consist of criteria of these two kinds. Such a system will be acceptable only when the reasonableness of its component criteria can be established. III. Essential Features of an Acceptable ELT Selection System

To qualify as reasonable, an ELT selection must meet two important "regulative" requirements: it must be simple enough to be readily intelligible, and it must be plausible, that is, patently reasonable in a way that can be apprehended easily and without involving ramified subtleties. Those medical administrators responsible for ELT choices must follow a modus operandi that virtually all the people involved can readily understand to be acceptable (at a reasonable level of generality, at any rate). Appearances are critically important here. It is not enough that the choice be made in a justifiable

Rescher: The Allocation of Exotic Medical Life-Saving Therapy

way, it must be possible for people—plain people—to "see" (i.e., understand without elaborate teaching or indoctrination) that it is justified, insofar as any mode of procedure can be justified in cases of this sort. One "constitutive" requirement is obviously an essential feature of a reasonable selection system: all of its component criteria—those of inclusion and those of comparison alike—must be reasonable in the sense of being rationally defensible. The ramifications of this requirement call for detailed consideration. But one of its aspects should be noted without further ado: it must be fair—it must treat relevantly like cases alike, leaving no room for "influence" or favouritism, etc. IV. The Basic Screening Stage: Criteria of Inclusion (and Exclusion)

Three sorts of considerations are prominent among the plausible criteria of inclusion/exclusion at the basic screening stage: the constituency factor, the progressof-science factor, and the prospect-of-success factor.

the institutionalization of medicine, moving it away from its original status as an art practised by an individual practitioner. B.The Progress-of-Science Factor

The needs of medical research can provide a second valid principle of inclusion. The research interests of the medical staff in relation to the specific nature of the cases at issue is a significant consideration. It may be important for the progress of medical science— and thus of potential benefit to many persons in the future—to determine how effective the ELT at issue is with diabetics or persons over 60 or with a negative RH factor. Considerations of this sort represent another type of legitimate factor in ELT selection. A very definitely borderline case under this head would revolve around the question of a patient's willingness to pay, not in monetary terms, but in offering himself as an experimental subject, say by contracting to return at designated times for a series of tests substantially unrelated to his own health, but yielding data of importance to medical knowledge in general.

A. The Constituency Factor

It is a "fact of life" that ELI can be available only in the institutional setting of a hospital or medical institute or the like. Such institutions generally have normal clientele boundaries. A veterans' hospital will not concern itself primarily with treating non-veterans, a children's hospital cannot be expected to accommodate the "senior citizen," an army hospital can regard college professors as outside its sphere. Sometimes the boundaries are geographic—a state hospital may admit only residents of a certain state. (There are, of course, indefensible constituency principles—say race or religion, party membership, or ability to pay; and there are cases of borderline legitimacy, e.g., sex.4) A medical institution is justified in considering for ELI only persons within its own constituency, provided this constituency is constituted upon a defensible basis. Thus the hemodialysis selection committee in Seattle "agreed to consider only those applications who were residents of the state of Washington. . . . They justified this stand on the grounds that since the basic research . . . had been done at . . . a state-supported institution, the people whose taxes had paid for the research should be its first beneficiaries."5 While thus insisting that constituency considerations represent a valid and legitimate factor in Eur selection, I do feel there is much to be said for minimizing their role in life-or-death cases. Indeed a refusal to recognize them at all is a significant part of medical tradition, going back to the very oath of Hippocrates. They represent a departure from the ideal arising with

C. The Prospect-of-Success Factor It may be that while the ELT at issue is not without some

effectiveness in general, it has been established to be highly effective only with patients in certain specific categories (e.g., females under 40 of a specific blood type). This difference in effectiveness—in the absolute or in the probability of success—is (we assume) so marked as to constitute virtually a difference in kind rather than in degree. In this case, it would be perfectly legitimate to adopt the general rule of making the ELI at issue available only or primarily to persons in this substantialpromise-of-success category. (It is on grounds of this sort that young children and persons over 50 are generally ruled out as candidates for hemodialysis.) We have maintained that the three factors of constituency, progress of science, and prospect of success represent legitimate criteria of inclusion for ELI selection. But it remains to examine the considerations which legitimate them. The legitimating factors are in the final analysis practical or pragmatic in nature. From the practical angle it is advantageous—indeed to some extent necessary—that the arrangements governing medical institutions should embody certain constituency principles. It makes good pragmatic and utilitarian sense that progress-of-science considerations should be operative here. And, finally, the practical aspect is reinforced by a whole host of other considerations—including moral ones—in supporting the prospect-of-success criterion. The workings of each of these factors are of course conditioned by the ever-present element of limited

II

all

IIILLI

7 Delivery of Health Care and Resource Allocation

availability. They are operative only in this context, that is, prospect of success is a leg:timate consideration at all only because we are dealing with a situation of scarcity. V.The Final Selection Stage: Criteria of Selection Five sorts of elements must, as we see it, figure primarily among the plausible criteria of selection that are to be brought to bear in further screening the group constituted after application of the criteria of inclusion: the relative-likelihood-of-success factor, the life-expectancy factor, the family-role factor, the potential-contributions factor, and the services-rendered factor. The first two represent the biomedical aspect, the second three the social aspect. A.The Relative-Likelihood-of-Success Factor It is clear that the relative likelihood of success is a legitimate and appropriate factor in making a selection within the group of qualified patients that are to receive ELT. This is obviously one of the considerations that must count very significantly in a reasonable selection procedure. The present criterion is of course closely related to item C of the preceding section. There we were concerned with prospect-of-success considerations categorically and en bloc. Here at present they come into play in a particularized case-by-case comparison among individuals. If the therapy at issue is not a onceand-for-all proposition and requires ongoing treatment, cognate considerations must be brought in. Thus, for example, in the case of a chronic ELT procedure such as hemodialysis it would clearly make sense to give priority to patients with a potentially reversible condition (who would thus need treatment for only a fraction of their remaining lives). B.The Life-Expectancy Factor Even if the ELT is "successful" in the patient's case he may, considering his age and/or other aspects of his general medical condition, look forward to only a very short probable future life. This is obviously another factor that must be taken into account. C.The Family-Role Factor A person's life is a thing of importance not only to himself but to others—friends, associates, neighbours, colleagues, etc. But his (or her) relationship to his immediate family is a thing of unique intimacy and significance. The nature of his relationship to his wife, children, and parents, and the issue of their financial and psychological dependence upon him, are

obviously matters that deserve to be given weight in the ELT selection process. Other things being anything like equal, the mother of minor children must take priority over the middle-aged bachelor. D.The Potential Future-Contributions Factor (Prospective Service) In "choosing to save" one life rather than another, "the society," through the mediation of the particular medical institution in question—which should certainly look upon itself as a trustee for the social interest—is clearly warranted in considering the likely pattern of future services to be rendered by the patient (adequate recovery assumed), considering his age, talent, training, and past record of performance. In its allocations of ELT, society "invests" a scarce resource in one person as against another and is thus entitled to look to the probable prospective "return" on its investment. It may well be that a thoroughly egalitarian society is reluctant to put someone's social contribution into the scale in situations of the sort at issue. One popular article states that "the most difficult standard would be the candidate's value to society," and goes on to quote someone who said: "You can't just pick a brilliant painter over a laborer. The average citizen would be quickly eliminated."6 But what if it were not a brilliant painter but a brilliant surgeon or medical researcher that was at issue? One wonders if the author of the obiter dictum that one "can't just pick" would still feel equally sure of his ground. In any case, the fact that the standard is difficult to apply is certainly no reason for not attempting to apply it. The problem of ELT selection is inevitably burdened with difficult standards. Some might feel that in assessing a patient's value to society one should ask not only who if permitted to continue living can make the greatest contribution to society in some creative or constructive way, but also who by dying would leave behind the greatest burden on society in assuming the discharge of their residual responsibilities? Certainly the philosophical utilitarian would give equal weight to both these considerations. Just here is where I would part ways with orthodox utilitarianism. For—though this is not the place to do so—I should be prepared to argue that a civilized society has an obligation to promote the furtherance of positive achievements in cultural and related areas even if this means the assumption of certain added burdens.' E. The Past Services-Rendered Factor (Retrospective Service) A person's services to another person or group have always been taken to constitute a valid basis for a claim upon this person or group—of course a moral and

Rescher: The Allocation of Exotic Medical Life-Saving Therapy not necessarily a legal claim. Society's obligation for the recognition and reward of services rendered—an obligation whose discharge is also very possibly conducive to self-interest in the long run—is thus another factor to be taken into account. This should be viewed as a morally necessary correlative of the previously considered factor of prospective service. It would be morally indefensible of society in effect to say: "Never mind about services you rendered yesterday—it is only the services to be rendered tomorrow that will count with us today" We live in very future-oriented times, constantly preoccupied in a distinctly utilitarian way with future satisfactions. And this disinclines us to give much recognition to past services. But parity considerations of the sort just adduced indicate that such recognition should be given on grounds of equity. No doubt a justification for giving weight to services rendered can also be attempted along utilitarian lines. ("The reward of past services rendered spurs people on to greater future efforts and is thus socially advantageous in the long-run future.") In saying that past services should be counted "on grounds of equity"8—rather than "on grounds of utility"—I take the view that even if this utilitarian defence could somehow be shown to be fallacious, I should still be prepared to maintain the propriety of taking services rendered into account. The position does not rest on a utilitarian basis and so would not collapse with the removal of such a basis.9 As we have said, these five factors fall into three groups: the biomedical factors A and B, the familial factor c, and the social factors D and E. With items A and B the need for a detailed analysis of the medical considerations comes to the fore. The age of the patient, his medical history, his physical and psychological condition, his specific disease, etc., will all need to be taken into exact account. These biomedical factors represent technical issues: they call for the physicians' expert judgment and the medical statisticians' hard data. And they are ethically uncontroversial factors—their legitimacy and appropriateness are evident from the very nature of the case. Greater problems arise with the familial and social factors. They involve intangibles that are difficult to judge. How is one to develop subcriteria for weighing the relative social contributions of (say) an architect or a librarian or a mother of young children? And they involve highly problematic issues. (For example, should good moral character be rated a plus and bad a minus in judging services rendered?) And there is something strikingly unpleasant in grappling with issues of this sort for people brought up in times greatly inclined toward maxims of the type "Judge not!" and "Live and let live!" All the same, in the situation that concerns us here such

distasteful problems must be faced, since a failure to choose to save some is tantamount to sentencing all. Unpleasant choices are intrinsic to the problem of ELT selection; they are of the very essence of the matter.'° But is reference to all these factors indeed inevitable? The justification for taking account of the medical factors is pretty obvious. But why should the social aspect of services rendered and to be rendered be taken into account at all? The answer is that they must be taken into account not from the medical but from the ethical point of view. Despite disagreement on many fundamental issues, moral philosophers of the present day are pretty well in consensus that the justification of human actions is to be sought largely and primarily— if not exclusively—in the principles of utility and of justice." But utility requires reference of services to be rendered and justice calls for a recognition of services that have been rendered. Moral considerations would thus demand recognition of these two factors. (This, of course, still leaves open the question of whether the point of view provides a valid basis of action: Why base one's actions upon moral principles?—or, to put it bluntly—Why be moral? The present paper is, however, hardly the place to grapple with so fundamental an issue, which has been canvassed in the literature of philosophical ethics since Plato.) VI. More Than Medical Issues Are Involved An active controversy has of late sprung up in medical circles over the question of whether non-physician laymen should be given a role in ELT selection (in the specific context of chronic hemodialysis). One physician writes: "I think that the assessment of the candidates should be made by a senior doctor on the [dialysis] unit, but I am sure that it would be helpful to him—both in sharing responsibility and in avoiding personal pressure—if a small unnamed group of people [presumably including laymen] officially made the final decision. I visualize the doctor bringing the data to the group, explaining the points in relation to each case, and obtaining their approval of his order of priority.',12 Essentially this procedure of a selection committee of laymen has for some years been in use in one of the most publicized chronic dialysis units, that of the Swedish Hospital of Seattle, Washington.° Many physicians are apparently reluctant to see the choice of allocation of medical therapy pass out of strictly medical hands. Thus in a recent symposium on the "Selection of Patients for Haemodialysis,"14 Dr Ralph Shakman writes: "Who is to implement the selection? In my opinion it must ultimately be the responsibility of the consultants in charge of the renal units. . . . I can

395

II

396

j11

IIILLI

7 Delivery of Health Care and Resource Allocation

see no reason for delegating this responsibility to lay persons. Surely the latter would be better employed if they could be persuaded to devote their time and energy to raise more and more money for us to spend on our patients."5 Other contributors to this symposium strike much the same note. Dr F.M. Parsons writes: "In an attempt to overcome . . . difficulties in selection some have advocated introducing certain specified lay people into the discussions. Is it wise? I doubt whether a committee of this type can adjudicate as satisfactorily as two medical colleagues, particularly as successful therapy involves close cooperation between doctor and patient."m And Dr M.A. Wilson writes in the same symposium: "The suggestion has been made that lay panels should select individuals for dialysis from among a group who are medically suitable. Though this would relieve the doctor-in-charge of a heavy load of responsibility, it would place the burden on those who have no personal knowledge and have to base their judgments on medical or social reports. I do not believe this would result in better decisions for the group or improve the doctor—patient relationship in individual cases."" But no amount o f flag waving about the doctor's facing up to his responsibility—or prostrations before the idol of the doctor—patient relationship and reluctance to admit laymen into the sacred precincts of the conference chambers of medical consultations—can obscure the essential fact that ELT selection is not a wholly medical problem. When there are more than enough places in an ELT program to accommodate all who need it, then it will clearly be a medical question to decide who does have the need and which among these would successfully respond. But when an admitted gross insufficiency of places exists, when there are 10 or 50 or 100 highly eligible candidates for each place in the program, then it is unrealistic to take the view that purely medical criteria can furnish a sufficient basis for selection. The question o ELT selection becomes serious as a phenomenon of scale—because, as more candidates present themselves, strictly medical factors are increasingly less adequate as a selection criterion precisely because by numerical category-crowding there will be more and more cases whose "status is much the same" so far as purely medical considerations go. The ELI- selection problem clearly poses issues that transcend the medical sphere because—in the nature of the case—many residual issues remain to be dealt with once all of the medical questions have been faced. Because of this there is good reason why laymen as well as physicians should be involved in the selection process. Once the medical considerations have been brought to bear, fundamental social issues remain to be

resolved. The instrumentalities of ELr have been created through the social investment of scarce resources, and the interests of the society deserve to play a role in their utilization. As representatives of their social interests, lay opinions should function to complement and supplement medical views once the proper arena of medical considerations is left behind.'8 Those physicians who have urged the presence of lay members on selection panels can, from this point of view, be recognized as having seen the issue in proper perspective. One physician has argued against lay representation on selection panels for hemodialysis as follows: "If the doctor advises dialysis and the lay panel refuses, the patient will regard this as a death sentence passed by an anonymous court from which he has no right of appeal."19 But this drawback is not specific to the use of a lay panel. Rather, it is a feature inherent in every selection procedure, regardless of whether the selection is done by the head doctor of the unit, by a panel of physicians, etc. No matter who does the selecting among patients recommended for dialysis, the feelings of the patient who has been rejected (and knows it) can be expected to be much the same, provided that he recognizes the actual nature of the choice (and is not deceived by the possibly convenient but ultimately poisonous fiction that because the selection was made by physicians it was made entirely on medical grounds). In summary, then, the question of ELr selection would appear to be one that is in its very nature heavily laden with issues of medical research, practice, and administration. But it will not be a question that can be resolved on solely medical grounds. Strictly social issues of justice and utility will invariably arise in this area—questions going outside the medical area in whose resolution medical laymen can and should play a substantial role. VII. The Inherent Imperfection (Non-Optimality) of Any Selection System

Our discussion to this point of the design of a selection system for ELT has left a gap that is a very fundamental and serious omission. We have argued that five factors must be taken into substantial and explicit account: a. Relative likelihood of success. Is the chance of the treatment's being "successful" to be rated as high, good, average, etc.?2° b. Expectancy of future life. Assuming the "success" of the treatment, how much longer does the patient stand a good chance (75 per cent or better) of living—considering his age and general condition?

Rescher: The Allocation of Exotic Medical Life-Saving Therapy

c. Family role. To what extent does the patient have responsibilities to others in his immediate family? d. Social contributions rendered. Are the patient's past services to his society outstanding, substantial, average, etc.? e. Social contributions to be rendered. Considering his age, talents, training, and past record of performance, is there a substantial probability that the patient will—adequate recovery being assumed— render in the future services to his society that can be characterized as outstanding, substantial, average, etc.? This list is clearly insufficient for the construction of a reasonable selection system, since that would require not only that these factors be taken into account (somehow or other), but—going beyond this—would specify

a specific set of procedures for taking account of them. The specific procedures that would constitute such a system would have to take account of the interrelationship of these factors (e.g., B and E), and to set out exact guidelines as to the relevant weight that is to be given to each of them. This is something our discussion has not as yet considered. In fact, I should want to maintain that there is no such thing here as a single rationally superior selection system. The position of affairs seems to me to be something like this: (1) It is necessary (for reasons already canvassed) to have a system, and to have a system that is rationally defensible, and (2) to be rationally defensible, this system must take the factors A—E into substantial and explicit account. But (3) the exact manner in which a rationally defensible system takes account of these factors cannot be fixed in any one specific way on the basis of general considerations. Any of the variety of ways that give A—E "their due" will be acceptable and viable. One cannot hope to find within this range of workable systems some one that is optimal in relation to the alternatives. There is no one system that does "the (uniquely) best"—only a variety of systems that do "as well as one can expect to do" in cases of this sort. The situation is structurally very much akin to that of rules of partition of an estate among the relations of a decedent. It is important that there be such rules. And it is reasonable that spouse, children, parents, siblings, etc., be taken account of in these rules. But the question of the exact method of division—say that when the decedent has neither living spouse nor living children then his estate is to be divided, dividing 60 per cent between parents, 40 per cent between siblings versus dividing 90 per cent between parents, 10 per cent between siblings—cannot be settled on the

basis of any general abstract considerations of reasonableness. Within broad limits, a variety of resolutions are all perfectly acceptable—so that no one procedure can justifiably be regarded as "the (uniquely) best" because it is superior to all others.21 VIII.A Possible Basis for a Reasonable Selection System Having said that there is no such thing as the optimal selection system for ELT, I want now to sketch out the broad features of what I would regard as one acceptable system. The basis for the system would be a point rating. The scoring here at issue would give roughly equal weight to the medical considerations (A and B) in comparison with the extramedical considerations (C = family role, D = services rendered, and E = services to be rendered), also giving roughly equal weight to the three items involved here (C, D, and E). The result of such a scoring procedure would provide the essential starting point of our ELT selection mechanism. I deliberately say "starting point" because it seems to me that one should not follow the results of this scoring in an automatic way. I would propose that the actual selection should only be guided but not actually be dictated by this scoring procedure, along lines now to be explained. IX.The Desirability of Introducing an Element of Chance The detailed procedure I would propose—not of course as optimal (for reasons we have seen), but as eminently acceptable—would combine the scoring procedure just discussed with an element of chance. The resulting selection system would function as follows: 1. First the criteria of inclusion of Section iv above would be applied to constitute a first-phase selection group—which (we shall suppose) is substantially larger than the number n of persons who can actually be accommodated with ELT. 2. Next the criteria of selection of Section v are brought to bear via a scoring procedure of the type described in Section vm. On this basis a second-phase selection group is constituted which is only somewhat larger—say by a third or a half—than the criticalnumber n at issue. 3. If this second-phase selection group is relatively homogeneous as regards rating by the scoring procedure—that is, if there are no really major disparities within this group (as would be likely if the initial group was significantly larger than n)—then the final selection is made by random selection of n persons from within this group.

II

311

111111

7 Delivery of Health Care and Resource Allocation

This introduction of the element of chance—in what could be dramatized as a "lottery of life and death"— must be justified. The fact is that such a procedure would bring with it three substantial advantages. First, as we have argued above (in Section vii), any acceptable selection system is inherently non-optimal. The introduction of the element of chance prevents the results that life-and-death choices are made by the automatic application of an ac mittedly imperfect selection method. Second, a recourse to chance would doubtless make matters easier for the rejected patient and those who have a specific interest in him. It would surely be quite hard for them to accept his exclusion by relatively mechanical application of objective criteria in whose implementation subjective judgment is involved. But the circumstances of life have conditioned us to accept the workings of chance and to tolerate the element of luck (good or bad): human life is an inherently contingent process. Nobody, after all, has an absolute right to ELT—but most of us would feel that we have "every bit as much right" to it as anyone else in significantly similar circumstances. The introduction of the element of chance assures a like

handling of like cases over the widest possible area that seems reasonable in the circumstances. Third (and perhaps least), such a recourse to random selection does much to relieve the administrators of the selection system of the awesome burden of ultimate and absolute responsibility. These three considerations would seem to build up a substantial case for introducing the element of chance into the mechanism of the system for ELT selection in a way limited and circumscribed by other weightier considerations, along some such lines as those set forth above.22 It should be recognized that this injection of man-made chance supplements the element of natural chance that is present inevitably and in any case (apart from the role of chance in singling out certain persons as victims for the affliction at issue). As F.M. Parsons has observed: "any vacancies [in an ELT program— specifically hemodialysis] will be filled immediately by the first suitable patients, even though their claims for therapy may subsequently prove less than those of other patients refused later."23 Life is a chancy business and even the most rational of human arrangements can cover this over to a very limited extent at best.

Notes 1. 2.

3. 4.

5.

Christine Doyle, (1967), "Spate-Part Heart Surgeons," in the Observer (12 May), 623. J.D.N. Nabarro, "Selection of Patients for Haemodialysis," in British Medical Journal, There are about 30 new cases per million of population—only 10 per cent of these can for the foreseeable future be accommodated with chronic hemodialysis. Kidney transplantation—itself a very tricky procedure—cannot make a more than minor contribution here. As this article goes to press, I learn that patients can be maintained in home didysis at an operating cost about half that of maintaining them in a hospital dialysis unit (roughly an $8,000 minimum). In the United States, around 7,000 patients with terminal uremia who could benefit from hemodialysis evolve yearly. As of mid-1968, some 1,000 of these can be accommodated in existing hospital units. By June 1967, a world-wide total of some 120 patients were in treatment by home dialysis. (Data from a forthcoming paper, "Home Dialysis," by C.M. Conty and H.V. Murdaugh. See also R.A. Baillod, et al., (1965), "Overnight Haemodialysis in the Home," in Proceedings of the European Dialysis and Transplant Association 6: 99ff. For the Hippocratic Oath, see .,Iippocrates: Works (Loeb ed.; London, 1959), I, 298. Another example of borderline legitimacy is posed by an endowment "with strings attacned," e.g., "In accepting this legacy the hospital agrees to admit and provide all needed treatment for any direct descer.dant of myself, its founder." Shana Alexander, (1962), "They Decide Who Lives, Who Dies," in life, 53 (9 Nov.), 10/-25 (see 107).

6. Lawrence Lader, (1968), "Who Has the Right To Live?," in Good Housekeeping (Jan.): 144. 7. This approach could thus be continued to embrace the previous factor, that of family role, the preceding item. 8. Moreover a doctrinaire utilitarian would presumably be willing to withdraw a continuing mode of ELT such as hemodialysis from a patient to make room for a more promising candidate who came to view at a later stage and who could not otherwise be accommodated. I should be unwilling to adopt this course, partly on grounds of utility (with a view to the demoralization of insecurity), partly on the non-utilitarian ground that a "moral commitment" has been made and must be honoured. 9. Of course the difficult question remains of the relative weight that should be given to prospective and retrospective service in cases where these factors conflict. There is good reason to treat them on a par. 10. This in the symposium on "Selection of Patients for Haemodialysis," in British Medical Journal (11 March 1967): 622-4. F.M. Parsons writes: "But other forms of selecting patients [distinct from first come, first served] are suspect in my view if they imply evaluation of man by man. What criteria could be used? Who could justify a claim that the life of a mayor would be more valuable than that of the humblest citizen of his borough? Whatever we may think as individuals none of us is indispensable." But having just set out this hard-line view he immediately backs away from it: "On the other hand, to assume that there was little to choose between Alexander Fleming and

Rescher: The Allocation of Exotic Medical Life-Saving Therapy

11. 12. 13. 14. 15.

16. 17.

18.

19. 20.

Adolf Hitler . . . would be nonsense, and we should be naive if we were to pretend that we could not be influenced by their achievements and characters if we had to choose between the two of them. Whether we like it or not we cannot escape the fact that this kind of selection for longterm haemodialysis will be required until very large sums of money become available for equipment and services [so that everyone who needs treatment can be accommodated)." The relative fundamentality of these principles is, however, a substantially disputed issue. J.D.N. Nabarro, op. cit., 622. See Shana Alexander, op. cit. British Medical Journal (11 March 1967): 622-4. Ibid., 624. Another contributor writes in the same symposium: "The selection of the few Ito receive hemodialysis) is proving very difficult—a true 'Doctor's Dilemma'—for almost everybody would agree that this must be a medical decision, preferably reached by consultation among colleagues" (Dr F.M. Parsons, ibid., 623). "The Selection of Patients for Haemodialysis," op. cit. (note 10 above), 623. Dr Wilson's article concludes with the perplexing suggestion—wildly beside the point given the structure of the situation at issue—that "the final decision will be made by the patient." But this contention is only marginally more ludicrous than Parson's contention that in selecting patients for hemodialysis "gainful employment in a well chosen occupation is necessary to achieve the best results" since "only the minority wish to live on charity" (ibid.). To say this is of course not to deny that such questions of applied medical ethics will invariably involve a host of medical considerations; it is only to insist that extramedical considerations will also invariably be at issue. M.A. Wilson, "Selection of Patients for Haemodialysis," op. cit., 624. In the case of an ongoing treatment involving complex procedure and dietary and other mode-of-life restrictions—and chronic hemodialysis definitely falls into this category—the patient's psychological makeup, his willpower to "stick with it" in the face of substantial discouragements, will obviously also be a substantial

Just Caring: In Defense of Limited Age-Based Healthcare Rationing Leonard M. Fleck Health Law The debate around age-based healthcare rationing was precipitated by two books in the late 1980s, one by

21.

22.

23.

24.

factor here The man who gives up takes not his life alone but (figuratively speaking) also that of the person he replaced in the treatment schedule. To say that acceptable solutions can range over broad limits is not to say that there are no limits at all. It is an obviously intriguing and fundamental problem to raise the question of the factors that set these limits. This complex issue cannot be dealt with adequately here. Suffice it to say that considerations regarding precedent and peopl6 expectations, factors of social utility, and matters of fairness and sense of justice all come into play. One writer has mooted the suggestion that: "Perhaps the right thing to do, difficult as it may be to accept, is to select [for hemodialysis) from among the medical and psychologically qualified patients on a strictly random basis" (S. Gorovitz, 1966, "Ethics and the Allocation of Medical Resources," in Medical Research Engineering 5: 7). Outright random selection would, however, seem indefensible because of its refusal to give weight to considerations which, under the circumstances, deserve to be given weight. The proposed procedure of superimposing a certain degree of randomness upon the rational-choice criteria would seem to combine the advantages of the two without importing the worst defects of either. "Selection of Patients for Haemodialysis," op. cit., 623. The question of whether a patient for chronic treatment should ever be terminated from the program (say, if he contracts cancer) poses a variety of difficult ethical problems with which we need not at present concern ourselves. But it does seem plausible to take the (somewhat anti-utilitarian) view that a patient should not be terminated simply because a "better qualified" patient comes along later on. It would seem that a quasi-contractual relationship has been created through established expectations and reciprocal understandings, and that the situation is in this regard akin to that of the man who, having under-taken to sell his house to one buyer, cannot afterward unilaterally undo this arrangement to sell it to a higher bidder who "needs it worse" (thus maximizing the overall utility). I acknowledge with thanks the help of Miss Hazel Johnson, reference librarian at the University of Pittsburgh Library, in connection with the bibliography.

Daniel Callahan' and the other by Norman Daniels.' These books ignited a firestorm of criticism,3 best captured in the claim that any form of age-based healthcare rationing was fundamentally ageist, discriminatory in a morally objectionable sense. That is, the elderly had equal moral worth and an equal right to life as the nonelderly. If an elderly and nonelderly person each had essentially the same medical problem requiring the same medical treatment, then they had an equal right

Source: From Cambridge Quarterly of Healthcare Ethics 10, 1 (2010), 27-37. Reprinted with permission.

Y 111 II

II

11

4111

HILL1

7 Delivery of Health Care and Resource Allocation

to receive that treatment no matter what the cost of that treatment. Alternatively, if cost was an issue because the benefits of the treatment were too marginal, then both the elderly and nonelderly patients requiring that treatment ought to be denied it. If there were something absolutely scarce about the treatment, then some fair process would have to be used to make an allocation decision (and that fair process could not use some age cutoff among the allocation criteria). The Beginning of the Age-Based Rationing Controversy

Callahan and Daniels may have opened themselves up to these criticisms because their defense of age-based rationing was too broad arid unqualified. Callahan, for example, can be fairly read as saying that beyond some age (75-80) individuals ought to be denied very expensive life-prolonging medical care. He appealed to the notion of a "natural life span" as a reference point for justifying this admonition. Once individuals had had the opportunity to achieve what it was reasonable to achieve over a natural life span, they no longer had a strong moral claim to expensive life-prolonging resources, especially if such claims threatened the financial resources needed to afford younger members of society the same opportunities already enjoyed by the now elderly. Callahan's hope was that a cultural shift in social expectations would freely accomplish this result rather than coercive government policies. Callahan did not try to make subtle distinctions among life-threatening circumstances faced by the elderly. Two 80-year-olds with the same life-threatening cardiac condition and surgical need might have radically different prospects for long-term survival (1 year vs. 10 years), but neither would have a stronger claim to those medical resources because both had exceeded a natural life span. Daniels offered a very different (but very general) argument in defense of age-based rationing, the "prudential life span account." What he wanted to avoid was a situation in which the healthcare needs of the young were pitted against the healthcare needs of the elderly. He asks us to imagine ourselves (in a Rawlsianlike thought experiment) as prudent young adults who have to allocate limited healthcare resources across our entire life span without knowing what specific health problems we might encounter that would threaten us with premature death unless we had assured access to some expensive lifesaving intervention. What he reasonably assumes is that prudent individuals (in this state of healthcare ignorance) would want to maximize the likelihood of their achieving old age, and, consequently, they would reduce substantially the amount

of life-prolonging resources they would allocate to their future possible elderly selves (especially if only very marginal benefits could be achieved regarding length or quality of life) and allocate those resources to younger stages of their life in order to diminish their risks of premature death or disability. This strategy was intended to defuse the criticism that his views were ageist, but it is a very broad general argument. John Harris has introduced into the literature the "fair innings" argument.4 The general idea is that individuals who have achieved a normal life span have had their "fair innings," and, consequently, they ought not to demand extra innings, especially if making such demands denied younger individuals the opportunity to have their fair innings This broad argument has intuitive moral legitimacy for many.5 However, Harris himself is a critic of this argument (and its ageist implications) except in a couple of very narrowly framed circumstances 6 My Key Claim

In this essay I want to defend a "limited" form of agebased rationing. One implication of this claim is that I do not regard age-based rationing as intrinsically morally objectionable (unjust). There are some circumstances in which an age-based rationing decision would be unjust but other circumstances in which such a decision would be just, or at least "just enough." In other words, it will sometimes be true that it will be morally permissible to deny an elderly individual a costly life-prolonging intervention that we would be morally obligated to provide a nonelderly individual in those same medical circumstances, and the primary reason for that denial will be the age of that individual as opposed to the probability of success of that intervention, or degree of effectiveness, or comorbidities, or any other morally and medically relevant considerations. If age-based healthcare rationing has such a morally obnoxious quality about it (but we need to find effective ways to control escalating healthcare costs), why should we not identify other approaches to rationing justly that avoid using any age-based reference points? The short answer I would offer to that question is that we would have to tolerate significant injustices in our healthcare system if we avoided altogether some forms of age-based rationing. Social Origins of the Age-Based Rationing Problem

In brief, here are the social, political, and economic circumstances that prompted Callahan and Daniels

Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing

to write their books in the late 1980s. First, the current elderly population in the United States represents about 13% of the overall population, but approximately 35% of all health dollars spent in the United States are devoted to meeting the healthcare needs of the elderly. The statistical fact is that, as things are now, the elderly consume a hugely disproportional share of health resources compared to the nonelderly. Second, in the United States and other advanced nations in the West, we are just beginning to experience the post—World War II "baby boom" reaching retirement age. In the United States this is a cohort of about 78 million individuals who will raise the fraction of our population over age 65 to about 21% in the year 2030.7 If that future generation were to consume healthcare resources at the same level as the current generation of the elderly, we would have extraordinarily difficult moral and political and economic challenges in meeting those demands without sacrificing the just claims (of many sorts) of the nonelderly. But, third, it is very unlikely that those future demands would be so restrained because of the continued proliferation of increasingly expensive medical technology throughout our healthcare system (as well as the healthcare systems of other advanced nations).8 The likelihood is that per capita healthcare spending on the elderly will increase very significantly and make even more difficult the resulting moral and political problems. In 1993 the Medicare program for the elderly (which only covers about half the healthcare expenses of the elderly) spent $150 billion. In 2009 the projections are for $503 billion, and in 2018 the projections are for $932 billion (still 12 years out from having the entire baby boom generation reach age 65).9 The conclusion I would draw from these three points is that whatever the data were that prompted Callahan and Daniels to propose age-based rationing in the late 1980s, those data have become substantially worse since then, thereby justifying examining age-based rationing again.

Medicare and Its Unintended Consequences There were good reasons for creating the Medicare program in the United States in 1965. Relatively few of the elderly then could afford to purchase health insurance from private insurance companies after they had retired (if insurance would even be offered to them) because they were seen as being at high risk for serious health problems. This was properly seen as a justice issue corrected by the creation of the Medicare program. In short, what we created was national health insurance for the elderly. At the time this did not represent

a threat to the just healthcare claims of the nonelderly because so little of the costly high-tech medicine we take for granted today had been invented or disseminated. But that dissemination has taken place over the past 40 years. The overall result is that the elderly are living longer with a greater burden of chronic illness that requires constant costly medical assessment and management.11) The Medicare and Medicaid programs (the latter covering long-term care for the impoverished elderly) represent public expenditures, which are adding to the current and projected deficit problems of the United States. As I write we are in the middle of wrestling with health reform, the hope being that virtually all the currently uninsured in the United States (47 million) would be provided with a substantial package of health benefits, equal at least to what we provide Medicare beneficiaries. But the total cost of the package to the federal government is now estimated at $1.2 trillion over a 10-year period. It is unthinkable that these costs would simply be added to the federal deficit. That means savings have to be achieved elsewhere (read: Medicare) from other federal healthcare expenditures in order to cover at least part of those costs. Critics will be prompted to ask whether that is fair to the elderly. The obvious retort is to ask whether the status quo (47 million uninsured) is fair to the nonelderly. To a large extent the elderly have access to whatever the healthcare system has available for meeting their health problems, no matter how costly, no matter how marginally beneficial." The relatively well-off elderly can demand these extraordinarily expensive cancer drugs that yield only extra weeks or months of life whereas the uninsured nonelderly with early-stage cancers will have no assured access to less expensive, more effective cancer treatments that can either save their lives or prolong them for very long periods of time. Many of these individuals will unnecessarily die prematurely.' A number of recent medical journal articles have been discussing a relatively dramatic increase in the number of individuals in their 80s and 90s who are having complicated surgical procedures for their heart problems." The logic behind these efforts is nicely captured by Shim: Together, the hope of medical intervention, the difficulty of prognostication and the actual gains of life that a great many cardiac procedures confer, contribute to the ethical viability and imperative of treating risk. For practitioners and patients alike, understandings of the mandate to treat at ever-older ages contribute, pragmatically, to the elimination of any significant deliberation about whether or not to treat. Instead, standard practice replaces choice."

II

JII

111111

7 Delivery of Health Care and Resource Allocation

The point is that the availability of the technology for prolonging life in the very old, together with assured funding for its use in the very old, and the perception of a likely net benefit create a clinical and (unreflective) moral imperative for its use that undermines our society's capacity (and willingness) to meet the just claims of the uninsured nonelderly. These are the clinical and social facts that generate a need for a renewed moral appraisal of the legitimacy of some forms of age-based healthcare rationing. The alternative is the continued toleration of the multiple, diffuse, invisible injustices that are inflicted on the nonelderly regarding their healthcare needs.' Arguments against Age-Based Rationing

Let us briefly rehearse the major moral arguments that speak against the legitimacy of any form of age-based rationing. Baruch Brody cor tends that allowing any elderly individual to die when we have the capacity to prolong that individual's life (and that individual wants that intervention) represents a failure to fulfill our "duty to rescue."16 He gives special moral weight to the fact that we are talking about identifiable individuals who are patients of physicians (as opposed to statistical lives that might be sacrificed as a result of a social nonfunding decision). We will spend millions of dollars to try to save the life of some hapless hiker in the Arctic wilderness, and, therefore, according to Brody, we have at least as strong an obligation to save older patients who are conveniently at hand in hospital beds. That duty to rescue does not have an age cutoff, which would represent a failure to accord equal respect to older individuals whose lives were in peril. Someone with a Rawlsian bent might be inclined to argue that the terminally ill elderly are among the "medically least well-off," and consequently, justice requires that they be provided with more in the way of medical resources, not less (as would be characteristic of age-based rationing decisions). Again, with reference to the terminally ill elderly someone could invoke a "last chance therapy" argument: It is one thing to insist that someone with a life-threatening medical need accept a less expensive effective intervention for their medical problem rather than a more expensive medical intervention that might be only marginally better. But when someone has only one therapeutic alternative or death, then we are morally obligated to provide that alternative no matter what the cost. The suppressed premise in this argument is that human life is priceless (no matter what the age of a person), and, hence, it is morally wrong to deny individuals with a terminal cancer these cancer drugs

that might cost $100,000 for a course of treatment that could yield only a few extra months of life. Yet another argument invoked in these circumstances is that denying the terminally ill elderly life-prolonging medical resources is just a form of involuntary euthanasia. For those who propound this line of argument the offering of hospice instead of the life-prolonging care these patients want is simply a more socially genteel form of involuntary euthanasia. Finally, John Harris defends what he calls an "age indifference" principle as a reference point for condemning virtually all forms of age-based rationing. He writes, "An individual's entitlement to the concern, respect, and protection of the community, as expressed in its public health care system does not vary with age or life expectancy. I believe this principle is the only one consistent with justice." Harris then adds as a supportive argument, "Those who believe in discriminating in favor of the young or against the old must believe that in so far as murder is an injustice, it is less of an injustice to murder the old than the young." But he adds that our common law tradition does not see things that way. Murder is just as bad, whether an individual is denied 40 additional years of life or a single hour of life.'' In Defense of Limited Age-Based Rationing

What we need to notice with regard to the above arguments is that they are perfectly general. They are not aimed at any particular types or circumstances in which age-based rationing might be considered. All age-based rationing is judged to be fundamentally flawed. Here, however, are some examples that ought to precipitate some critical reflection. We should start by taking note of the fact that subtle age rationing is already part of accepted (and morally defensible) medical practice. We typically limit lots of screening tests by age, such as prostate cancer screening. We stop doing screening mammograms beyond age 70. We focus certain public health programs, such as antismoking campaigns, on the relatively young, believing (correctly) that we are much more likely to achieve substantial positive results there than in someone with a 30-year habit of smoking.18 We use 10-year duration hip prostheses in patients over age 80 (as opposed to much more costly longer duration prostheses) when it is unlikely those patients will survive more than 10 years due to assorted comorbidities. This is morally warranted so long as the resources saved are reallocated to meet higher priority healthcare needs of the elderly. Still, my critic might argue that these examples are not persuasive enough because they

Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing

not weighty enough; age-based rationing matters most when life itself is at stake. So we shall consider some of those cases. are

The DeBakey Case

First, consider the case of Dr. Michael DeBakey, the famed cardiac surgeon who pioneered a number of important surgical procedures. He diagnosed himself with an aortic aneurysm at age 98. He knew that in theory it was surgically repairable, but he feared the brain injury such surgery could occasion. His family and colleagues convinced him to have the surgery. The surgery was successful, but Dr. DeBakey spent eight months in the hospital at a cost of more than $1 million (and he survived for a year after that). Our question is: Would Dr. DeBakey have been treated unjustly if he had been denied that surgery, largely because of his advanced age and the likely excess costs associated with a prolonged hospitalization? Several of the critical arguments advanced above could be invoked to justify providing Dr. DeBakey with the surgery. That is, someone could argue there was a duty to rescue, or that this was a last chance intervention, or that Dr. DeBakey was among the medically least well. However, the very generality of each of these rationales has the morally problematic implication of justifying virtually every costly marginally beneficial lifeprolonging intervention for every patient on Medicare, thereby adding to the problem of escalating health costs and diminishing the likelihood of health reform that would provide health insurance for America's uninsured. Invoking the "duty to rescue" argument makes moral sense when the numbers are small and the circumstances rare, even if the costs per rescue are extraordinarily high. But in the world of life-prolonging medical technology we have today, the opportunities for rescue are ubiquitous among the elderly, and the likelihood is that a majority of Medicare recipients would need to be rescued at great expense several times before disease overmastered them. This is a radically different situation from the "usual" circumstances in which we invoke the "duty to rescue." The command of public resources by the Medicare elderly, especially under this alleged ethical obligation, is very much a threat to the just claims of the uninsured and marginally insured nonelderly. A similar analysis will apply if the ethical argument involves an appeal to last chance therapies or to the status of being medically least well-off. There is a range of circumstances in which these moral considerations will be legitimate and determinative of a just outcome. For example, a relatively small number of children are born each year in the United States with either Gaucher's

or Hunter's syndrome. These are metabolic disorders that will result in very premature death. Today we have the drugs imiglucerase and idursulfase, each of which costs $300,000 per person per year. But they are effective in preventing very premature death for these young afflicted individuals. These are last chance therapies; there are no alternatives. Further, these children can be regarded as being among the medically least well-off. What makes the critical moral difference in these cases is that these interventions are very effective in sustaining both length and quality of life, and these individuals are very far from having had their fair innings. The same cannot be said for Dr. DeBakey. My critic might respond that this is not really a matter of age-rationing rather, what matters is whether a life-sustaining intervention is going to be sufficiently effective in a very old individual. So the proverbial vigorous elderly individual (Hazel Homer [age 99] in note 13) who gets 5 extra years of life from an expensive cardiac intervention has a just claim because that intervention proved to be so effective. However, I would argue that is too quick a moral judgment. What can be argued instead is that there are compelling considerations of compassion that would warrant giving Hazel her bimodal cardiac resynchronization device at $60,000. Compassion must not be confused with justice, nor will compassion trump justice if a compassionate response to the health needs of the very old threatens to fail to meet the just health claims of the relatively young. We can illustrate this last point with our second critical example. The Case of the Artificial Heart

In the United States we are currently clinically testing totally implantable artificial hearts [Timis]. These devices may be regarded as last chance therapies for patients in end-stage heart failure. Each year in the United States 550,000 individuals enter early-stage heart failure. It is expected that improved models of this device will yield 5 extra years of life expectancy on average for patients who would otherwise die of heart failure. Computer modeling suggests that we could implant as many as 350,000 of these devices per year at a cost of $300,000 each. That would add $105 billion to the cost of healthcare in the United States, roughly 70% of which would be added to the cost of Medicare. About 100,000 of these devices would be needed by those below age 70. If the societal decision were made that we could afford only 100,000 of these devices, do we want to argue that they ought to be distributed by lottery over the elderly and nonelderly (so long as minimal criteria for likelihood of success are met)? Or would we have good reasons for saying that the nonelderly had

403

N NI

ill

IIILL1

7 Delivery of Health Care and Resource Allocation

the stronger just claims because they had not yet had their fair innings?" Should we be morally comfortable with accepting Harris's age indifference principle as justification for using a lottery to distribute access to TIAHs? If we were, then a 40-year-old in heart failure facing death in 6 months with a prospect of 20-plus years survival with a TIAH would be entitled to no more a chance for that TIAH than a 90-year-old in heart failure with only a 2-year prospective gain in life expectancy. This does not look like a morally plausible result that most of us would be comfortable endorsing. We must ask ourselves how the 90-year-old managed to achieve that age. He might have gotten there because substantial and costly healthcare resources got him there, the very sorts of resources that the 40-yearold might now be denied. Alternately he might have simply been the beneficiary of an excellent genetic endowment (whereas our 40-year-old was the unfortunate recipient of a deficient genetic endowment). From the perspective of healthcare justice, what would justify society conferring additional life-years on the fortunate 90-year-old at the expense of life-years lost by the 40-yearold? Neither egalitarian nor utilitarian considerations would warrant such an allocation. Harris's view is that both such individuals equally desire to experience "the rest of their lives" and this is what requires equal moral respect. But that expression of desire looks more like plain self-interest than any consideration having substantial moral weight. If we perform a Rawlsian-like mental experiment with two 20-year-olds and tell them that one of them will have the fate of our 90-year-old whereas the other will have the fate of our 40-year-old but we must decide now which of those two future possible selves will have assured access to the TIAH, I am confident that the impartial answer that would emerge would be that the 40-year-old gets the TIAH. Harris is correct that there are numerous rationing scenarios in which age indifference would be a moral requirement. But that does not imply that age indifference is required in every rationing scenario. Implantable cardiac defibrillators (Icps) cost $40,000 each. They are intended to prevent sudden death from an arrhythmia In the United States we may be implanting 200,000 per year. However, 8].% of these devices never fire over a 5-year period, at which point batteries need to be replaced at a cost of $20,000. There are now tests that can determine with 98.7% accuracy which patients will likely not suffer an arrhythmic event over the next 2 years. We could reduce by 70,000 the number of devices implanted and save $3 billion per year, but there would have been 800 (in theory) preventable deaths as a trade-off for saving those dollars. In this scenario

age indifference would be a moral requirement. The 60-year-old and the 80-year-old would both be justly denied an ICD if the relevant medical test indicated it was extremely unlikely the ICD would make a difference. In both cases the marginal benefit is too small to justify the cost for either individual. Higher priority health needs of both the elderly and nonelderly could be better served by the achieved savings. Harris invokes the analogy with murder as a basis for rejecting all age-based rationing that involves loss of potential life-years. However, this is an inapt analogy. The murderer is violating the liberty rights of his victim. His victim would have gone on living happily but for his murderous intervention. His victim does not require any expensive social resources for continued survival. No question of distributive justice is raised in the case of murder. But that is precisely the moral issue at stake with respect to the problem of healthcare rationing. We have only limited resources to meet virtually unlimited healthcare needs. What criteria may we use to determine which needs are fairly met or unmet? If we spend $300,000 to give a child with Gaucher's an extra year of life, are we equally morally obligated to spend that same amount to give an 80-yearold an extra week of life, or an extra day of life, as Harris's use of the murder analogy seems to imply? This is neither just nor reasonable. The loss of that extra week of life is rightly regarded as being unfortunate, not unjust. There are numerous reasonable and morally relevant considerations that might be invoked for making fair healthcare rationing decisions at the level of social policy. These considerations will often need to be balanced against one another in complex circumstances. We have no reason to believe there exists one "perfectly just" balancing that could be achieved through human reason in all these circumstances.2° Numerous options for allocating artificial hearts can be reasonable and "just enough." But it would be unwise and unfair if such allocation decisions were left to the moral intuitions of tens of thousands of individual physicians, all vulnerable to the normal range of personal biases. Rationing and Rational Democratic Deliberation

The better option, as I have argued at length in a recent book,21 would be a public process of rational democratic deliberation. This would not be a Rawlsian mental experiment; this is a real world option. The vast majority of us in middle age are completely ignorant of our future possible health needs, and in that respect we are capable of the moral impartiality required by these choices. Further, we are mindful of what we regard as reasonable financial expenditures of our money (taxes) for healthcare.

Fleck: Just Caring: In Defense of Limited Age-Based Healthcare Rationing

That creates the budgets (limits) within which we can collectively set healthcare priorities and determine what rationing protocols we are willing to impose upon our future possible selves that we judge to be "fair enough." If we decide that we are willing to pay for only 100,000 artificial hearts and that these should not be available at social expense to those over age 70, we have made an age-based rationing decision. But it is not a decision we are imposing on others we deem unworthy of equal moral respect. And it is a judgment we make with the understanding that this represents a last chance therapy for a future possible version of myself. We recognize that we are telling future caregivers that we have freely given up a right to be rescued with an artificial heart at age 77 because we have chosen, in effect, to allocate limited and expensive life-prolonging healthcare resources to earlier stages in our life when they would do much more good and be judged to be much more valuable to us. Such judgments represent fair terms of cooperation, commendable moral reciprocity. They cannot properly be described as nonvoluntary forms of euthanasia." If we collectively decided, we could pick age 75 instead of age 70, or we could choose age 80 as an absolute limit for an artificial heart at social expense. We could choose not to use any age at all, relying instead upon some complex mix of clinical criteria related to probability of survival for some number of years. But that would create considerable fuzziness and uncertainty (and consequent risk to fairness) in actual practice. This is what Callahan refers to as the "ragged edge" problem.23 The virtue of a definite age limit in this context is that it creates a bright line that might better protect fairness. What a fair deliberative process allows us to say at the very least is that choosing a specific age beyond which individuals would not have a claim to an artificial heart at social expense is not unjust. Another circumstance in which we may have to invoke some form of age-based rationing would be a

serious influenza pandemic, comparable in severity to the 1918 Spanish flu that may have killed 50 million people. If we did not have enough respirators to save the lives of all who would otherwise be virtually certain to die, then it would be morally defensible (not unjust) if we were to deny respirators to those over the age of 70. The substantive moral justification for such a choice would be the fair innings argument. The procedural moral justification would be a rational democratic deliberative process that endorsed such a policy, even if the endorsement was as the least unjust option available. We might wonder how this could be fair to those who are currently in their 60s or 70s. I will mention that I am 65 years old, but I would certainly endorse this policy if any alternative arrangement put my children at risk of being denied a life-saving respirator (say, as a result of a lottery process). I suspect this same commitment would be endorsed by the vast majority of my fellow nearly senior citizens. Conclusion

What we are defending in this essay is the need for surgically precise age-based rationing protocols that have been fairly and democratically legitimated in a range of rationing circumstances where an age-based criterion would yield a more just outcome. We are not denying that the greater health needs of the elderly justly command more social resources to meet those needs. But we are saying that there must be just limits to meeting those needs. We must also note the morally relevant point that some of the savings that result from agebased rationing protocols redound to the benefit of the elderly in the form of health services that yield more health good for them at a lower cost. Purchasing very expensive marginally beneficial life-prolonging healthcare can adversely affect the just claims of the elderly to needed healthcare as much as the just claims of the nonelderly. This last point applies as much to the US healthcare system as to European healthcare systems.

Notes 1. 2.

3.

Callahan D. Setting Limits: Medical Goals in an Aging Society. New York: Simon and Schuster; 1987. Daniels N. Am I My Parents' Keeper? An Essay on Justice between the Young and the Old. New York: Oxford University Press; 1988. This could be a very long list, but we will simply list more prominent critiques. See Binstock RH, Post SG, eds. Too Old for Health Care? Controversies in Medicine, Law, Economics, and Ethics. Baltimore, MD: Johns Hopkins University Press; 1991; Moody H. Aging: Concepts and Controversies. Thousand Oaks, CA: Pine

4. 5.

Forge Press; 2000; Moody H. Ethics in an Aging Society. Baltimore, MD: Johns Hopkins University Press; 1992; Kilner J. Life on the Line: Ethics, Aging, Ending Patients' Lives, and Allocating Vital Resources. Grand Rapids, MI: W.B. Eerdmans Publishing Co.; 1992; Smeeding T. Should Medical Care Be Rationed by Age? Totowa, NJ: Rowman and Littlefield; 1987. Harris J. The Value of Life. London: Routledge and Kegan Paul; 1985. Williams A. Intergeneration equity: An exploration of the "fair innings" argument. Health Economics 1997;6:117-32.

■Y

1

I

II

.111

111111

7 Delivery of Health Care and Resource Allocation

6. Harris J. Cardiac surgery in the elderly. Heart 1999;82:119-20. 7. Europe is facing a comparable problem with the aging out of its population. "In Europe, predictions from the Statistical Office of the European communities (EUROSTAT) estimate that from 2010 to 2030 the population of octogenarians in 25 countries of the European Union will increase by 57%. By 2050, the population of those aged 80 and older will increase by 180%." See Cvachovec K. Coronary artery bypass sargery in the very old: Light at the end of the tunnel or a dead-end road? Journal of Cardiothoracic and Vascular Anesthesia 2007;21:781-3 at pp. 781-2. 8. As numerous policy analysts have noted, advancing medical technologies are the primary driver of the problem of escalating healthcare costs. See Bodenheimer T. High and rising health care costs. Part 1: Seeking an explanation. Annals of Internal Medicine 2005;142:847-54; Bodenheimer T. High and rising health care costs. Part 2: Technologic innovation. Annals of Internal Medicine 2005;142:932-7. See also Aaron HJ, Schwartz WB, Cox M. Can We Say No? The Chal'enge of Rationing Health Care. Washington, DC: Brookings Institute; 2005. In 1993 in the United States we were spending $912 billion on healthcare. The estimate for 2009 is that we will spend $2.5 trillion on healthcare and $4.4 trillion in 2018 (or about 20.3% of our projected Gross Domestic Product at that time). See Sisko A, Truffer C, Smith S, Keehan S, Cylus J, Poisal J, et al. Health spending projections through 2018: Recession effects add uncertainty to the outlook. Health Affairs 2009;28:w346-57. 9. See Note 8, Sisko et al. 2009. 10. Almost 82% of the elderly have at least one chronic condition, 65% have at least two chronic conditions, and 25% have four or more chronic conditions. The reader should think of heart disease or cancer or lung diseases or kidney disease or diabetes or stroke or dementias or arthritis or sensory deficits, and so forth. Although many of these conditions are ultimately fatal, the period of time for which individuals can survive has increased dramatically as a result of costly contemporary medicine. This is reflected in the Medicare cost statistics cited earlier. See Wolff J, Starfield B, Anderson G. Prevalence, expenditures and complications of multiple chronic conditions in the elderly. Archives of Internal Medicine 2002;162:2269-76. 11. This sentence needs a qualification because the Medicare program has significant copayments and deductibles. Those serve as an effective barrier for the elderly in roughly the lower half of the income spectrum. They cannot demand everything, no matter what the cost. More economically well-off elderly have supplementary insurance that eliminates those barriers and allows them to demand whatever they see as medically beneficial for themselves. 12. The most often cited estimate of that annual number in the United States is 22,000 premature deaths attributable to being uninsured. See Dorn S. Uninsured and dying because

of it Updating the Institute of Medicine analysis on the impact of uninsurance on mortality; 2008 Jan; available at httpl/www.urban.orgiurl.cfm?1D5411588&renderforprint=1 (last accessed 19 July 2009). 13. See, for example, Huber CH, Goeber V, Berdat P, Carrel T, Eckstein E Benefits of cardiac surgery in octogenarians-A postoperative quality of life assessment. European Journal of Cardio-Thoracic Surgery 2007;31:1099-105; Filsoufi F, Rahmanian PB, Castillo JG, Chikwe J, Silvay G, Adams DH. Results and predictors of early and late outcomes of coronary artery bypass graft surgery in octogenarians.

Journal of Cardiothoracic and Vascular Anesthesia 2007;21:784-92. See also the case of Hazel Homer who, at age 99, received an advanced pacemaker and defibrillator to assist a failing heart. She is alive today at 104; Hartocollis A. Rise seen in medical efforts to treat the very old. New York Times 2008 Jul 18. 14. Shim JK, Russ AJ, Kaufman SR. Risk, life extension and the pursuit of medical possibility. Sociology of Health and Illness 2006;28:479-502, at 496. 15. For a profoundly effective understanding of what it means to be uninsured, the following Kaiser Family Foundation website ought to be visited for in-depth interviews with a number of such families. These are the kinds of injustices that are essentially morally invisible at a broader social level. See Shirk M. In Their Own Words:

The Uninsured Talk about Living without Health Insurance; available at httpl/www.kff.org/uninsured/2207-index.cfm (last accessed 19 July 2009). 16. Brody B. The macro-allocation of health care resources. In Sass HM, Massey R, eds. Health Care Systems: Moral Conflicts in European and American Public Policy. Dordrecht: Kluwer Academic Publishers; 1988:213-36. 17. See note 6, Harris 1999:119. 18. See the fuller exposition of this argument in Dey I, Fraser N. Age-based rationing in the allocation of health care. Journal of Aging and Health 2000;12:511-37. 19. It should be noted that this particular issue will be just as much an issue in Europe as in the United States. Simply having a universal healthcare system does not address this issue as a justice issue. 20. This is what John Rawls refers to as the "burdens of judgment." Our moral arguments and moral theories will often not be powerful enough in complex circumstances to yield a uniquely correct moral response. See his Political Liberalism. New York: Columbia University Press; 1993. 21. See my book Just Caring: Health Care Rationing and Democratic Deliberation. New York: Oxford University Press; 2009, especially Chapter 5, for a fuller exposition of the role of rational democratic deliberation in yielding just healthcare rationing policies. 22. I address this issue more comprehensively in an earlier essay: Just caring: Assisted suicide and health care rationing. University of Detroit Mercy Law Review 1995;72:873-99. 23. See Callahan D. What Kind of Life: The Limits of Medical Progress. New York: Simon and Schuster; 1990, especially Chapter 2.

Kishore: Human Organs, Scarcities, and Sale: Morality Revisited

7.5 Increasing Resources

Human Organs, Scarcities, and Sale: Morality Revisited R.R. Kishore

The recent exposure of an international racket in organ trafficking, extending from Brazil to South Africa, has prompted me to write this paper.' Kidney vendor Alberty Jose da Silva and the American woman who bought the organ were both in the same boat. To many the process may sound iniquitous and even sinful but, in fact, it is fair and natural and is consistent with normal human behaviour. The shortage of available organs is a global feature of organ transplantation and has been a challenge almost since its inception. In the US, as at 10 July 2004, 86,173 people are on the nation's organ transplant waiting list and on average 17 patients die every day while awaiting an organ—one person every 85 minutes. On average, 115 people are added to the nation's organ transplant waiting list each day—one every 13 minutes. In 2001, 6,251 individuals died on the US organ transplant waiting list because the organ they needed was not donated in time.2 The situation is no better in Europe. In certain countries of the Eurotransplant area—Austria, Belgium, Luxembourg, Germany, the Netherlands, and Slovenia—as at 1 July 2004, there were 15,585 people on the waiting list. In the year 2002, 12,644 patients were on the waiting list for kidneys but only 3,043 could get an organ. In the year 2003, in the Netherlands and Germany, 1,182 and 9,479 patients, respectively were on the waiting list for kidney transplantation but only 406 and 2,111 transplantations could be conducted.3 In India there is no database and reliable studies are yet to be conducted but in view of the country's large population of nearly 1.2 billion and the rising incidence of end stage renal disease (ESRD), the requirement for kidney transplantation alone is expected to be around 80,000 per year but not even 5,000 transplants are conducted. Strategies such as liberalization of the concept of brain stem death, introduction of presumed consent, routine harvesting, required request, mandated choice, raising the donor's upper age limit, relaxation of restrictions imposed on donations among family members, and

allowing altruistic donations from strangers have not resolved the problem. Organ scarcity continues to prevail, leading to inequitable therapeutic dispensation, escalating costs, trade, crime, and premature death. In India there are periodic reports of organ trafficking involving clinicians, managers of clinical centres, middle men, and even state officials; several cases are at present under active investigation or at trial. The "worldwide shortage of kidneys from cadavers has resulted in illicit organ sales and even kidnapping and murder of children and adults to 'harvest' their organs."4 Millions of people are suffering, not because the organs are not available, but because "morality" does not allow them to have access to the organs. The question is what is good and what is bad. How are we to measure the moral content of a particular act? Morality is always contextual. It depends on how and in what context we interpret values. The famous Roman physician Galen—for example, did most of his anatomy research on pigs and dogs as it was regarded as immoral to dissect humans at that time,' but, subsequently, dissection of human cadavers during medical education became a routine practice. The issue of biotechnological achievements and their social assimilation contemplates a much deeper dialogue than is being conducted in contemporary ethical discussions. This paper is an attempt in that direction. Essential Questions and Values

The basic ethical principle involved in organ transplantation is whether a person has a right to enjoy life on the basis of organs belonging to others. Once we choose to answer this question in the affirmative we concede that we are prepared to inflict harm on others in order to improve our health or to prolong our life. Thus we sacrifice the long cherished principle of nonmaleficence in medicine. Whether the organs come through donation, gift, or sale is a matter of individual choice and circumstances. Even if a person gives his organ willingly and without any thought as to recompense he suffers harm to his body. It is a fact that in every gift or donation some kind of expectation is involved, though it may not be a material consideration. In the case of live organ donors

Source: From Journal of Medical Ethics: The Journal of the Institute of Medical Ethics 31, 6 (2005), 362-5. Copyright © 2005 BM J Publishing Group Ltd. Reprinted with permission.

II AI

IIIIJJ

7 Delivery of Health Care and Resource Allocation

the organ is donated to a particular person who, in fact, may not be the neediest or the most deserving bearing in mind the seriousness of illness, period of waiting, age, family circumstances, capability to afford posttransplantation therapy, and other criteria. This means that the act of donation is tainted with considerations of personal relationship, choice, and preference. In other words the donation is not a candid act of altruism or human solidarity, but rather :.s motivated by the desire to save the life of a near and dear one, which may, at times, be to secure one's own comfort and future. Such urges and motives also consnute considerations other than altruism since they are aimed at pleasure and fulfillment. Even a donation made to a stranger is not without considerations of possible benefit. Such cases may be motivated by the desire to discharge a religious duty, to correct a wrong done in the past, to gain mental or moral satisfaction, or to be seen as a good Samaritan. Once the practice of organ donation by the genetically related and also by strangers, based on altruism, has been accepted as ethically sound, the following components of organ removal stand morally vindicated: a.

a person's expectation to enjoy life with the help of organs belonging to others is valid, and b. the breach of a donor's bodily integrity and the consequent harms are permissible.

Judged on these values a person's act of severing his/her organ in order to liberate a fellow being from a terminal illness or to save his/her life cannot be dubbed as immoral simply because the act is accompanied by a reasonable material consideration. "When a person sells an organ he or she acts both selfishly, in advantaging him or herself, and altruistically, in contributing to a public good."6 The presence of considerations is not a sufficient reason to transform a simple act into a sin. Otherwise, selling water to the thirsty would be an equally big sin—in fact, rather a bigger one. To equate an organ vendor with a criminal committing a heinous crime, as is reflected in the legislative strategies of many jurisdictions, is misconceived. Several jurisdictions provide stringent punishment for organ sale in utter disregard of the circumstances that compel a person to sell her/his organ. The punishment ranges from three months' imprisonment and/or a fine, as in the UK,7 to eight years' imprisonment and/or a fine, as in Venezuela.8 In India the punishment may extend up to seven years' imprisonment and/or a fine up to Rs20,000.9 Such punishments are prescribed only for serious offences, and it is thus clear that organ sale is treated as a serious offence, worldwide. Such an approach does not seem to be correct. A person who

sells his organ does so because he knows that his organ is going to save the life of a fellow human being and as such he is convinced that he is not doing anything immoral or inhuman. Had he known that his act would lead to loss of life, property, or inflict any other kind of injury on the buyer he would not have sold his organ. A criminal has no such moral conviction or justification and commits the act solely for his personal gain, without caring for the loss or injury suffered by the victim. As such, it seems that the legislative strategies in the area of organ transplantation have not been realistic. Arguments against Financial Incentive and Sale

Policies on organ transplantation reflect a unique social paternalism. Objections against the sale of organs such as "(1) the dilution of altruism in society, (2) the risk that the quality of donated organs would decrease; (3) doubts about the voluntariness of those who accept financial incentives for donation, and (4) the treatment of human beings and their parts as commodities' do not reflect an objective approach. Recent critics of markets in organs give two main reasons to support their opposition: (1) "the integrity of the human body should never be subject to trade," and (2) a system is unethical "when it penalises the weakest people and exacerbates discrimination based on census" and generates "the risk of exploitation of vulnerable donors."" Some are more skeptical and feel that the "poor of the developing world could become a vast reservoir" of organs for the developed worldu and that the poor in a "starved country" can never be "fully informed and autonomous donors.' Others feel that it amounts to "exploitation of potential donors."H Arguments against organ sale are thus grounded in two broad considerations: (1) sale is contrary to human dignity, and (2) sale violates equity. Let me examine these one by one. Is Sale Contrary to Human Dignity?

In contemporary ethical deliberations human dignity has become a very handy tool to measure the ethical content of biotechnological applications, at times, without appreciating its true nature, ambit, and implications. It is not within the scope of this paper to deal with human dignity in its entirety but it may, however, be worthwhile to know what it means in essence. Essentially speaking, human dignity is an expression of the human content of Homo sapiens. It is an expression of the properties or virtues due to which a human creature is known as a human being. These are the characteristic or attributes that are unique to the human race and not possessed by any other living form. What

Kishore: Human Organs, Scarcities, and Sale: Morality Revisited

are these virtues? These virtues, known in Vedic thought as dharma, are ten in number—namely, love, trust, righteousness, compassion, tolerance, fairness, forgiveness, beneficence, sacrifice, and concern for the weak With these human virtues in mind, any act done to save the life of a human being or to liberate him from suffering cannot be construed as contrary to human dignity. The presence of a consideration does not alter the basic content of an act such as an organ sale, which is grounded in the need to save at least two human lives, one from terminal illness (the recipient), the other from hunger (the donor). The concept of human dignity does not demand that people should be forced to die a premature death where an illness can be cured nor that people who donate organs should die of hunger and their families be left to starve. Rather, it will be contrary to human dignity to promote such an act. The matter of payment is a logistical dimension, not the substantial aspect of the transaction. Retrieval of organs from the dead by presuming consent on their behalf or the act of declaring a person brain stem dead in order to remove organs from his body are devices designed solely to augment organ supply but they are not regarded as contrary to human dignity. Prohibition on sale of organs makes matters worse by restricting transparency, fairness, and choice and by generating arbitrariness, fear, and bribes. Vendor and buyer are rendered vulnerable because of the introduction of an unwarranted legal component that brings in many players such as police, lawyers, adjudicators, and social activists, each with their own philosophy and interest, thereby transforming a simple activity into a highly complex exercise. The argument that there cannot be genuine and free consent to the sale of organs is not well founded. In fact such an argument is an antithesis of the concept of autonomy. The decision to sell an organ, taken by a person after considering all circumstances, consequences, options, and possibilities, cannot be disregarded by others on the ground that it has been taken under undue influence or inducement. The individual is the best judge as to what is best for him in a given situation and so long as his decision does not affect others he cannot be stopped from acting upon his decisions. If the vendor is not able to give free and informed consent because of the pressures of poverty and the lure of money, the buyer is also not able to give such consent because of the pressures of illness and the urgent need to save his life. The donor too is unable to give free and informed consent because of the fear of losing a near relative and possibly the consequent loss of support and security. Thus none of the parties involved is capable of giving free and informed consent because of the compelling circumstances in which they find

themselves. This means that the whole exercise of organ transplantation is inconsistent with the principle of free and informed consent and therefore is unethical. Why then should only the vendor be dedared an offender? Arguments linking a person's autonomy to bodily "integrity" or "fullness" and on this basis declaring organ sale as "misuse" of "our autonomy"15 appear misconceived because if that is the case "integrity" or "fullness" is also breached in the case of donation, which is not considered to be unethical. In order to justify the failure to provide organs to the needy, various considerations, such as old age, associated diseases, poor prognosis, and irresponsible behaviour, are brought up in order to exclude a number of potential organ recipients, knowing fully that in such cases "the alternative to transplantation is death."16 Organ scarcity has polluted the moral concepts involved in transplantation and some feel "alcoholics should be given lower priority for a liver because of their moral vice of heavy drinking"17 and "lung transplantations" should not be offered "to people who smoke or have other substance abuse in the last six months."18 Despite the fact that brain stem death and human death are not the same, the definition of death has been liberalized in order to give an "incessant push to expand the pool of potential organ donors."19 Thus, the prohibition on sale is not without heavy costs, which include untimely death, poor quality of life, higher disease burden, moral bias, and premature certification of death. Is this consistent with human dignity? It is significant that the concept of human dignity is being selectively applied in the case of certain tissues only. Blood, bone marrow, sperm, and eggs are being openly sold and a woman can "command $50,000 for her donated eggs" (Kahn, 14: p. 1) but their sale, it seems, does not attract notion of human dignity. Is Sale Violative of Equity?

(1) The apprehension that organs will become costly, going beyond the reach of the common man is unfounded. Costs can always be controlled by the state through the use of regulatory mechanisms, as is done in the case of other goods and services. Furthermore, socioeconomic inequalities are present in all walks of life, not only in organ transplantation. The whole healthcare system is subject to market forces. Many drugs, many pieces of equipment, appliances, procedures, and services are prohibitively costly and are not accessible to all those who need them. In many jurisdictions, they have to be imported from the developed and industrialized countries, which, at times, monopolize their trade. Despite strong objections from the public who desire health care to be available as a welfare measure there has been an increasing commercialization of

II

J11

1111111

7 Delivery of Health Care and Resource Allocation

healthcare services. Organ transplantation is also a part of this overall milieu. (2) The purchase of organs is likely to have only a marginal impact on the cost of -xansplantation procedures. In many countries, including India, where there have been reports of organ trafficking, kidneys are sold for as little as US$400-500, while reports on the total cost of a kidney transplantation vary widely, ranging from US$1,0002° to US$8,000.21 Dr Raymond Crockett, debarred from practising in Britain in 1990, for professional misconduct, arranged for kidneys to be bought from Turkish people for £2,500-3,500, but charged each patient £66,000 for the transplantation.22 This shows that the cost of organs is just a fraction of the total transplantation cost, which, in fact, is much higher if the pos:-transplantation immunosuppressive therapy and the other follow-up care is taken into account. The apprehension in some quarters that organ sales will create a market mechanism that will greatly increase transplantation costs is therefore misconceived. Rather, the free availability of organs will reduce the costs of transplantation by curbing the expenses incurred in clandestine operations and the middle men who are invariably associated with the organ trade, as has been made clear by the recent exposures in London23 and the Punjab.24 Organ transplantation is a costly medical intervention, mainly because of the high fees of the surgeons and others involved in the process. Organ sale is unlikely, therefore, to increase the transplantation costs substantially. Moreover, if organ sale is legitimized the cost can always be regulated by the state, as is being done in the case of other commodities. A better option to reduce transplantation costs is to regulate surgeons' fees, nursing home charges, and the price of equipment, appliances, and drugs. (3) With regard to the concern that once organ sale is legitimised organs will mainly be sold by "those who cannot afford to keep their organs,"25 the matter requires deeper examination. The evidence thus far shows that in almost all cases, organs have been sold by persons in a state of abject poverty. In one case, in India, when I asked an organ seller why he had sold his kidney his reply was devastating "I had nothing else to sell!" People surrounded by such brutal poverty and social deprivation do not have many options. Even when their organs are intact their lot is miserable because they suffer from hunger, diseases, and scorn. Society has so far done nothing to alleviate their suffering, The selling of an organ may provide them with some additional means and prolong their existence. If the sale of organs amounts to exploitation of the poor it is no more than a continuum of the long drawn-out process of their exploitation, which has been watched by society for centuries. The poor have been selling all that they possess for centuries in order to continue

their existence. If, finally, selling their organs is the only way to get the money they need to prolong their existence, even if only temporarily, how can society stop them from doing so when society itself is unwilling to provide them with adequate means to survive. Why this sudden concern for the poor? A society passive to their problems for ages has no authority to interfere with the arrangement evolved by them to safeguard their survival. If, however, society truly feels they should not sell their organs, then their genuine needs should be addressed so that they are not forced to sell their organs. What kind of morality is it which snatches from the poor the only asset offered to them by nature? The prohibition on the sale of organs has worsened the lot of the poor. Buyers quite often refuse to pay or do not pay the agreed price. The vendor cannot assert his claim because of the fear of being prosecuted. Thus the strategy that was evolved for protecting the poor has been causing just the opposite effect. There is one more aspect to the above issue. If a person who is not poor and in whose case there is no possibility of being exploited chooses to sell his organ will he be allowed to do so? If not, this means that the reasons for prohibiting organ sale are grounded not in the concern for the poor, but in some other considerations. (4) The argument that permitting organ sale is not an equitable proposition since it restricts availability of organs only to affluent sections of the population is misconceived. How can it be fair to deny health care to those who wish to buy it using their own money, earned by honest means, on the ground that it is not available to others because of their inability to pay? If that is fair, why have we chosen a system of pricing for health care when many cannot pay even for the basic necessities, such as food and drinking water? Ours is an essentially heterogeneous society and equity has to be defined in realistic and pragmatic terms. (5) Prohibition on organ sale generates inequity by exerting undue pressure on the near relative who may "feel compelled to overlook the risks of organ donation when their loved one stands to receive so great a benefit."26 This pressure amounts to coercion, which is as bad as that exerted by the poverty and as such "should equally rule out donation."' The sale provides a wider choice, the vendor may be much healthier than the donor, and his or her organ may be much more compatible with the recipient. Furthermore, [during] removal of a vital part like a kidney the donor is subjected to a major surgical intervention which is not without risk. The recipient who is already sick is also exposed to a major surgical procedure with possibilities of complications. This means that two members of the same family are placed in a vulnerable

Kishore: Human Organs, Scarcities, and Sale: Morality Revisited

situation thereby affecting the fate of other members of the family also. In case of any complications or untoward outcome the said family may be the victim of serious adversities. As such it seems safer to accept donation from a member of a different family? (6) There is also a concern that there may be transnational movement of organs. The affluent countries, with the power of their money, may drain organs from the poorer countries thereby making the populations of the poorer countries even more vulnerable. This again is a problem of regulation, which can always easily be handled by banning exports of human organs or by adopting other suitable strategies. (7) The purpose of allowing the sale of organs is not to improve the health status of the sellers or to award them "a long term economic benefit," as conceived by some.29 Such economic or health benefit does not occur in altruistic donations either, but they are permissible. The reasons for permitting the sale of organs are grounded in the concern to save the lives of terminally ill patients with the help of available medical knowledge and technology by curbing "a contrived shortage created by existing organ procurement policies."30 As regards to poverty and ill health, they are rooted in factors other than organ sale and need to be addressed accordingly. Conclusion

Those who were earlier destined to die carry a hope to survive, provided the biotechnology is allowed to unfold itself. It may be worth appreciating that medicine is always need-based-that is, it is an aid to overcome

physical or mental disability or disease. It cannot be equated with justice, art, or spirituality. It is also worth remembering that the right to relief from pain and suffering is intrinsic-that is, it is non-divestible. Any social policy leading to infringement of such a valuable right has to be founded on equally vital considerations. Donor and recipient are the major stakeholders in policies relating to organ transplantation and as such any policy that does not take into account their perspectives and views is not valid. The sale of organs is essentially rooted in the urge to survive. The recipient wants to survive the threat of a terminal illness, the seller wants to survive the threat of poverty. Ethicists continue to debate the moral content of biotechnological promises, subordinating therapeutic advantages to "higher" goals but I am forced to draw the same conclusion that I drew in 1995-that is: Neither the diseased persons nor the genetic relations provide an answer to trading in human body parts. The live human body constitutes a vital source of supply of organs and tissues and the possibilities of its optimum utilisation should be explored. There is no scope for dogmatic postures and open mindedness should be the approach while dealing with the issue of organ transplantation. Society owes a duty to save the life of a dying man and in the event of failure to do so, it is absolutely immoral to interfere with his own arrangements by making unrealistic laws. The scarcity needs to be urgently overcome otherwise unwarranted trade and crime are liable to thrive. Commercialisation should be curbed by making the enforcement agencies efficient and not by depriving a needy person of his genuine requirements.28

Notes 1. Rohter, L. 2004. 'Tracking the Sale of a Kidney on a Path of Poverty and Hope," in New York Times (23 May): 1. 2. The United Network for Organ Sharing (uNos), [online], accessed 10 July 2004 at wwwunos.org. 3. Eurotransplant Foundation, [online], accessed 1 July 2004 at wwwtransplant.org. 4. Siegel-Itzkovich, J. 2003. "Israel Considers Paying People for Donating a Kidney" in stij 326: 126-7. 5. Tuffs Heidelberg, A. 2003. "German Surgeon under Investigation over Organ Trading," in BMJ 326: 568-9. 6. Harris, J., and Erin, G. 2002. "An Ethically Defensible Market in Organs," in BMJ 325: 114-15. 7. Human Transplant Act 1989 s 1(5). 1989. Hmso: London. 8. Law of 19 July 1972 of Venezuela on Organ Transplantation. 1972. International Digest of Health Legislation 23: 636.

9. Transplantation of Human Organs Act 1994. Gazette of India 1995 Feb 4: Part 2, section 3, sub-section (i). 10. American Medical Association: Council on Ethical and Judicial Affairs. 1995. "Financial Incentives for Organ Procurement Ethical Aspects of Future Contracts for Cadaveric Donors," in Archives of Internal Medicine b 5.581-9. 11. Marino, I.R., Cirillo, C., and Cattoi, A. 2002. "Market of Organs Is Unethical under Any Circumstances," in stij 325: 835. 12. Wigmore, S.J., Lumsdaine, J.A., and Forsythe, J.L.R. 2002. "Defending the Indefensible," in BMJ 325: 835-6. 13. Moslmann, E 2002. "The Right to Buy or Sell a Kidney," in Lancet 360: 948. 14. Kahn, J. 1999. 'Wanted: Tall, Smart and Fertile," in Bioethics Examiner 3: 1-4, at 4. 15. Cohen, C.B. 1999. "Selling Bits and Pieces of Human to Make Babies: The Gift of the Magi Revisited," in Journal of Medical Philosophy 24: 288-306, at 295.

II UM

11141.1

7 Delivery of Health Care and Resource Allocation

16. Schmidt, V.H. 1998. "Selection of Recipients for Donor Organs in Transplant Medicine," in Journal of Medical Philosophy 23: 50-74, at 52. 17. Glannon, W 1998. "Responsibility, Alcoholism, and Liver Transplantation," in Journal of Medical Philosophy 23: 31-49, at 31. 18. Anon. 2001. "No Transplants for Smokers," in The Hindustan Times (9 Feb.): 12. 19. Youngner, S., and Arnold, RM. 2001. "Philosophical Debates about the Definition of Death: Who Cares?," in Journal of Medical Philosophy 26: 527-37, at 527. 20. Swami, P 2003. "Punjab'; Kidney Industry," in Frontline (14 Feb.): 115-17. 21. Anon. 2000. "Kidney Transplant Racket Busted. Delhi Surgeon, Five Donors from Andhara Pradesh among Nine Arrested," in The Hindustan Times (7 Dec.): 1. 22. Anon. 2000. "Doctor Appeals against Ban on Kidney Sales," in The Times (23 Sept.): 11.

23. Dyer, 0.2002. "GP Struck off after Offering to 'Fix' Kidney Sale," in BMJ 325: 510. 24. Kumar, S. 2003. "Police Uncover Large Scale Organ Trafficking in Punjab," in BMJ 326: 180. 25. Kluge, E.-H. 2000. "Improving Organ Retrieval Rates: Various Proposals and their Validity," in Health Care Analysis 8: 279-95, at 283. 26. Kahn, J. 2003. "Dying to Donate," in Bioethics Examiner 7: 1-4, at 4. 27. Radcliffe Richards, J. 1996. "Nefarious Goings On," in Journal of Medical Philosophy 21: 375-416, at 377. 28. Kishore, A.R. 1996. "Organ Transplantation: Consanguinity or Universality?," in Medical Law 15: 93-104. 29. Goyal, M., Mehta, R.L., and Lawrence, J.S., et al. 2002. "Economic and Health Consequences of Selling a Kidney in India," in fAMA 288: 1589-93. 30. Barnet, A.H., Blair, R.C., and Kaserman, D.L. 1992. "Improving Organ Donation: Compensation versus Markets," in Inquiry 29: 372-8.

7.6 Cases

Case 1

Optimal Care versus Cost Containment: What Is a Doctor to Do? In a BC Supreme Court case, :he plaintiff sued doctors for negligence claiming that her husband died because the doctors had not taken a CT scan and thus failed to provide timely, appropriate, and skillful emergency care. In the course of the judgment against the doctors, Mr Justice Spencer of the British Columbia Supreme Court commented upon the relationship between budget reductions in health care and physicians' responsibilities toward their patients: I must observe that throughout this case there were a number of times when doctors testified that they feel constrained by the British Columbia Medical Association standards to restrict their requests for CT scans as diagnostic tools. No doubt such sophisticated equipment is limited and costly to use. No doubt there are budgetary restraints on them. But this is a case where, in my opinion, those constraints worked against the patient's interest by

Note 1.

Law Estate v Simice (1994), 21 C.C.L.T. (2d) 228 (B.C.S.C.).

inhibiting the doctors in their judgment of what should be done for him. That is to be deplored. I understand that there are budgetary problems confronting the health care system. I raise it in passing to point out that there were a number of references to the effect of financial restraint on the treatment of this patient. I respectfully say it is something to be carefully considered by those who are responsible for the provision of medical care and those who are responsible for financing it. I also say that if it comes to a choice between a physician's responsibility to his or her individual patient and his or her responsibility to the medicare system overall, the former must take precedence in a case such as this. The severity of the harm that may occur to the patient who is permitted to go undiagnosed is far greater than the financial harm that will occur to the medicare system if one more CT scan procedure only shows the patient is not suffering from a serious medical condition.' Judge Spencer's reasoning can be usefully compared to the discussions of physicians' roles in rationing health care resources, discussed also in Chapter 4.

Cases 1 413

Case 2

Having an Estate Sale of One's Organs Fletcher knows that his death is imminent, and he's been putting his affairs in order. He wishes he had more money to leave his family, especially his granddaughter, who would make a great doctor if she only had medical school tuition, and his eldest grandson, who would love to go to New York to drama school. While watching a news report about transplant tourism and the extreme lack of organs in Canada, Fletcher

Case 3

Rationing Services to an Elder Who Is Responsible for His Medical Condition Frank J. is 63 years old and semi-retired, working part-time at a non-strenuous job to support his invalid wife. The couple have no children or other family members. Frank is overweight but not obese; he smokes one-and-a-half to two packages of cigarettes a day, and he has been an alcoholic for the last 30 years, drinking whisky slowly but steadily from morning to night. He has had good health all his life in spite of his sedentary lifestyle. He has never been hospitalized and in fact has never had an illness more serious than the flu. His mother's death by medical mistake and his wife's chronic poor health have given him a distrust of medical professionals and an intense dislike of hospitals. Frank has come to his doctor with shortness of breath and intermittent chest pains. Testing indicates that Frank is suffering from congestive heart disease and that he will need a triple-bypass surgery. Frank is frightened, both of the prospect of heart surgery and of

Case 4

Buying a Kidney in India but Requesting Canadian After-Care Forty-four-year-old Harjumar B. has been suffering from kidney disease for years. He is fed up with having dialysis every other day while he waits up to six years for a donor kidney, he wants a more immediate solution. After liquidating some of his RSP savings, Mr B. travels to India where a broker has arranged for him to

gets an idea: he could hold an estate sale of his organs after his death. The highest bidder would get the organ, while his estate would get the highest price. "I think it's a great idea," he tells his wife. "I don't know," she responds hesitantly, "I think that sort of thing is illegal in Canada." "Why should it be?" Fletcher demands indignantly. "It's my body, my organs. I can auction off my car, my piano, and my mother's wedding jewelry, so why cant I auction off my organs? It's not like I'd be hurting anyone, in fact I'd be making sure some people who might not get an organ get one sooner. I think it's a win-win situation."

the year-long recovery period. He is reluctant to have the surgery, stating that it's his life, his choice, and that he may choose not to have the surgery and "let whatever happens, happen." His wife is uncomfortable with that choice because his death would leave her without income or a caregiver. Frank reluctantly agrees to see the surgeon to discuss his options. The surgeon tells Frank that in order to receive the surgery, he must give up smoking and drinking and must agree to certain lifestyle changes, including a healthier diet and more exercise. This ultimatum enrages Frank, who states that smoking and drinking are his only enjoyments in life and if those are gone, he has nothing left to live for. However, now that he knows that the surgery is no longer entirely his choice, he demands the surgery without limitations of any kind. He tells the surgeon that he's contributed to "the [health care] system" all of his life and taken nothing from it, so it's wrong of the ficPs to deny him treatment the first time he needs it. The surgeon explains that the very invasive surgery will not be as successful if he continues to smoke, drink, and live a sedentary lifestyle afterwards. Frank says he doesn't care—the system "owes" him and cannot ref use his need.

receive a kidney from an Indian woman. Usually patients never meet the donors, but a nurse at the hospital reveals to Mr B. that the donor is a healthy widow with five children who is using the money to help pay her bills. The surgery goes well, but upon his return to Canada Mr B. quickly becomes very sick. He goes to his local hospital where it becomes apparent that he has developed a serious infection that will require a lot of medication and time in the hospital to treat. As Mr B.'s daughter is passing the nursing station one day

II

414

,JII

1111.1.1

7 Delivery of Health Care and Resource Allocation

on the way to visit her father. she overhears the nurses talking about her fathen case. "Why should he be able to jump the line and go out of the country to buy his kidney, when others can't afford to do it?" wonders one nurse. "I don't know," says the other one, "I'd probably do it if I was dying and could afford it. After all, it's your life, you know" The first nurse replies:

"I don't think it's fair at all. Our medical care wasn't good enough for him when he decided he wanted a kidney—why is it good enough now when he's got an infection? He wouldn't even have the infection if he hadn't had the surgery outside the country. Why should Canadian taxpayers pay for something he's done to himself?"

Case 5

Security and the Guaranteed Income Supplement— which together totaled $1,169.47 monthly after he had reached the age of 65. On June 1, the federal government terminated Olsons payments. Human Resources Minister Diane Finley called the fact that he had been receiving them "outrageous" and "offensive," stating that "It's wrong and obviously unfair that prisoners who break the rules receive the same entitlements [as everyone else] ."2 In September 2011, it was revealed by the commissioner for corrections that Olson was suffering from metastatic terminal cancer and had been transferred to a hospital in Laval, Quebec, to undergo treatment. He died September 30 at the age of 71. This information was revealed to the victim's families under their right to know Olsons movements; however, out of respect for Olson's privacy, the commissioner's office would not release information to the media. The idea that Olson's privacy was worthy of respect and that he was receiving medical care—whether treatment for cancer previously or palliative care currently—enraged many Canadians. Some argued that Olson didn't deserve any treatment at all, that he should suffer and die like the beast he was. Some argued that if we live according to a social contract, then Olson broke that contract and is not entitled to the same treatment as those who keep it. On the other hand, some argued that Olson deserved the best medical treatment Canadian health care could offer, the same as any other patient.3 No consideration of his past actions should be included in decisions about current health care treatment.

Does Clifford Olson Deserve Medical Treatment? In 1980-1, Clifford Olson strangled, bludgeoned, and stabbed 11 British Columbia children and youths between the ages of 9 and 18 to death. After his arrest in August 1981, Olson made a cash-for-bodies deal that paid $10,000 per body located into a trust for Olson's wife and infant son; Joan Hale was paid $100,000. Olson pled guilty to 11 counts of murder and was given 11 concurrent life sentences and "dangerous offender" status, meaning that he would likely never have been released from prison. By his own admission, this self-proclaimed "beast of British Columbia" was guilty of between 80 and 200 murders and would kill again if ever released. While in prison he revelled in his multiple serial murderer status, including writing book manuscripts and making a series of video tapes describing the acts of torture and sexual assault he perpetrated on his victims, which included driving nails into their skulls, then asking them how it felt. Five years to the day after they found the body of Daryn Johnsrude, Olson continued his torment by sending a letter to the victim's• mother and stepfather describing in exact detail what he did to their 16-yearold son.' In March 2010, it was revealed that Olson had been receiving two federal government benefit cheques per month while in prison—the Canadian Old Age

Notes 1.

2.

CBC News, "In Depth: Clifford Olson, The Beast of British Columbia," 19 July 2006, http://www.cbc.ca/news/ background/olson-cliff ord Jane Taber, "Harper Cuts Clifford Olsons Government Pension Payments," The Globe and Mail 1 June 2010, http•1/ www.theglobeandmail.com/news/politics/ottawa-notebook/

3.

harper-cuts-clifford-olsons-government-pension-payments/ article1587995. Bill Kaufmann, "Olson deserves 'best medical treatment': Lawyer," The Toronto Sun 22 September 2011, http://www.torontosun.com/2011/09/22/ olson-deserves-best-medical-treatment-lawyer.

Study Questions

7.7 Study Questions 1. What is the egalitarian system of providing health care? What is the libertarian? What is the liberal? What are the strengths and weaknesses of each? Which do you prefer, and why? 2. Which areas of the Canadian health care system would you like to see improved (and how) and which left alone? 3. How are elements of egalitarian, liberal, and libertarian approaches to macroallocation reflected in Ronald Dworkin's approach to macroallocation of health resources? What do you think Dworkin's criticism of the libertarian approach to allocation of health resources would be? 4. What are the problems with using "the rescue principle" (also called "the rule of rescue") as a means of allocating scarce resources? 5. What are the different metrics for measuring the health benefits of spending scarce medical resources, and what are their advantages and disadvantages? 6. What is the complex criteria method of microallocation? What is the random selection method? Do you agree with either, or would you prefer a system that also takes into account age and responsibility for one's condition (or anything else)? 7. What is age-based rationing, and what are the arguments for and against it? Which side do you take, and why? 8. Do you think that persons should be offered reduced health care for conditions brought on by such things as smoking, drinking, overeating, eating fatty foods, underexercising, engaging in risky activities, or working too hard?

7.8 Suggested Further Reading Health Care Delivery Systems Armstrong, Pat. 1997. "Managing Care the Canadian Way" Humane Health Care International 13 (1): 13-14. DeGrazia, David. 1996. "Why the United States Should Adopt a Single-Payer System of Health Finance." Kennedy Institute of Ethics Journal 6 (June): 145-60. Freedman, Benjamin, and Francoise Baylis. 1997. "Purpose and Function in GovernmentFunded Health Coverage."Journal of Health Politics, Policy and Law 12 (1): 97-122. Neilsen, Kai. 1989. "Autonomy, Equality, and a Just Health Care System." The International Journal of Applied Philosophy 4: 39-44. Woolhandler, Steffie, and David U. Hiimmelstein. 1989. "A National Health Program: Northern Light at the End of the Tunnel."JAMA 262: 2136-7. Macroallocation Daniels, Norman. 2007. Just Health: Meeting Health Needs Fairly. Cambridge: Cambridge University Press. Englehart, Tristram. 1986. The Foundations of Bioethics, ch. 8. New York: Oxford University Press. Grogan, Collen M. 1992. "Deciding on Access and Levels of Care: A Comparison of Canada, Britain, Germany and the United States." Journal of Health, Politics, Policy, and Law 17 (2): 213-32. Gutman, Amy. 1981. "For and against Equal Access to Health Care." Millbanh Quarterly 59: 542-60. Nozick, Robert. 1971. Anarchy, State, and Utopia. New York: Basic Books.

II

416

JZI

11123.1

7 Delivery of Health Care and Resource Allocation

Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Sting], Michael. 1996. "Equality and Efficiency as Basic Social Values." In M. Stingl and D. Wilson, eds, Efficiency vs. Equality: Health Reform in Canada, 7-19. Halifax: Fernwood. Wiklar, Daniel, and Sarah Marchand. 2009. "Society's Allocation of Resources for Health." In Helga Kuhse and Peter Singer, eds, A Companion to Bioethics, 2nd edn. Oxford: Wiley Blackwell. Mesoallocation/Rationing Browne, Alister. 2010. "The Ethics of Aggressive Discharge Planning." Cambridge Quarterly of Healthcare Ethics 19 (1): 75-85. - --. 2008. "Violating Rules of Rationing." Canadian Journal of General Internal Medicine 3 (3). - --T. Anderson, D. Brown, C. Cooledge, B. Leal, D. McDonald, and M. Saxe-Braithwaite. 2005. "How to Make Allocation Decisions: A Theory and Test Questions." Healthcare Management Forum 18 (1): 32-3. Daniels, Norman. 1994. "Four Unsolved Rationing Problems: A Challenge." Published with replies to the challenges by Frances M. Kamm, "To Whom?"; Eric Rakowski, "The Aggregation Problem"; John Broome, "Fairness versus Doing the Most Good"; Mary Ann Baily, "The Democracy Problem." Hastings Center Report 24 (4): 27-42. - --. 1991. Is the Oregon Rationing Plan Fair? JAMA 265(17): 2232-5. Klevit, H.D., A.C. Bates, T. Castanares, E.P. Kirk, R. Sipes-Metzler, and R. Wopat. "Prioritization of Health Care Services: A Progress Report by the Oregon Health Services Commission." Archives of Internal Medicine 151: 912-16. Martin, Doug, and Peter Singer. 2003. "A Strategy to Improve Priority Setting in Health Care Institutions." Health Care Analysis 11 (1): 59-68. Menzel, Paul T. 1990. Strong Medicine: The Ethical Rationing of Medical Care. New York: Oxford University Press. Nelson, James Lindmann. 1996. "Measured Fairness, Situated Justice: Feminist Reflections on :-.1ealth Care Rationing." Kennedy Institute of Ethics Journal 6: 53-68. Wikler. Daniel. 1992. "Ethics and Rationing `Whether,' How,' or 'How Much'?" Journal of the American Geriatrics Society 40 (4): 398-403. Microallocation Broome, John. 1990. "Fairness." Proceedings of the Aristotelian Society, New Series 91: 87-101. Browne, A. 2013. "Blood in a Time of Scarcity" Cambridge Quarterly of Healthcare Ethics 22 (2): 159-69. Childress, James E 1970. "Who Shall Live When Not All Can Live?" Soundings 53: 339-55. Shaw, George Bernard. 1946 (originally published 1911). The Doctor's Dilemma. Harmondsworth, Middlesex: Penguin Books. White, D.B., M.H. Katz, J.M. Luce, and B. Lo. 2009. "Who Should Receive Life Support during a Public Health Emergency? Using Ethical Principles to Improve Allocation Decisions." Annals of Internal Medicine 150 (2): 132-8. Age-Based Rationing Brody, Baruch. 1988. "The Macro-allocation of Health Care Resources." In H.M. Sass and R. Massey, eds, Health Care Systems: Moral Conflicts in European and American Public Policy, 213-36. Dordrecht: Kluwer Academic Publishers. Callahan, Daniel. 1995. Setting Limits: Medical Goals in an Aging Society (with "A Response to my Critics"). Washington, DC: Georgetown University Press.

Suggested Further Reading

Childress, James F. 1984. "Ensuring Care, Respect, and Fairness for the Elderly" in Hastings Center Report 14: 27-31. Daniels, Norman. 1988. Am I My Parent's Keeper? An Essay on Justice between the Young and the Old. New York: Oxford University Press. Fleck, Leonard M. 2009. Just Caring: Health Care Rationing and Democratic Deliberation. New York: Oxford University Press. Harris, John. 1985. The Value of life. London: Routledge and Kegan Paul. Jecker, Nancy S. 1991. "Age-Based Rationing and Women." Journal of the American Medical Association 266: 3012. Williams, Alan. 1997. "Intergenerational Equity: An Exploration of the 'Fair Innings' Argument." Health Economics 6: 117-32.

Responsibility for Condition/Luck Egalitarianism Alvin H. Moss and Mark Siegler. 1991. "Should Alcoholics Compete Equally for Liver Transplantation?" Journal of the American Medical Association 265 (10): 1295-8. Anderson, Elizabeth. 1999. "What Is the Point of Equality?" Ethics 109: 287-337. Arneson, Richard.1989. "Equality and Equal Opportunity for Welfare." Philosophical Studies 56: 77-93, Atterbury, C.E. 1986. "The Alcoholic in the Lifeboat: Should Drinkers be Candidates for Liver Transplantation?" Journal of Clinical Gastroenterology 8 (1): 1-4. Katharine Browne. 2016. "VoluntarySterilisationandAccesstoIFVinQuebec ." http://jme.bmj .com/content/early/2017/08/05/medethics-2016-103726 Cohen, G.A. 1989. "The Currency of Egalitarian Justice." Ethics 99: 906-44. Dworkin, Ronald. 2003. "Equality, Luck and Hierarchy." Philosophy & Public Affairs 31: 190-8. Nielsen, Morten Ebbe Juul, and Martin Marchman Andersen. 2014. "Should We Hold the Obese Responsible? Some Key Issues." Cambridge Quarterly of Healthcare Ethics 23 (4): 443-51. Scheffler, Samuel. 2003. "What Is Egalitarianism?" Philosophy & Public Affairs 31: 5-39.

Pubk i_r ean 8.1 Introduction Public health is the part o f health care dealing with anything affecting the health o f populations that can be addressed through collective or government action. It is often contrasted with clinical medicine or health care, segments of health care considered in many of the other chapters in this text. The definition of public health has two important components. One is that the patient in public health is not an identified individual like someone in a hospital bed. Instead, the "patient" is a large population or the community as a whole. The other component is that the decision-makers are not the individual patient or his or her family in consultation with a medical professional deciding on a treatment that will affect only that person. Instead, the decision-maker is usually a department of the government— federal, provincial, or municipal—deciding on a policy that will affect many people. What kinds of moral issues are there in public health? Many of them concern the conflict between an individual's rights or freedoms and the well-being of other members of the community or population. Others are about how to balance harms and benefits to determine which programs will be for the good of the greatest number or which are most fair and just. These problems are illustrated in the four topics we've selected for this chapter: collective action problems in health care, medical screening, harm reduction programs, and inequalities in health prospects among different groups of people in Canada.

Collective Action Problems Types of Goods rival (or rivalrous) good a good where one person's benefit reduces or eliminates a benefit to others. h a non-rival good, one person's benefit or consumption does not affect another person's.

excludable good a good that people can be prevented from benefiting from without great cost. A non-excludable good is one where people can be stopped from consuming it only at great cost.

Economists sometimes divide goods—things we might want or need—into four different kinds.' The division is based on two features. A good can be rival or not in the benefits it provides to users or consumers. An organ transplant is an example of a health care good that is rival. If one patient is transplanted with a particular kidney (or "consumes it" in the words of an economist), then another cannot have that transplanted kidney. But there are health care goods that are not rival. The knowledge gained from medical research can benefit (or be consumed by) one patient without subtracting from any other patient's benefit. The other feature dividing types of goods is that getting benefits from the good may be excludable or not. Suppose medical researchers have found a new drug to treat multiple sclerosis. Access to the new drug is excludable; it can easily be denied to anyone not willing or able to pay, even when it would be medically beneficial. But access to the informat:.on about the drug's effectiveness is not excludable. Once the information about its effectiveness is available to some members of the general public, there is no realistic way to

Collective Action Problems 419

Table 8.1 Types of Goods Excludable

Not Excludable

Rivalrous

Private Goods hospital beds; tangible products of medical research (e.g., MRI machines)

Common-Pool Goods blood collected by Canadian Blood Services for transfusion; organs for transplant; access to family doctor appointments if eligible

Not Rivalrous

Club Goods electronically stored medical records; access to electronically stored medical journals

Public Goods knowledge from medical research; "herd immunity"; food and drug safety

common-pool resource good a good that is rival (in consumption) but is not excludable. club good a good that is not rival (in consumption) but is excludable.

prevent others from also finding this out (or "consuming" this knowledge). The combination of these two features (rival/non-rival and excludable/non-excludable) gives us the four categories of goods in Table 8.1.

Collective Action Problems in Health Care Collective action problems appear for goods in the second column. Because people cannot be excluded from consuming or benefitting from the good once it is available, there can be free riders and there can be overconsumption or overuse. Free riders use or consume the good without paying or contributing sufficiently to its production. Overusers consume or use it in such great amounts that production or supply of the good cannot keep up. In Chapter 4, Milton Weinstein in "Should Physicians Be Gatekeepers of Medical Resources?" (see page 181) considers overconsumption of the medical services provided by, for example, a Canadian provincial or territorial health plan. In this chapter, Katharine Browne in "The Measles and Free Riders" (see page 427) and Walter Glannon in "Free Riding and Organ Donation" (see page 431) consider free rider examples. It is easy to understand how someone can free ride when it comes to an organ for transplant or blood for transfusion. Others donate their organs or their blood. These goods are now available to anyone in medical need. No one in Canada is excluded from the benefit because they do not have a signed organ donor card or have not recently given blood. Anyone who has an organ transplant or a blood transfusion without contributing enough in some way to maintain the pool of organs or a blood supply is a free rider. But how can there be free riders when it comes to vaccinations against an infectious disease like measles, mumps, or polio? Vaccinations produce two benefits. The person vaccinated has their own risk of being infected reduced. This private benefit is consumed by the person. But the person's risk of infecting others is also reduced. This secondary benefit goes to others. These others can't be excluded from getting or consuming it. When enough individuals are vaccinated these secondary benefits add up to a collective or "herd" immunity, as Browne explains. Someone who gets this secondary benefit from others being vaccinated without themselves being vaccinated is a free rider. Assuming that it is wrong to free ride in these cases, both Browne and Glannon consider what is to be done to discourage free riding. In the case of vaccinations and organ donation, it is often suggested that medical professionals could encourage people to be vaccinated or to sign a card saying they agree to donate their organs after they die. Some are not vaccinated because they believe the risk of bad side effects from certain vaccines is so great that the benefits to themselves or their children, and perhaps even the secondary benefits to others, do not outweigh the costs to them and their families of being vaccinated. Although some people have religious or other convictions that forbid them from donating an organ upon death, in many cases not signing a donor card and free riding is not the result of considered thought. It turns out, unfortunately, that medical professionals are mostly unsuccessful at persuading these individuals to change their minds.2 Given this failure, an

collective action problem any failure of a group of individuals to achieve a result (for example, the production of a good) that everyone values at least as much as the result of everyone acting in their own self-interest. free rider a person who benefits from or consumes a good without contributing sufficiently to its production or continued existence.

II

MI

111Mial

8 Public Health

idea that Glannon favours to increase the number of organs donated after death is to adopt what is called an "opt-out" system. The system now in place in Canada is an "opt-in" one. Under an opt-in arrangement, people have to indicate—by signing an organ donor card—a willingness to donate their organs after death. Under an opt-out scheme, they would have to indicate—perhaps by going somewhere and signing up—an unwillingness to donate. A far greater percentage of people than are currently registered to donate their organs say that they like the idea of donating their organs after death.3 An opt-out system would make them walk their talk Another possibility is to have the government provide incentives for people to contribute to the production of the good, either the existence of herd immunity in the case of vaccination, or a pool of solid organs for transplant in the case of organ donation. This is a solution Browne investigates for vaccinations in her article. In some jurisdictions, children who are not vaccinated are not allowed to attend school.4 The legal incentives for organ donation can be as strong. In Israel, even if one patient has spent less time on the waiting list for a transplant than another, or is not as good a match for an available organ, priority can go to the one who has indicated a willingness to donate organs after death.5 AT there differences between organ or blood donation on the one hand and vaccination on the other that will alter what should be done about free riding? Most of us think that donating an organ or blood is a nice thing to do but nothing we are morally required to do. We do not harm others unjustly by not donating, Not being vaccinated, however, is different. Anyone not vaccinated risks infecting others. They risk wrongly infringing the rights others have not to be harmed. This is the kind of circumstance where John Stuart Mill (as Chapter 1 explains) would say that the government should intervene. But this difference between causing harm and not bringing or doing good may not be so great. Imagine, says James Rachels in "Active and Passive Euthanasia" in Chapter 6 (see page 301), two very similar but slightly different cases. In one, person A opens the bathroom door and pushes a baby into a bathtub where it drowns. In the other, person B opens the bathroom door, finds that a baby is drowning in a bathtub, and does nothing to save it. Person A is surely violating the baby's right not to be harmed. A non-vaccinating individual is like person A. Person B simply does not help the baby. This is like the non-organ-donating person But is person A really acting in any morally worse way than person B in Rachels's examples? If so, is anyone who is not vaccinated acting in a morally worse way than someone who does not donate organs after death or blood before death? There are many more examples of collective action problems in health care. Another example is related to the topic of Chapter 9—research with humans as subjects. This example is mentioned by Glannon in his article on organ donation. As Table 8.1 states, the knowledge gained from medical research is a benefit often available to everyone. (Some people, it is true, will benefit more than others.) If the right way for most people to contribute to medical research is to be a subject, then not being a subject is to be a free rider. Being a subject in medical research is usually thought of as something wonderful to do but not wrong not to do. But perhaps free riding in this case is as wrong as it is in the case of not signing an organ donation card.°

Screening screening test a medical test that looks for predictors or the early stages of illness when a person shows no symptoms.

In medical screening, people who have no symptoms of an illness are tested for predictors, precursors, or even the early presence of the illness. For example, screening tests for many medical problems are offered to all newborns in all provinces across Canada as part of provincial health care programs.' Very rarely are these tests mandatory, parents can decline to have them done if they wish. Most states in the United States, however, require that some of these tests be done.8 Another example of a screening test is the one for type II diabetes. This is recommended by Diabetes Canada for certain individuals who are 40 years of age or older or who are at high risk of getting this illness.9 Provincial health plans cover the cost of this screening test.

Screening

Screening programs certainly seem to be a good thing. What parents would not want to have their baby tested for a treatable illness like phenylketonuria (an inherited disorder that causes the build-up of the amino acid phenylalanine, which in turn leads to serious health problems)? This disease is part of newborn screening in every province. But even if this particular screening program is beneficial, this does not tell us that every screening test is worth encouraging people to have or worth spending public funds on. It also certainly does not settle whether being screened should be a voluntary choice, left up to an individual to decide on, or something that is mandatory. Some of these topics are taken up in the articles included in this chapter. Let's focus on whether publicly funded voluntary programs should be offered—that is, programs in which no one is compelled or required to be screened. What are the benefits of these programs? Sometimes illnesses that are not life-threatening can be prevented or at least have their damage reduced through early screening. This happens in the case of glaucoma (a condition that causes damage to the optic nerve). Sometimes screening saves lives through the early detection of an illness, such as cancer. Sometimes screening saves the lives of many by detecting a communicable infection like HIV or TB. But do these programs always save lives? Sometimes the way numbers about survival and success rates are presented for these programs can be misleading regarding lives saved and can lead an unwarranted enthusiasm for a screening program. This is what Darren Shickle and Ruth Chadwick call "screeningitis" in "The Ethics of Screening" (see page 436). There are also harms from mass screening programs. One type of harm is overdiagnosis and overtreatment, as they are sometimes called. This kind of problem is considered by the Shickle and Chadwick article and by Timothy Caulfield in "Direct-to-Consumer Genetics and Health Policy" (see page 434). Caulfield concentrates on genetic screening tests like those provided by 23andme and other companies here in Canada and around the world, but the problem also occurs for other screening tests. Another kind of harm that can result from mass screening programs comes from the mistakes that screening tests make. Four different outcomes or results are possible with any screening test. The result of a screening test for a disease may be positive (the disease is present according to the test) or negative (it is not present according to the test). Each of these results may be correct or true on the one hand or incorrect or false on the other hand. A correct or true result means that what the test says is right, but most medical screening tests get it wrong sometimes, producing a false positive or false negative result. Of course, no one knows when using the test whether a particular result is a true or false one. All that is known is what the screening test says and, from research done on the test, the rate at which it gets it correct or not. (If other information from a patient's medical record is also available, it can help to resolve some of the uncertainty.10) This issue is discussed by Shickle and Chadwick in a general way. Case 3 in this chapter considers what happens in a specific example (breast cancer screening). Aside from considerations about these harms and benefits for publicly funded screening programs, what other moral issues might there be? One comes up in the chapter on medical information (Chapter 3). Suppose people—for example, the parents of babies in Canada— have a positive right to information about their own health and that of their small children. This would seem to mean that they have a right that public health authorities, someone, or some group provide them with effective screening tests. On the other hand, if these people have only a negative right, then all that public health authorities are morally required to do is avoid getting in the way of people paying privately for their own screening tests. This is the main issue considered by Caulfield in "Direct-to-Consumer Genetic Testing."

Harm Reduction Programs Harm reduction or minimization programs are public health measures aimed at reducing the harmful effects resulting from people's unhealthy or risky behaviours. These types of

overdiagnosis a correct diagnosis of illness (for example, cancer) where the illness will do no harm to the person during that person's lifetime.

overtreatment medical treatment for an illness given when the illness will do no harm to the person during that person's lifetime.

false positive a test result that is positive but incorrect. A test result is positive when thetestsays a patient has the illness being tested for or a predictor for it.

false negative a test result that is negative but incorrect. A test result is negative when the test says a patient does not have the illness being tested for or a predictor for it.

positive right a right to be provided with something (a good or benefit) by some institution or person or group. It corresponds to a positive duty on the part of that institution or person or group to provide the good or benefit to those with the right.

negative right a right not to be stopped by some institution or person or group from getting something (a good or benefit). It corresponds to a negative duty on the part of that institution or person or group not to interfere with those with the right from getting thegood or benefit.

II

422

j11

1111.1,1

8 Public Health

programs are not to be confused with ones that have the goal of eliminating unhealthy behaviours, which are called "cessation" or "abstinence" or perhaps "zero-tolerance" or "war on" (as in "war on drugs") programs. An example of a cessation program would be to forbid bike riding on city roads because people can get hurt riding their bikes. An example of a harm minimization program, however, would be to allow bike riding while adding special bike lanes, encouraging riders to wear helmets, and requiring that bikes have appropriate lights and reflectors for night-time riding. Harm reduction programs, at least of the kind we are concerned with here, offer medical and health care services to people who behave in risky or unhealthy ways. Most of us will think of methadone maintenance programs for those addicted to heroin or needle-exchange programs and supervised injection sites for intravenous drug users when it comes to these kinds of programs. Other kinds of harm reduction programs in which medical professionals participate include measures and interventions to help sex trade workers avoid sexually transmitted diseases, college students to overcome the after-effects of binge drinking sessions, and competitive athletes to deal with side effects from performance-enhancing drug use. None of these programs are attempts to stop the activities; they just deal with the harmful results. There are also less obvious examples of harm reduction programs. Suppose you are diabetic, have a heart condition, or have high blood pressure partly as a consequence of an unhealthy diet. Or suppose you will need a liver transplant if you continue to drink alcohol in large amounts. In these scenarios, any medical services provided to you to reduce or help combat the harmful effects of your behaviour—so long as they aren't aimed primarily at getting you to stop the behaviour—will be an example of harm reduction or minimization. There are many arguments against harm reduction programs such as safe injection sites and against medical professionals participating in them. One is that health care professionals will have "dirty hands" if they provide services in these kinds of harm reduction programs.'2 They are participants or helpers in something that is morally bad, and so they are dirtying their hands by being involved. For this argument, the real question is whether the unhealthy behaviour is immoral. In some cases it is certainly illegal. This is why safe injection sites and some other kinds of harm reduction programs need special legal permissions, as noted by Dan Small in "Canada's Highest Court Unchains Injection Drug Users" (see page 443). But are the behaviours—using addictive drugs, for example—immoral? One possibility is that people who use these drugs can't control their behaviour. If these individuals are not able to control their actions voluntarily, they are not acting immorally. After all, as the philosopher Immanuel Kant famously said, "Ought implies can."12 If the behaviours are not immoral, then helping people to avoid the harmful effects of the behaviour leaves one's hands clean. This line of thinking is considered by Dan Small in his article (see page 443) in this chapter. A different argument is considered in "Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate" by Timothy Christie, Louis Groarke, and William Sweet (see page 451). This is an argument concerned with the consequences of having these harm reduction programs. The idea is that by removing or helping to minimize some of the bad effects of unhealthy behaviour like intravenous drug use, health care professionals and these programs encourage the behaviour and maybe even increase it. This idea is not completely silly. Going back to the example of riding a bike on city roads, researchers have found that some bike riders compensate for the safety benefits wearing a helmet should bring them by riding in riskier ways. In other words, harm reduction encourages unhealthy, dangerous behaviour by reducing or removing effects that would otherwise discourage the behaviour.13 Does the same thing happen with drug users and sex trade workers? If it does, then harm reduction programs are not so beneficial after all. Fortunately, the factual evidence we have shows that methadone maintenance or

Harm Reduction Programs

needle-exchange programs and supervised injection sites do not increase drug use. Some of this evidence is provided in the article by Small. Some other arguments say there is nothing morally wrong with harm reduction programs, or with medical professionals offering their services in these programs, and that these programs should exist. One of these arguments is found in Small's article. It relies on the reasoning in a judgment in a Supreme Court of Canada case. Small argues—as noted a few paragraphs ago—that users of addictive drugs like heroin do not freely participate in their unhealthy behaviour; it is outside their control So, like others whose medical needs are the result of factors outside their control, addictive drug users should receive as large a share as any other medical need when the budget for scarce medical resources is being allocated. Furthermore, their share should be provided in a way that is most medically beneficial and efficient. That way is harm reduction efforts like methadone maintenance programs for heroin addicts and clean needle-exchange programs and supervised injection sites for intravenous drug users.

Equality in Health As long ago as 1986, Health Canada recognized that certain population groups in Canada have significantly lower life expectancy, poorer health, and a higher prevalence of disability than the average Canadian.14 The map in Figure 8.1 has more recent data from Statistics Canada that shows some of what Health Canada was talking about. In this case, the population groups are people living in different health regions across the country." The map (Figure 8.1) shows that there are differences in expected life spans—an important measure of general health—in different health regions across Canada. Some areas are much better than the Canadian average and some much worse. These differences are

health region a geographical area defined for administrative purposes by a provincial or territorial ministry of health.

Compared to the national average of 80.7 years Significantly higher M Not significantly different =I Significantly lower Sparsely populated

Regina

titer* . • .47

Winnipeg

11.011).. AA.Charlottetown

Fredericton 1/7 Halifax Ottawa

Figure 8.1 Expected Life Span Differences across Canada Source: Statistics Canada. Vital Statistics. cansinATable 102-4307. Life expectancy. Produced by the Geography Division for Health Statistics Division, Statistics Canada, 2011.

II

.111

1111:41

8 Public Health

greater than just accidental chance would allow. This phenomenon is repeated on a smaller scale too—that is, inside what Statistics Canada calls health regions. In Vancouver, where the editors of this textbook live, the Vancouver Coastal Health regional health authority says that those living in Vancouver—Westside are expected to live 11.5 per cent longer than those in Vancouver—Downtown-Eastside.i° Are these differences or inequalities in health unjust or unfair? Differences, even ones as dramatic as these, do not by themselves mean that there is injustice or unfairness. As Norman Daniels tells us in "When Are Health Inequalities Unjust?" (see page 458), they are unjust only if what causes them is something that could have been changed or done differently. So what is responsible for these differences in health status in various regions of the country? Are they things that could have been done differently? We know, for example, that there are great differences in the distribution of medical care, differences considered in Chapter 7. Urban areas usually have much better health services—larger, easily accessible hospitals, among other things—than rural or sparsely populated areas. This situation is certainly part of the story, and it is of course something that could be changed by government action. But according to the Canadian Medical Association (CMA) and many social scientists, the distribution of medical care is actually only a small part of the explanation for the dramatic differences among health regions.'7 Figure 8.2 shows what the CMA tells us is responsible for the health of Canadians and for the great population-level differences in average health. As Figure 8.2 notes, a substantial proportion of what contributes to individual and population-level differences in health are factors like income, education, housing

WHAT MAKES CANADIANS SICK?

50% >

YOUR LIFE

INCOME EARLY CHILDHOOD DEVELOPMENT DISABILITY EDUCATION SOCIAL EXCLUSION SOCIAL SAFETY NET GENDER EMPLOYMENT/WORKING CONDITIONS RACE ABORIGINAL STATUS SAFE AND NUTRITIOUS FOOD HOUSING/HOMELESSNESS COMMUNITY BELONGING

25% >

YOUR HEALTH CARE

ACCESS TO HEALTH CARE HEALTH CARE SYSTEM WAIT TIMES

15% 10%

~•> YOUR BIOLOGY YOUR ENVIRONMENT

BIOLOGY GENETICS AIR QUALITY CIVIC INFRASTRUCTURE

Figure 8.2 What Makes Canadians Sick Source: Canadian Medical Association. 2013 ."Health Equity a n d the Social Determinants of Health:' https://www.cma.ca/En/ Pages/health-equity.aspx.

Equality in Health 425

conditions, and other aspects of a person's daily life. Some of these are what social scientists call the social determinants of health.'7 But Statistics Canada tells us that there are disadvantages in these social determinants in the health regions where people have shorter life expectancies and corresponding advantages in the health regions where people have longer life expectancies.'8 The deprived regions have in common lower incomes, high long-term unemployment rates, lower levels of education, small immigrant populations, and relatively large Aboriginal populations. So a substantial proportion of what contributes to the population-level health differences shown in the map in Figure 8.1 are factors like these advantages or disadvantages in social determinants. Can these factors, which are largely not the result of the behaviour or habits of those living in these regions, be changed and improved? Surely some can be, as Daniels tells us. But Statistics Canada also notes that these disadvantaged regions also tend to be ones with some of the highest rates of smoking, obesity, and heavy drinking rates in Canada.18 Thus, some of the factors that produce the health problems in these locations are the result of behaviour. Can these behaviours be changed? Perhaps they can be—as Daniels suggests—and largely by improvements in social and economic circumstances. Clearly, some things could be changed to eliminate the differences in health among people living in these regions. However, that fact is not enough to prove that the inequalities are unjust inequities needing to be changed. What more is needed to prove that point? Two quite different answers to this question can be found at various points in the article by Daniels. One of the answers is backwards-looking. It argues that populationlevel differences in health are unfair and should be corrected if these differences resulted from controllable and unjust past disadvantages for the populations with worse health— for example, if they were produced by laws or practices that discriminate unfairly on the basis of race, ethnicity, social class, or gender. The other answer is forwards-looking: population-level differences in health are unfair and should be corrected if these differences produce future controllable and unjust disadvantages for the populations with worse health. For example, the inequalities are unjust in a forwards-looking way if they put people at a disadvantage when it comes to competing on equal terms with others for economic and social goods like a spot in medical school or a good job after graduation. These are not the only possible answers. Some authors (not included here) defend more radical ideas.'9 For example, according to one radical backwards-looking answer, population-level differences in health are unfair and should be corrected if these differences resulted from any past disadvantages for the populations with worse health. On this view, it doesn't matter whether the disadvantages were unfair or unjust. (They do on one of the answers Daniels provides.) They could, for instance, be the result of bad luck rather than factors like discrimination. One question that can be asked about all of these views—backwards-looking or forwards-looking, radical or not so radical—is whether what is important is absolute or comparative health deprivation. Here is what this means. The average life span for Canadians living in disadvantaged health regions is just over 71 years, whereas the expected life span is nearly 85 years in the most advantaged health regions.18 A moral approach that focuses on comparative health deprivation sees this inequality or difference as the key focus. If the expected life span in every health region were pretty much the same—say, around 74 years—there would be nothing to complain about morally. There would be no unfair difference because there would be no difference. Of course, it might be better if the average life span were longer, perhaps even nearing the 84 or 85 years of the most advantaged health regions today, but in terms of comparative inequality in health, a longer life span would not be a moral improvement or anything required by justice. On the other approach, absolute health deprivation, the key focus is on the fact that the health of some groups—their expected life spans, rates of disability, and other aspects of average general health status—is much worse than it could be. The better health status of those in advantaged regions shows how much better the health status of the disadvantaged groups could

social determinants of health the social and economic conditions in which people are born, grow, live, work, and age that have effects on their health.

II

.1II

IIILLI

8 Public Health

be if something in their lives or situations were different and improved. Therefore, justice requires not so much erasing the differences between those health regions as bringing all regions up to the level that those regions with the best of health show is possible.

Notes 1. Gregory Mankiw, Microeconomics, 5th edn, ch. 11 (Mason, OH: South-Western Cengage Learning, 2009). 2. Brendan Nyhan and John Reifler, "Does Correcting Myths about the Flu Vaccine Work? An Experimental Evaluation of the Effects of Corrective Information," Vaccine 33 (2015), 459-64; Eve Dube et al., "Understanding Vaccine Hesitancy in Canada: Results of a Consultation Study by the Canadian Immunization Research Network," PlosOne 11 (2016). 3. Richard H Thaler, "Opting in vs Opting out," The New York Times (26 Sep 2009). Ipsos Reid, "Canadians willing to donate their organs in the case of their death" (11 Jul 2006). Available at http://www.marketwired.com/press-release/canadians-willing-to-donate-their-organs-in-the-case-oftheir-death-603162.htm. 4. Heather MacDougall, Laurence Monnais, and Monica Brown, "The Big Picture: Historical, Social and Cultural Features of Vaccine Hesitancy," presentation to the Canadian Public Health Association conference, Toronto, June 2016. 5. Jacob Lavee, Tamar Ashkenazi, Gabriel Gurman, and David Steinberg, "A new law for allocation of donor organs in Isreal," The Lancet vol 375 (2010), 1131-1133. 6. Fr tz Allhoff, "Free Riding and Research Ethics," American Journal of Bioethics 5 (2005), 50-1; G. Owen Schaefer, Ezekiel Emanuel, and Alan Wertheimer, "The Obligation t o Participate in Biomedical Research," Journal of the American Medical Association 302 (2009), 67-72. 7. Lift Grainger, "Newborn Screening in Canada: What to Expect," Today's Parent 20 October 2014; Canadian Organization for Rare Disorders, "Newborn Screening in Canada Status Report," Canadian PKJ and Allied Disorders, September 2015. 8. National Newborn Screening and Global Resource Center, "US Newborn Screening Programs," http://genes-r-us.uthscsa.edu/resources/consumer/statemap.htm. 9. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee, "Screening for Type 1 and Type 2 Diabetes," http://guidelines.diabetes.ca/browse/chapter4. 10. This uses information from the medical record to establish a prior probability for having or not having the condition being screened for. 11. John Kleinig, 'The Ethics of Harm Reduction," Substance Use and Misuse 43 (2008), 1-16. 12. Immanuel Kant, 1998, Critique of Pure Reason, p. A548/13576 (Cambridge and New York: Cambridge University Press). 13. Jake Olivier and Prudence Creighton, "Bicycle Injuries and Helmet Use: A Systematic Review and Meta-Analysis," International Journal of Epidemiology 46 (2017), 278-92; Tim Gamble and Ian Walker, "Wearing a Bicycle Helmet Can Increase Risk Taking and Sensation Seeking in Adults," Psychological Science 27 (February 2016), 289-94; Ross Phillips, Aslak Fybri, and Fridulv Sagberg, "Risk Compensation and Bicycle Helmets, Risk Analysis 31 (2011), 1187-95. 14. J. Epp, Achieving Health for All: A Framework for Health Promotion (Ottawa: Health and Welfare Canada, 1986); Dennis Raphael, "Health Inequalities in Canada: Current Discourses and Implications for Public Health Action," Critical Public Health 10 (2000), 193-216. 15. Other population-level differences exist in Canada. See Juha Mikkonen and Dennis Raphael, Social Dererminants of Health: The Canadian Facts (Toronto: York University School of Health Policy and Management, 2010); Margo Greenwood, Saraj de Leeuw, Nicole Lindsay, and Charlotte Reading, ed5.., Determinants of Indigenous Peoples' Health in Canada, (Toronto: Canadian Scholars' Press, 2015); Pat Armstrong and Ann Pederson, eds, Women's Health, 2nd edn. (Toronto: Women's Press., 2015). 16. Ken MacQueen, "Vancouver's Downtown Eastside Gets New Lease on Life," Maclean's 15 October 2012. 17. Dennis Raphael, ed., Social Determinants of Health: Canadian Perspectives, 3rd edn., (Toronto: Canadian Scholars' Press, 2016). 18. Lawson Greenberg and Claude Normandin, "Disparities in Life Expectancy at Birth" (Ottawa: Statistics Canada, 2015), http://www.statcan.gc.ca/pub/82-624-x/2011001/article/11427-eng.htm. 19. Shlomi Segall, "Affirmative Action in Health," in Shlomi Segall, Equality and Opportunity, (New York: Oxford University Press, 2013), ch. 3.

Browne: The Measles and Free Riders: California's Mandatory Vaccination Law

8.2 Collective Action Problems

The Measles and Free Riders: California's Mandatory Vaccination Law Katharine Browne Introduction

In response to the measles outbreak—famously linked to Disneyland—and to the growing number of unvaccinated children, California recently passed legislation to eliminate all nonmedical exemptions for school-mandated vaccines.' According to the new legislation, children who do not receive schoolmandated vaccines and who do not have a medical excuse will be prohibited from attending public and private schools. This makes California the largest state to restrict vaccine exemptions to those that are medically necessary and could set a precedent that would pave the way for other states to follow suit. However, the legislation has been met with significant resistance. Some opponents to it have claimed that implementing mandatory vaccinations violates medical freedom and parental autonomy, emphasizing that wherever there are risks associated with vaccination, there should be choice regarding whether to vaccinate. In this article I take up a game-theoretic perspective on the problem of vaccination compliance. Such an approach characterizes parental decisions to vaccinate their children as a collective action problem. Collective action problems reveal the presence of an incentive to free ride— to benefit from the efforts of others who vaccinate their children without carrying any of the costs of vaccinating one's own. I argue that California's legislation represents a reasonable means to overcome the free rider incentive and improve compliance with vaccination requirements. Although the focus of this article is on California's legislation, this conclusion has implications for other jurisdictions where vaccination compliance is a problem.

of further spread of the disease is minimized, because there are fewer transmission points (i.e., nonimmune individuals) by which the virus can spread? Achieving herd immunity helps to protect vulnerable members of the population who cannot be vaccinated, such as infants and individuals with certain health issues. The percentage of the population that needs to be vaccinated for herd immunity to obtain varies, depending on the illness. In most cases, herd immunity obtains when at least 90% of the population is vaccinated against an illness. In order to achieve herd immunity from the measles, which is highly contagious, 96%-99%3 of the population needs to be vaccinated. The World Health Organization (WHO) estimates that the United States national average for vaccination against the measles is 91%.4 In 2013-2014, 92.3% of kindergarten children in California were vaccinated with the vaccine against measles—the MMR (measles-mumps-rubella) vaccine.5 A recent analysis of the data from the Disneyland-linked measles outbreak revealed an estimated vaccination rate of 50% to 86% among those exposed to the illness.6 It is thus clear that vaccination levels are lower than would be desirable to guarantee herd immunity against the measles. What is not clear is what should be done about this. The debate over mandatory vaccination is a heated one, often characterized as one between science and its opponents. Antivaccinators are often characterized as selfish, ill-informed, and irrational. And, although many of the claims on which the antivaccine movement rests are arguably without scientific credibility,' it is not true that vaccines carry with them no risk, and there remains a legitimate question about the extent to which the state can interfere with people's choices to vaccinate. On the one hand, if fewer people are vaccinated, herd immunity drops, and this puts everyone at risk. On the other hand, it is desirable for coercive measures to be kept to a minimum whenever possible. Thus, it would be best if compliance with vaccination requirements could be secured voluntarily. Securing Voluntary Compliance

Herd Immunity, Vaccination, and the Law

The Holy Grail in the vaccine world is to achieve what is known as "herd immunity" This describes the point at which the majority of a population is immune to a particular illness. Once herd immunity is achieved, the risk

To see if and how voluntary vaccine compliance might be achieved, consider the following. Parent A is deciding whether to have her children vaccinated against the measles. Factors relevant to her decision may include the following: (1) the risk associated with the vaccine,

Source: From Cambridge Quarterly of Healthcare Ethics 25, 3 (2016), 472-8. Reprinted with permission.

II 'Al

HILL!

8 Public Health

(2) the effectiveness of the vaccine in protecting against the measles, (3) the risk of contracting the measles without the vaccine, and (4) the risk of complications arising from the measles. The antivaccine movement tends to latch on to the first factor concerning vaccine safety and has been fueled largely by the (discrechted8) link drawn between the MMR vaccine and autism. Research concerning parental attitudes toward vaccination reveals that parents may also deem vaccination unnecessary, either because they assign a low probability to the possibility of their children contracting the disease or because they believe they can control whether their children are exposed to the disease.9 The legacy of successful vaccination programs is that the dangers of illnesses like the measles are unfamiliar to parents and risks of exposure and complications may be perceived as low. Insofar as these factors impede achieving herd immunity, one strategy to boost voluntary vaccination compliance would be through public health efforts designed to educate parents about the risks and benefits of vaccination and the dangers of contracting the measles. However, it is not clear that such education programs will be sufficient, for two reasons. First, there is little empirical evidence that parental attitudes can be changed through public education efforts. Although there is a growing literature concerning reasons for parental refusal of vaccination, there remains surprisingly little data on which interventions are most effective in reversing vaccination ref usal.'° Furthennore, the data that do exist suggest that parental attitudes concerning vaccination are not readily changeable. A recent American study of vaccine promotion revealed that education about vaccination tends not to influence parents to choose to vaccinate and, in fact, often has the opposite effect.il Second, educational efforts may not address all relevant factors involved in parents' decisions to vaccinate their children. Indeed, it is the third factor named previously—the risk of an unvaccinated individual contracting the measles—that is especially problematic. More specifically, the facts that (1) the risk of contracting the measles is dependent on herd immunity and that (2) herd immunity is dependent, largely, on whether other parents choose to vaccinate their children pose a serious challenge to securing voluntary vaccination compliance.

Game Theory and Vaccination The interdependence of parental choices to vaccinate (that is, the fact that one parent's choice to vaccinate will be influenced by the choice of another, and vice versa) sets the stage for a game-theoretic analysis of

those choices.'2 Game theory is the study of strategic interaction among rational individuals. The players in the game are taken to be ideally rational, where this is usually understood to mean they are expected utility maximizers. In a strategic choice situation involving two or more rational players, an individual will choose on the basis of what she expects her opponent(s) to do: her action will constitute a best reply to the action she anticipates the other player(s) will choose. Consider Parent As decision whether to vaccinate her children against the measles with the MMR vaccine. Assume, for the sake of the argument, that the MMR vaccine is 100% effective" at guarding against the measles; that when more than 96% of the population vaccinates their children, herd immunity is achieved; that wherever there is herd immunity there is a 0°k chance of contracting the measles; that when less than 96% of the population vaccinates their children, there is no herd immunity, and that where there is no herd immunity the chance of contracting the measles is 100%. The matrix in Table 1 represents Parent As choice situation, where "benefit" stands for the benefit of immunity from the measles, "cost" for the myriad costs associated with vaccination, and "measles" represents harms associated with contracting the measles. Parent A can choose between two strategies: "vaccinate" or "don't vaccinate." The first column of the matrix represents Parent As choice situation when there is herd immunity. In such a case, the option to vaccinate carries with it both the benefit of immunity from vaccination and the costs of vaccination. This cost can range from the inconvenience of going to the doctor to obtain the vaccine or more serious side effects associated with the vaccine. Importantly, however, when there is herd immunity the individual benefit of immunity obtains regardless of whether Parent A vaccinates, because protection against the measles is already achieved by the efforts of the 96% of the population who vaccinate their children. Because, in this scenario, "don't vaccinate" imposes no cost to Parent A but permits enjoyment of the benefit of herd immunity, Parent A should choose "don't vaccinate." In so doing, Parent A will be free riding off of the efforts of the 96%—enjoying the benefits of herd immunity from vaccination without bearing any of the costs associated with vaccinating.

Table 1 Parent A's Choice Situation > 96% vaccinate (herd immunity) Vaccinate

Benefit + cost

Don't vaccinate Benefit + no cost

< 96% vaccinate (no herd immunity) Benefit + cost No cost + no benefit + measles

Browne: The Measles and Free Riders: California's Mandatory Vaccination Law

The second column in the matrix represents Parent A's choice where there is no herd immunity. Here "vaccinate" carries with it the same costs as before but confers the benefit of immunization that would otherwise be unavailable. If a parent chooses the option "don't vaccinate," she and her children avoid the costs associated with vaccinating, but the children face the potential harms associated with contracting the measles. Her choice to vaccinate in this situation then depends on whether the potential harms associated with the measles are greater than the costs associated with vaccination. In this matrix, when 96% of the population vaccinates, Parent A, insofar as she is rational, will reason to the conclusion that it is to her advantage to refrain from vaccinating. That way she could enjoy the benefits of herd immunity without bearing any of the costs associated with receiving the vaccine. This is a free rider temptation. Thus, the basic logic in the case of vaccines is this. If a sufficient number of others get vaccinated, then everyone will be better off. But it would be best for any particular individual if all others got vaccinated and he or she did not. The trouble is that if everyone thought that way, no one would become vaccinated. There would be no herd immunity, and everyone would be at risk of falling ill. In this picture of vaccination as a collective action problem, it emerges that public education about the safety and efficacy of vaccination is of limited value in encouraging parents to vaccinate their children. Until the threshold for herd immunity is met, all unvaccinated individuals are at risk of contracting the measles, and education can show parents that on a simple costbenefit analysis they should vaccinate their children. However, once herd immunity is achieved, it is rational for parents to refrain from vaccinating their children, and no amount of education about the risks and benefits of vaccination can show otherwise. Policy and Free Riders

Ensuring that populations are adequately protected against vaccine-preventable illnesses like the measles requires a means to both achieve and maintain herd immunity. One policy option is to continue with efforts at education about the risks and potential benefits of vaccination but allow the exemption of conscience (i.e., for philosophical and religious reasons) and otherwise take no measures that go beyond the voluntary. That would preserve individual freedom (keeping antivaccinators happy) and may help to achieve herd immunity but would do nothing to maintain it once it is achieved. The likely result would be

occasional measles outbreaks like the one that started in California. But this still leaves open the possibility of free riding, and there are good reasons to close that opportunity. First, eliminating free riding ensures that no one takes unfair advantage of the collective efforts of others and everyone shares in the burden. Second, permitting free riding increases the risk of infection in the community and thus puts vulnerable members of the population—those who, for medical reasons, cannot be immunized—at risk of contracting the illness. Finally, the risks of measles for anyone should not be minimized. Although there have been no measles-related deaths reported to date in 2015 in the United States, from a global perspective there has been significant mortality over the years. For example, in 2013, there were 145,700 measles-related deaths worldwide.14 Meanwhile, the WHO has estimated that vaccination prevented 15.6 million measles-related deaths between 2000 and 2013.15 There are also (often serious) complications that accompany the measles, such as hearing loss, respiratory infections such as pneumonia, and encephalitis.16 An alternative, or possible supplement, to educational efforts would be to appeal to the common good in order to encourage vaccination. The free rider problem emerges because of a tension between individual selfinterest and the good of the collective. If that tension could be resolved, then the problem would disappear. So if people can be encouraged to care about the welfare of the collective rather than just their own (or that of their children), then vaccination may be achievable without the use of force. But to the extent that parents are concerned with their own children's welfare and not necessarily the common good (or the welfare of other people's children), and to the extent that such concerns are not always readily changeable, appealing to those elements will prove ineffective. Add to this equation a risk of serious side effects from vaccinating one's own children, and appeals to the collective good will quickly lose force. This suggests that heavier-handed interventions may be needed to ensure vaccination compliance. Such measures are common in other policy fields. Take, for example, littering. Although it is best for everyone if people refrain from littering in the streets, an individual person's small contribution of litter makes minimal difference to the cleanliness of her city's streets. There is a temptation to free ride. But if everyone reasons this way, no one will use a trashcan, and the city's streets will become full of litter. Because self-interest will not guarantee clean streets, and because appeals to the common good of keeping the city's streets clean sometimes prove ineffective, fines are implemented for littering.

II J11

111111

8 Public Health

Following this logic, we might say that similarly heavy-handed interventions may be needed in the case of vaccines. At the extreme end of the spectrum of such measures would be forced vaccination—administering vaccination against the will of parents or children. But to the extent that public policy should aim at the mildest measure possible to produce the desired result, such an intervention should be a last resort, and milder measures should be tried first. California's proposed legislation is one such measure. To be sure, it does have disadvantages. It runs the risk that parents who remain opposed to vaccination will remove their children from the school system instead of complying with mandates. The alternative to public or private schooling is home schooling. If unvaccinated children are homeschooled together, this may jeopardize herd immunity, especially given clustering effects, whereby unvaccinated individuals tend to pool together, which increases the probability of a disease outbreak.17 Furthermore, this alternative runs the danger of disadvantaging the unvaccinated children by depriving them of an education to which they are entitled. Children who are homeschooled may also be deprived of socialization tha: they would receive in the school system. There are alternatives to California's proposed legislation. Australia, for example, has recently announced its "no jab, no pay" policy, which prevents parents who refuse vaccination for philosophical or religious reasons from receiving annual tax and childcare benefits.' The benefits forgone by parents who refuse immunization may amount to as much as $11,000 per child.19 Whether this intervention will be effective in boosting vaccination compliance is yet to be seen. However, independently of its effectiveness, the main drawback to Australia's policy—or, for that matter, to any kind of intervention that relies on economic

incentives to generate the desired result—is that it privileges those who can afford to forgo the state-sponsored financial benefit. Consequently, the burdens of securing the public good of herd immunity may not be equally shared across the population because of disparities in wealth. The rich can free ride off the efforts of the poor. Conclusion

A full comparative analysis o f the myriad mechanisms by which governments can and should encourage measles vaccination lies beyond the scope of this article. Having said that, a game-theoretic approach demonstrates that legislative measures similar to those implemented in California, although perhaps imperfect, are defensible on both pragmatic and moral grounds. Pragmatically, they provide a good chance of achieving herd immunity in the presence of free rider temptations. Morally, closing the loopholes that permit free riding ensures that everyone shares in the burden equally. Furthermore, in the context of the United States, Californiab legislation has the advantage over Australia's policy in being an extension of the familiar and generally accepted practice of requiring those who attend school to be vaccinated, altering only the grounds on which one can object to vaccination. To be sure, voluntary compliance with vaccination requirements would be most desirable. By removing all nonmedical vaccination exemptions, the proposed legislation infringes on the conscience of parents (as any policy that toughens up on exemptions would). However, insofar as the alternative to closing those exemptions is periodic outbreaks—and that is unacceptable—compromises to consciences are unavoidable. California's legislation is arguably the best of imperfect alternatives and represents a defensible first attempt to achieve and maintain herd immunity

Notes 1.

2. 3.

4.

S.B. 277,2015-16 Leg., Reg. Sess. (Cal. 2015); available at http://www.leginfo.ca.gowpub/15-16/ bil]/sen/sb_02510300/sb_277_bill_20150219_introduced.pdf (last accessed 1 Sept 2015). Fine P, Eames K, Heymann DL "Herd immunity": A rough guide. Clinical Infectious Diseases 2011;52(7):911-16. Plans-Rubi6 P Evaluation of the establishment of herd immunity in the population by means of serological surveys and vaccination coverage. Human Vaccines & Immunotherapeutics 2012;8(2):184-8. World Health Organization. Global Health Observatory repository: Measles (mcv) data by country; available at http://apps.who.int/gho/data/node.main.A826 (last accessed 1 Sept 2015).

5.

6.

7.

Seither R, Masalovich S, Knighton CL, Mellerson J, Singleton JA, Greby SM. Vaccination coverage among children in kindergarten—United States, 2013-14 school year. Morbidity and Mortality Weekly Report 2014 Oct 17;63(4):913-20; available at http://wwwcdc.gov/mmwr/ preview/mmwrhtml/mm6341a1.htm (last accessed 1 Sept 2015). Majumder MS, Cohn EL, Mekaru SR, Huston JE, Brownstein JS. Substandard vaccination compliance and the 2015 measles outbreak. Journal of the American Medical Association Pediatrics 2015;169(5):494-5. Wakefield AJ, Murch SH, Anthony A, Linnell J, Casson DM, Malik M, et al. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder

Browne: The Measles and Free Riders: California's Mandatory Vaccination Law

8.

9.

10.

11.

12.

in children. Lancet 1998;351(9103):637-41. Retraction in: Lancet 2010;375(9713):445. E.g., Taylor B, Miller E, Farrington C F, Petropoulos MC, Favot-Mayaud I, Li], et al. Autism and measles, mumps, and rubella vaccine: No epidemiological evidence for a causal association. Lancet 1999;353(9169):2026-9; Jain A, Marshall], Buikema A, Bancroft T, Kelly JP, Newschaffer CJ. Autism occurrence by MMR vaccine status among US children with older siblings with and without autism. JAMA 2015;313(15):1534-40. Smailbegovic MS, Laing GL, Bedford H. Why do parents decide against immunization? The effect of health beliefs and health professionals. Child: Care, Health and Development 2003;29(4):303-11. Sadaf A, Richards JL, Glanz J, Salmon DA, Omer SB. A systematic review of interventions for reducing parental vaccine refusal and vaccine hesitancy. Vaccine 2013;31(40):4293-304. Nyhan B, Reiner J, Richey S, Freed GL. Effective messages in vaccine promotion: A randomized trial. Pediatrics 2014;133(4):e835-42. For other game-theoretic treatments of vaccination, see Bauch CT, Earn DJD. Vaccination and the theory of games. Proceedings of the National Academy of Sciences of the United States of America 2004;101(36):13391-4; Shim E, Grefenstette]] , Albert SM, Cakouros BE, Burke DS. A game dynamic model for vaccine skeptics and vaccine believers: Measles as an example. Journal of Theoretical Biology 2012;295:194-203.

Free Riding and Organ Donation Walter Glannon

With the gap between the number of transplantable organs and the number of people needing transplants widening, many have argued for moving from an opt-in to an opt-out system of deceased organ donation. In the first system, individuals must register their willingness to become donors after they die. In the second system, it is assumed that individuals wish to become donors unless they have registered an objection to donation. Opting out has also been described as presumed consent. Spain has had the most successful presumed consent policy, resulting in a substantial increase in the donation rate.' Despite support for an opt-out system from the British Medical Association and other groups since 1999, the report of a task force delivered in November 2008 recommended that the current opt-in system in the UK be retained.2 The USA has also retained an opt-in system.3 Two prominent American bioethicists, Tom Beauchamp and James

13. According to the Centers for Disease Control and Prevention, one dose of the MMR vaccine is 93% effective and two doses are 97% effective. Centers for Disease Control and Prevention. Measles Vaccination; 2015 Apr 7; available at http://www.cdc.gov/measles/vaccination.html (last accessed 1 June 2015). 14. World Health Organization. Measles fact sheet; 2015 Feb; available at http://www.who.int/ mediacentre/factsheets/ fs286/englast accessed 1 Sept 2015). 15. See note 14, World Health Organization 2015. 16. See note 14, World Health Organization 2015. 17. Salathe M, Bonhoeffer S. The effect of opinion clustering on disease outbreaks. Journal of the Royal Society Interface 2008;5(29):1505-8. 18. Ireland J. Abbott government closes religious loophole for children's vaccinations. Sydney Morning Herald 2015 Apr 19; available at httpf/ www.smh.com.aufiederal-politics/political-news/ abbottgovernment- closes-religious-loophole-forchildrens-vaccinations-20150419-1mo4rv.html?utm_ content=bufferd74d36zutm_medium=social&utm_ source=twitter.com&utm_campaign=buffer (last accessed 1 Sept 2015). 19. Izadi E. Religious vaccination exemptions will completely end in Australia. Washington Post 2015 Apr 20; available at httpi/www.washingtonpost.com/news/ acts-of-faith/wp/2015/04/20/ religious-vaccinationexemptions-will-completely-end-in-australia/ (last accessed 1 Sept 2015).

Childress, claim that an opt-out policy "would not likely be adopted in the United States, and if it were adopted, it probably would not increase the number of organs for transplantation because so many citizens would opt out."4 Offering financial incentives such as paying funeral expenses for a deceased donor's family is one way of increasing the number of organs. A regulated market in organs is another.' In the UK at least, there is broader support for presumed consent. Among the alternatives to opting in, this system has the best prospect of eventually being adopted. Presumed consent would not obviate but retain the basic idea of consent, because individuals would still have the right to opt out and choose not to donate.6 One argument for presumed consent is that it would reduce the incidence of free riding. For a free rider, it is not rational to contribute to the organ pool because one's donation has a negligible effect on the number of transplantable organs, the personal cost of donating is too high, and one could refuse to donate but have equal access to an organ transplant if one

Source From J Med Ethics (2009), 35, 10: 590-591. Reprinted with permission.

II

au

111111

8 Public Health

developed organ failure.' The third reason is especially significant because in the current opt-in system access to a transplant is not conditional on donation. If it were, then it would be rational to donate. In the absence of such a condition, and given differences in people's beliefs, it may be too strong to claim that one is rationally required not to donate. A weaker and more plausible claim is that not donating is rationally permitted. Many refuse to donate organs after death because of ungrounded fears about the transplant system. On rational grounds, though, some may find the current opt-in system too costly in terms of the time and effort required to register as an organ donor. For those with strong religious objections to deceased organ donation, refusing to donate is rational because this decision is consistent with their belief in the sanctity and integrity of the body after death. There are also divergent views on donation among families. Whereas some families may find meaning in donating a deceased child's or spouse's organs, others may be distressed by the thought that donating would violate their belief in the bodily integrity of a loved one. This psychological cost would not be outweighed by any benefit to transplant recipients or any psychological benefit to families from knowing that the transplanted organs saved other people's lives. For those who object to donation, presumed consent would not threaten individual autonomy and the right to make decisions regarding one's body. They would have to make the time and effort to express an objection to having their organs procured for transplantation. In all of these cases, Tnrsonal cost rather than risk is the core issue, as there is no risk for an individual once he or she has been declared dead. This distinguishes deceased organ donation from live organ donation. The risk of short and long-term morbidity and mortality in live kidney donation, for example, is low; but it is not zero. Rationality alone is sufficient to explain why so many people do not donate. It is not because people fail to act on a moral duty but because donation is not in their own best interests. The main problem with free riding is rot that any benefit from contributing to the organ pool is not worth the cost. Rather, free riders know that they can benefit from others donating without having to incur any costs of their own. This may be unfair to those who donate, but it seems rational for one not to contribute to a collective good to which one will have access if one needs a transplant. This idea rests on the crucial assumption that one's willingness to donate makes no difference to one's access to organs. Eitl- er a sufficient number of other people will contribute to the organ pool or

they will not. In either case, not donating seems to be a rational choice for an individual. Dead donors cannot benefit from an increase in the supply of organs, because by the time they give up their organs they have ceased to exist. Mere willingness to donate does not make a difference to the organ supply; only actual donation does. However, this assumption is mistaken because the effect of donating or failing to donate is not limited to the time when one dies. If a person is in organ failure and needs a transplant, then his or her access to a transplant now is possible because of the decision of another person to donate before the need for the transplant arose. Stating on a donor registry or driver's licence that one would like to give up one's organs for transplantation does make a difference. Many donors die unexpectedly, and stating at an earlier time that one wishes to donate one's organs after death enables organ procurement. The result of more individuals stating that they wish to be deceased donors is more transplantable organs. The effect of an individual refusing to donate might seem negligible with respect to the total number of organs, but many people opting out would create a collective action problem, the result of which would be fewer organs than the already limited number currently available.8 Fewer organs would mean that a person developing organ failure and needing a transplant would have to wait longer to receive one. One's condition could deteriorate while waiting to the point at which one would no longer be a candidate for a transplant and would die. In light of the increasing probability of organ failure as one ages and the collective effect of many individuals refusing to donate, free riding by opting out of organ donation may not be a rational choice. Hypertension and type 2 diabetes have been increasing among adults in developing and developed countries, and these conditions are the main causes of kidney failure. In many cases, these conditions are not predictable and can develop in a short period of time. Hepatitis C and other viral infections have increased the incidence of liver failure, although this is not as high as the incidence of kidney failure. One might think that it is irrational to contribute to the total number of organs because it is unlikely that one would develop a condition requiring a transplant. Healthy young adults may reason in this way. Yet as we age the probability of diseases such as hypertension and diabetes increases among the general population, and the probability of an individual developing one of these diseases and kidney failure increases as that individual ages. Refusing to donate may be rational the younger one is because it is unlikely that one would develop kidney

Glannon: Free Riding and Organ Donation 433

failure at an earlier stage of life. Yet given the increasing probability of kidney failure as we age, refusing to donate may not be so rational at a later stage of one's life. This probability may be low in absolute terms. Yet in light of the collective effect of many people failing to donate over time, the cost of donating relative to the cost of not receiving a transplant makes donation a rational choice. The effect of many people opting out would further reduce the availability of organs and limit access to a transplant if one developed kidney or other types of organ failure. Each person may appear to be better off by acting alone; but all people are worse off by acting collectively in the same way. Clearly, the cost of not receiving an organ transplant and dying prematurely is greater than any cost associated with having one's organs taken from one's body after death. The organ shortage is due largely to the failure of people to donate in the current opt-in systems in the UK and USA. If Beauchamp and Childress4 are correct, the situation would not improve in an opt-out system, because many people would still choose not to donate. The collective action problem of opting out of organ donation is analogous in many respects to the collective action problem of opting out of medical research. If asked to participate in a clinical trial comparing a standard therapy with an experimental one, an individual could free ride by refusing to participate in the trial. One might believe that being a research subject would have a negligible effect on the scientific knowledge gained from the research and would not be worth the cost of participating. One might also believe that one could benefit subsequently from treatments deriving from the collective acts of other research subjects without having to be a subject. In both deceased organ donation and medical research, the benefits of an individual's participation accrue to other people. On these grounds, it seems at least rationally permissible not to participate. Yet scientific knowledge and new therapies based on that knowledge would be limited if many people refused to participate in research. Some have argued that we have an obligation to participate in medical research because participants in the past put themselves at some risk and incurred costs for scientific knowledge from which we have benefited.9 We should do the same now to benefit people in the future.1° One could present a similar argument that we have an obligation to donate our organs after death because organ donation and transplantation are two components of health care from which most of us have benefited over the course of our lives. An

optout system would be consistent with this sense of obligation, but many will fail to discharge any presumed obligation to donate. Moral obligations are not socially or legally enforceable and are not subject to sanctions if they are not discharged. Because no one can be forced to give up their organs, and because access to a transplant is not conditional on willingness to donate, those who choose not to donate would not be denied a transplant for making this choice. Still, the question is not whether there is a moral obligation to donate, but whether it is rational for selfinterested individuals not to donate. Regarding rationality, there are two differences between participation in medical research and organ donation. It may be rational for one not to participate in research because, even if the contribution is not negligible, any risk to which one may expose oneself will probably not be outweighed by any benefit. Although a subject in the active arm of a randomized controlled phase II or III drug trial may benefit if the drug is more effective than standard therapy, generally any benefit from being a subject would probably not accrue to that subject but to future patients. In contrast, there is no risk in deceased organ donation and thus no risk-benefit calculation for an individual. Yet one who refused to donate organs could be adversely affected during one's life if one developed organ failure and did not receive an organ transplant. This would be a consequence of a shortage of organs, the collective effect of many individuals refusing to donate. Those who already have chronic conditions such as hypertension or diabetes would be excluded from deceased organ donation and thus would not be part of this collective action problem. The critical age group would be those roughly between 30 and 55 years of age who are generally healthy and have none of the obvious risk factors for organ failure, because this is the period when most people decide to donate. It may seem rational for each individual in this group to free ride by ref using to donate. Yet individual actions cannot be separated from the collective actions of others. As the collective effect of many people opting out of donation would be fewer organs for transplantation, an individual who may eventually need a transplant may not receive one and thus may not be acting rationally by opting out. As noted, it may be rational for those with certain religious beliefs not to donate. For others, though, the absence of any risk, the relatively low personal cost, the limited supply of organs, and the increasing probability of developing organ failure and needing a transplant as we age make deceased organ donation a rational choice.

II

ail

111111

8 Public Health

Notes 1. Quigley M, Brazier M, Chadwick R, et al. The organs crisis and the Spanish model: theoretical versus pragmatic considerations. J Med Ethics 2008;34:223-4. 2. Organ Donation Task Force. The potential impact of an opt-out system for organ donat on in the UK: an independent report from the Organ Donation Task Force. London: ODTE 17 November 2008. 3. Childress], Liverman C. Organ donation: opportunities for action. Institute of Medicine. Washington, DC National Academies Press, 2006,205-28. 4. Beauchamp T, Childress]. Principles of biomedical ethics, 6th edn. New York Oxford University Press. 2009:110.

5. Erin CA, Harris J. An ethical market in human organs. J Med Ethics 2003;29:137-8. 6. Harris J. Organ procurement: dead interests, living needs. J Med Ethics 2003;29:130-4. 7. Tuck R. Free riding. Cambridge, MA: Harvard University Press, 2008. 8. Olson M. The logic of collective action. Cambridge, MA: Harvard University Press, 1965. 9. Harris]. Scientific research is a moral duty. J Med Ethics 2005;31:242-8. 10. Schaefer GO, Emanuel EJ, Wertheimer A. The obligation to participate in biomedical research. JAMA 2009;302 67-72.

8.3 Screening for Disease

Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario? Timothy Caulfield

There is currently little evidence that the information provided by personal genomics companies—such as 23andMe and Navigenics—on a direct-to-consumer (DTC) basis, has any real health value. To be fair, research directly on-point is lacking. However, emerging analysis suggests that the personal genomic information that can be obtained from these companies is simply not of a nature that can support useful health applications (Kraft and Hunter 2009; Jakobsdottir et al. 2009; Magnus et al. 2009). Despite this skepticism, reflected at times in both the popular press and government reports (e.g., Federal Trade Commission [ETC] 2006), DTC personal genomics receives significant attention and was even selected as the 2008 Invention of the Year by Time Magazine (Hamilton 2008). II also seems to engender a reasonable amount of interest in the general public, as highlighted by McGuire and colleagues' survey (2009). Significantly, this survey revealed that the public appears to be interested in the testing because they view it as valuable health information. Of course, this finding makes perfect sense. Given that people hear in the media that genes influence popularity, ability to empathize, and likelihood of cheating while in a relationship, it is unsurprising that they expect genetic data to provide significant health information. The McGuire and colleagues' study

highlights the degree to which this is so. This study found that an astounding 78% of those who would pursue testing would ask their physician for assistance with interpreting the data Further, 61% of respondents felt that physicians have a professional obligation to help with this interpretive process. From a health policy perspective, this data suggests the potential for a worrisome trend: the driving up of healthcare utilization costs with little actual health gain. The trend could play out as follows. Enticed by marketing strategies and the high profile of genetics in popular culture, individuals access the services of personal genome companies, hoping to obtain useful health information. They take this information to a healthcare professional (e.g., family physician) for help interpreting the data. The healthcare professional likely does not currently have the skills or knowledge to assist (Baars et al. 2005). The patient is left to interpret the data on his/her own, or with a varying degree of help from the company that sold the service. Since the data is of only marginal health value, little or no health benefit is achieved. But the interaction with the physician costs the system both time and money. One could argue that despite the lack of a scientifically verified value to personal genomic services, the results may still motivate healthy lifestyle behaviors by suggesting areas where change would be advantageous (e.g., "your genes say you should eat more broccoli, exercise and sleep more!"). While this result is possible (indeed, some people appear to be motivated by horoscopes), the evidence on behavior change is not encouraging. Bottom line It is tough to get people to change

Source: From The American Journal of Bioethics 9 (2009), 6-7,48-50.

Caulfield: Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario?

unhealthy behaviours, even in the face of health information that tells them change is a good idea (Caulfield et al. 2009). There is some evidence that genetic risk information will motivate non-health behavior change, such as the purchasing of insurance (Zick et al. 2005), but a meaningful and sustained change in diet and exercise likely requires more than receiving a test result (Story et al. 2008). This seems particularly so because the risk information provided by DTC companies generally has minimal predictive power (Magnus et al. 2009). For example, would an increased risk of 0.5% over a background risk of 1% motivate long-term behaviour change? It is also worth considering whether test results might have the opposite impact, providing individuals with a justification for fatalistic inaction (Claassen et al. 2009) or a false sense of security via the presence of a "health protective" gene. In other words, could genomic tests lead to less healthy behaviour? A speculative concern, certainly, but humans do have a bit of a tendency toward complacency and the adoption of unhealthy habits. It may not take much—even a genetic test of marginal predictive value may suffice—to nudge us further toward a sloth-like lifestyle. To classify this situation as a "worst-case scenario," as I do in my title, is likely a bit alarmist. However, there seems little doubt that a health policy issue is germinating, if not already starting to bloom. The direct costs to the healthcare system, such as the cost of a visit to a family physician or, perhaps, anxiety-driven follow-up testing, may not be substantial. Yet, in this time of fiscal restraint, can any cost be ignored? In countries with a public healthcare system, such as my home country of Canada, this situation is particularly worrisome. Consumers would be paying for information from private

providers and then accessing the publicly financed system for interpretive help. What is needed to respond to and moderate this issue? It seems inevitable that healthcare providers will be asked by patients to assist in the analysis of the information generated by DTC companies. Accordingly, it is axiomatic that physicians should be provided with the resources necessary to deal with patients' questions (Guttmacher et al. 2007). While this measure will not erase the cost of a physician visit, it will help make the visit more useful and, perhaps, lesson the likelihood of future costs. Even more preferably, as noted by others, individuals should visit a knowledgeable healthcare provider before ordering a test, to discuss the potential health implications and limits of DTC testing (FTC 2006). The need to provide resources and education to physicians is a common recommendation (Burke et al. 2002; and Burke et al. 2009) and is supported by the McGuire data (McGuire et al. 2009). In addition, companies should be encouraged to offer comprehensive and accurate information about the limitations of their services. Finally, there is a need for publicly available and widely disseminated independent information about genetics and DTC testing services, something that is already starting to emerge (FTC 2006). If framed appropriately, these suggestions may help to lower the health expectations of future consumers. This brief comment ends with the traditional academic plea for more research. So many of the concerns associated with DTC genetic testing remain largely conjecture. As it appears here to stay, we need more research on its potential impact on the healthcare system (Mouchawar et al. 2005) and on individuals.

References Baars, M.J., Henneman, L., and Ten Kate, L.P. 2005. Deficiency of knowledge of genetics and genetic tests among general practitioners, gynecologists, and pediatricians: A global problem. Genetics in Medicine 7 (9): 605-160. Burke, W, Acheson, L., Botkin, J., et al. 2002. Genetics in primary care: A USA faculty development initiative. Public Health Genomics 5 (2): 138-146. Burke, S., Martyn, M., Stone, A., Bennett, C., Thomas, H., and Farndon, P 2009. Developing a curriculum statement based on clinical practice Genetics in primary care. British Journal of General Practice 59 (559): 99-103. Caulfield, T., Shelley, J., Alfonso, V., and Bubela, T. 2008. Nutrigenomics and the promise of prevention: Representations and realities. Health LawJournal special edition: 41-66. Claascen, L., Henneman, L De Vet, R, Knol, D., Marteau, T., and Timmerman, D. 2009. Fatalistic responses to

different types of genetic risk information: Exploring the role of self-malleability. Psychology & Health (January 16, 2009). Available at: http://www.tandfonlinecom/doi/ abs/10.1080/08870440802460434 (accessed March 7, 2009). Federal Trade Commission (FTC). 2006. At-Home Genetic Tests: A Healthy Dose of Skepticism May Be the Best Prescription. Available at: httpi/www.ftc.gov/ bcp/edu/pubs/consumer/health/hea02.pdf (accessed March 7, 2009). Guttmacher, A.E., Porteous, M.E., and McInerney, J.D. 2007. Educating health-care professionals about genetics and genomics. Nature Reviews Genetics 8: 151-157. Hamilton, A. October 30, 2008. TIME's Best Inventions of 2008: Invention of the Year—The retail DNA test. Time Magazine. Available at: http://www.time.com/time/specials/packages/ article/0,28804,1852747_1854493,00.html (accessed March 8, 2009).

II JII

111111

8 Public Health Jakobsdottir, J., Gorin, M.B., Conley, Y.P., Ferrell, R.E. and Weeks, D.E. 2009. Interpretation of genetic association studies: Markers with replicated highly significant odds ratios may be poor classifiers. PLoS Genetics 5(2): e1000337. Kraft, P., and Hunter, D. 2009. Genetic risk prediction—Are we there yet? New England Journal of Medicine 360: 1701-1703. Magnus, D., Cho, M. K., and Cook-Deegan, R. 2009. Directto-consumer genetic tests: Beyond medical regulation? Genome Medicine 1: 17. McGuire, A., Diaz, C., Wang, T., and Hilsenbeck, S. 2009. Social networkers' attitudes toward direct-toconsumer personal genome testing. American Journal of Bioethics 9 (6-7): 3-10.

The Ethics of Screening: Is "Screeningitis" an Incurable Disease? Darren Shickle and Ruth Chadwick

Most Western societies are concerned about health care expenditure in view of an ageing population and a growth in the number of effective health care interventions. Scarcity of resources means that health care systems are expected to produce more health gain with relatively fewer resources. It is assumed that f "prevention is better than cure," then it will also be "cheaper." Cost benefit analysis tends to consider tangible costs and benefits which accrue directly to the health service since these are easier to measure and are the only ones that will appear on balance sheets at the end of the financial year. However, there are ethical implications that arise from ignoring the intangible and indirect aspects of this equation. Screening may be defined as a selection procedure for further investigation, applied to a population of asymptomatic individuals, with no personal or family history to suggest that they are at a higher risk of the disease than the rest of the population. It may be a form of "secondary prevention," which aims to detect disease in presymptomatic individuals in order to provide more effective treatment in the early stages of disease, or "primary prevention," which aims to identify risk factors or carrier states. Individuals offered screening would usually consider themselves to be "healthy" (unless they have coincidental illness). We tend not to consider our health until either we or someone close to us loses it. The offer of screening implies that we may not be healthy after all. This may result in a rise in anxiety, even if small and short-lived, until the result

Mouchawar, J., Hensley-Alford, S., Laurion, S., et al. 2005. Impact of direct-to-consumer advertising for hereditary breast cancer testing on genetic services at a managed care organization: A naturally-occurring experiment. Genetics in Medicine 7 (3): 191-197. Story, M., Kaphingst, K.M, Robinson-O'Brien, R., and Glanz, K. 2008. Creating healthy food and eating environments: Policy and environmental approaches. Annual Review of Public Health 29: 253-272. Zick, C.D., Mathews, C.J., Roberts, J.S., Cook-Deegan, R., Pokorski, R.J., and Green, R.C. 2005. Genetic testing for Alzheimer's disease and its impact on insurance purchasing behavior. Health Affairs (Millwood) 24 (2): 483-490.

comes back "negative." This anxiety-state may be described as that of the "worried well" (figure 1). Consider a population, some of whom have a disease, while others do not. The diagnosis of "disease" may require extensive investigation (or may only be made at postmortem). Such a "gold standard" test is important prior to embarking on treatment. Screening should be a simpler, cheaper, and quicker test to select for further investigation. However, cases with a positive screening-test finding may not have the disease (falsely labelled as "positive"), and similarly, disease may be present in individuals who have a negative screening result, i.e., "false negatives" (figure 2).

STOP! Are YOU Healthy? ARE YOU SURE?? Did you know that there is a malignant disease called Screeninoitis ?

DON'T PANIC There is a screening test available ... ... and it is accurate... ... MOST of the time ... ... BUT the treatment is PAINFUL!! ... AND HAS A LOW SUCCESS RATE!!!

Figure 1 Screening and the "Worried Well"

Source: From Journal of Medical Ethics 20,1 (1994), 12-18. Reprinted with permission.

Shickle/Chadwick: The Ethics of Screening: Is "Screeningitis" an Incurable Disease?

Disease

positive Finding

Present

Absent

true positive A

false positive B

false true negative negative negative D C A+C

A +B

C+D

B+D

Sensitivity = Proportion of patients with the disease in whom the finding is positive A A+C Specificity

Proportion of those without the disease in whom the finding is negative D B+D

Figure 2 Possible Outcomes of Screening Test

It is not always possible to make an absolute distinction between those individuals in whom the disease is present or absent even with a "gold standard." For some conditions there will be a continuous distribution of variables, such that at one extreme, individuals could be considered "diseased," while those at the other are "healthy," for example, in the context of hearing or visual acuity. For other biological variables, for example, blood pressure or haemoglobin, the extremes at both ends of the distribution would be "unhealthy," while those in the middle would be "normal." For the purposes of screening there should be a threshold to trigger further investigations or treatment. The position of this cut-off in the distribution should be based on the associated risk of morbidity or mortality that warrants further intervention. The choice of threshold may therefore be arbitrary or depend on the resources available. For example, a blood cholesterol of 6.5 mmo1/1 or above is considered to be associated with a significantly high risk of coronary heart disease.' However, between 25 and 36 per cent of the UK population' have a blood cholesterol above this threshold, and it would be unrealistic to treat this number of people. A screening test with high specificity would tend to detect only people with the disease, i.e., there would be relatively few false positives. Raising the threshold

would raise specificity and hence target resources at those with the highest risk of morbidity and potentially the most to gain from treatment. In contrast, a test with high sensitivity would tend to maximise identification of diseased people in the screened population, i.e., there would be relatively few false negatives, but there would also be unnecessary investigations or treatment for others. Thus an increase in sensitivity will be at the expense of specificity and vice versa. There are costs and benefits associated with each of the four permutations of true/false positive/negative (figure 3). When a screening programme is evaluated, the benefits of being a true positive tend to be considered and sometimes the reassurance associated with a negative result. However, associated costs tend to be ignored. If costs outweigh benefits, then either the balance between sensitivity and specificity should be altered or else the screening test should not be used at all. The preference for specificity or sensitivity should depend on the consequences of making the diagnosis or not. High specificity should be desirable if: 1. the diagnosis is associated with anxiety or stigma; 2. further investigations are time-consuming, painful, and/or expensive; 3. cases are likely to be detected by other means before it is "too late" for effective treatment; and 4. treatment, especially if painful or expensive, is to be offered without further investigations. In contrast, high sensitivity would be desirable if: 1. adverse consequences of missed diagnosis for the individual, for example, late treatment, might be significantly worse than early, 2. adverse consequences of missed diagnosis for society, for example, with a serious communicable disease; and 3. diagnosis is to be confirmed by other investigations so that the period of anxiety is short, or the correct diagnosis is given before treatment is started. For some screening tests the choice between sensitivity and specificity is difficult. Consider HIV (Human Immunodeficiency Virus) screening. The diagnosis is associated with high levels of anxiety and stigma, there may be no other investigations to refute the diagnosis, treatment is expensive, with many side-effects, and may be ineffective. However, there are important consequences of a missed diagnosis for society. The ethicist has a role in resolving dilemmas arising from the conflict of sensitivity and specificity. Moral conflict can arise in circumstances where an

II

AI

.III

IJILLI

8 Public Health

intervention is associated with both good and harm for an individual. Utilitarianism is an attractive theory in such circumstances, since the correct solution would depend on whether the utility arising from the sum of the benefits outweighs the disutility associated with harm. The person usually considered best placed to assess utility is the person him/herself; however, does the effect of illness cloud the person's judgement? In any case, should the autonomy of an individual with a communicable disease have primacy over the desire of society to place that person in quarantine to prevent other individuals from becoming infected? Similarly, should a couple be allowed to have a handicapped child, when prenatal screening may have detected an abnormal fetus which could have been terminated? The professional duty to both non-maleficence ("do not harm") and beneficence ("do good") is enshrined in the Hippocratic oath2: "I will use treatment to help the sick according to my ability and judgement,

but I will never use it to injure or wrong them." However, in circumstances where both good and harm are likely to result from an intervention, an objective rule may be required to resolve such an ethical dilemma. There are frequent appeals for funds to allow a child to receive a "life-saving" treatment that is not available from the UK National Health Service. It is difficult for a health planner to withdraw funding from a programme that can save life because of the consequent publicity that would be generated through "shroud waving." Vested interest groups would argue that the benefits of screening to detect one individual who may otherwise have died without early treatment would justify any harm caused. The principle of double effect permits harm to others provided the harm is indirect and was not intended, although it may have been foreseen. However, this principle has been widely criticised3" and is unhelpful in the resolution of the dilemma of whether to screen or not. It can apparently excuse the harm caused both by

True Positive

False Positive Benefits

Costs May be no increase in life expectancy if treatment is of no effect but patient may have to live longer with diagnosis because it is made earlier (lead time bias) Anxiety:"Worried ill" Stigma Reduced quality of life, for example, due to side-effects or complications of treatment

Unnecessary anxiety and stigma until correct diagnosis is made

Benefits of"sick role" (excused social responsibilities provided seek and comply with treatment)

Unnecessary investigations and treatment (timeconsuming; expensive; pain; side-effects; complications)

Explanation of symptoms: knowing the cause of symptoms so that they can be "treated" can relieve anxiety

Lingering doubts: patients tend to believe doctors! May be difficult to reassure patient that initial diagnosis was incorrect

Costs

Benefits

False reassurance ("Unworried

Spared anxiety if treatment of no benefit

May legitimizeun healthy lifestyle" Later treatment (unpleasant; more expensive; worse prognosis)

Provides an opportunity to counsel on unhealthy lifestyles:"next time could be f or real"

True Negative

False Negative

ill")

Benefits

Costs

Earlier treatment (cheaper; more pleasant; better prognosis)

Costs Anxiety while awaiting results of test ("Worried weir)

Benefits Reassurance

May legitim ise"un healthy lifestyle:for example, If my chest x-ray is normal, then smoking cannot be doing me any harm!"

Increased cost per desired outcome

Figure 3 Costs and Benefits Associated with Being a True Positive, False Positive, False Negative or True Negative

Shickle/Chadwick: The Ethics of Screening: Is "Screeningitis" an Incurable Disease?

providing and withholding screening. For example, it is accepted that all screening procedures are associated with false positives and negatives. This harm, however, although foreseen, can be described as an unintended side-effect. On the other hand, if screening is withheld, individuals who could have benefited from screening, via earlier treatment of their disease, will be denied these advantages. Thus the decision to withhold arguably causes harm which is foreseen but not intended. The principle of acts and omissions might be invoked to escape responsibility for the harm that results from a decision to withhold screening. According to this principle, failing to prevent some harm is less bad, morally speaking, than actively bringing it about,6 and the withholder has not, it might be claimed, actively caused the harm. Like the principle of double effect, this principle has been heavily criticised.' In the light of the evidence that haemophiliacs can be infected with the HIV virus as a result of transfusions of infected blood, for example, it is arguably just as bad to fail to screen the blood, with the result that, say, 10 per cent will be infected, as it would be to inject those 10 per cent with the virus deliberately. If we fail to prevent something bad from happening where we are in a position to do so, we are responsible. Harris8 identifies "positive" moral responsibility where something happens because someone did something and "negative" responsibility because someone allowed or permitted or let it happen. He concludes that there is equal responsibility for what happens because of our actions and what happens because of our inaction or non-action. Thus, it would be an inadequate defence to argue against a screening procedure on the grounds of the need for a needle-prick or anxiety that may be caused to the "worried well," since there is "negative" responsibility for the "true positives" who would otherwise have been identified. Unlimited Moral Demands The problem that arises from the abandonment of the principle of acts and omissions is that it seems to place unlimited moral demands upon us. It might indicate a moral requirement to carry out any screening programme that could conceivably prevent harm. This has led to the search for ways of placing limits on what we are required to do by attempting to draw moral distinctions between different kinds of cases. In the screening context it might be argued that it is morally significant that those who are harmed through screening are identifiable, whereas those that are harmed through failure to screen are not.

For it is not possible to know which members of a population will be "true positives" before screening, and whether these individuals will benefit from the treatment offered. Glover has shown that this consideration is irrelevant, morally speaking: whether the harmed individual can be identified prospectively or retrospectively makes no difference to the wrong of harming them.9 Harris maintains that an individual remains "negatively" responsible even if he was not aware that he could have prevented harm, although he recognises that the individual should not be blamed. Another attempt to limit the moral demands upon us is to try to establish priorities between different kinds of moral requirements. Frankena, for example, proposed that beneficence includes four elements'°: 1. 2. 3. 4.

One ought not to inflict evil or harm (what is bad). One ought to prevent evil or harm. One ought to remove evil or harm. One ought to do or promote good.

Frankena ranked the four elements such that in cases of conflict, non-maleficence took moral precedence over beneficence. Foot's proposed priority of what she calls "negative" duties to refrain from injuring someone and "positive" duties to help them is similar.' There are at least two problems in applying these distinctions to the decision concerning whether or not to carry out a screening programme. First, the primacy of non-maleficence is not absolute. For example, a surgical incision will cause injury, but the consequence of the operation may be that the patient's life is saved. Thus, the injury caused by taking a blood sample as part of a screening programme may be considered trivial in comparison to the potential benefits. Second, it is not always clear whether a decision not to do something, such as carry out a screening procedure, should be described as a maleficent infliction of injury or a failure to do good. It is easy to imagine contexts in which the failure to screen blood before transfusion to haemophiliacs could accurately be described as a maleficent infliction of injury. It is difficult, therefore, to accept any of these proposals for drawing moral distinctions. There must be room for a trade-off between the harm brought about by screening and the harm brought about by failure to screen. This suggests that cases should be considered on their merits. Even if we accept, as we should, that there is negative responsibility for the harm we fail to prevent by withholding screening, this does not mean

II ill

111111

8 Public Health

that we should carry out every screening programme that could conceivably prevent harm. A useful test is that proposed by Singer: "If we :an prevent something bad without sacrificing anything of comparable moral significance, we ought to do it."12 There is a danger that advances in medical technology will be translated into screening services and offered by "enthusiastic" scientists to an "unsuspecting" public, without consideration of the consequences. Even if the society is consulted, there is a general expectation among the public that medical advance is generally "a good thing." The harm consequent from a screening test for any individual will usually be trivial in comparison with the potential harm from not offering screening and saving a life by providing effective treatment. However, the way in which individuals perceive risk is complex: some are "risk lovers" while others are "risk averse" or "risk neutral."13 For example, the risk of an HIV-positive doctor infecting one of his/her patients is very small, yet there has been demand within the media for regular, universal HIV testing of healthcare professionals. It has been argued that patients are more concerned with the nature of risk than with its precise numerical value: "let us not get so preoccupied with the risks that we lose sight of the stakes."14 However, individuals may place differing amounts of disutility on these states. For example, some patients may be happy to receive treatment for mild hypertension, even though there is only a small decrease in risk of a "stroke" or "heart attack" following treatment; other individuals may be less willing to accept the label of "ill-health." An individual may choose screening even though there is a high risk of relatively trivial harm and only a small chance that he/she will benefit. Using utilitarian principles, the net utility or disutility for society arising from an intervention is equivalent to the sum of the change in utility to its individual members. However, the sum of trivial disutility for the many may exceed the large utility gain for the few. Others have raised similar concerns about the potential net disutility associated with health promotion. For example, Skrabanek pointed out that "the proverb, 'a stitch in time saves nine,' may be sound advice for mending socks but it makes little sense if a thousand people need one stitch (in its medical equivalent) :o save one person from nine stitches." It is, however, unreasonable to expect the individual to include "opportunity costs" in his/her consideration of providing consent, i.e., resources can only be spent once, and hence benefits may have to be foregone by others.

Counselling Should Not Be Prescriptive There are further problems in placing the responsibility of whether benefits outweigh costs upon the individual. Counselling should, as far as possible, facilitate the autonomous decisions of the individual, for example, allowing informed consent to be obtained for screening. It is unlikely however, that any individual could absorb all the necessary information during a few counselling sessions, especially if this is during a time of anxiety. While counselling should not be prescriptive, the fact that a provider of health care places a positive value on screening such that the service is offered may give the counsellee the impression that the "correct choice" of action would be to accept screening. Thus arguably, even if a counsellor attempts to give information in a neutral fashion, counselling cannot be truly "non-directive." We would not wish to argue that screening should not be offered because information about risk, utility, opportunity costs, etc. is too complex, nor that we should not encourage individuals to include such considerations in their decision-making. Unless specified in law, we believe that screening should not be overtly directive, although some may disagree if the consequences of the condition in question are sufficiently serious. However, it should be recognised that the obtaining of apparently informed consent does not remove any responsibility for harm from those who offer screening. There is a duty on health-service planners, however, to offer screening tests that the majority of the target population would wish to accept. The principles proposed by Wilson and Jungers' in the mid-1960s have attracted much criticism. However, these criteria may be useful guidelines for the development of screening programmes: 1. The condition sought should be an important problem. 2. There should be an acceptable treatment for patients with recognised disease. 3. Facilities for diagnosis and treatment should be available. 4. There should be a recognised latent or early symptomatic stage. 5. The natural history of the condition, including its development from latent to declared disease, should be adequately understood. 6. There should be a suitable test or examination. 7. The test or examination should be acceptable to the population.

Shickle/Chadwick: The Ethics of Screening: Is "Screeningitis" an Incurable Disease?

8. There should be agreed policy on whom to treat as patients. 9. The cost of case-finding (including diagnosis and subsequent treatment of patients) should be economically balanced in relation to the possible expenditure as a whole. 10. Case finding should be a continuous process and not a "once and for all" project. Wilson and Junger used the term "principles" for "ease of description rather than from dogma." It is unlikely that any screening programme will be able to fulfil all of these criteria to everyone's satisfaction. The question therefore arises as to whether each criterion has equal merit or whether there is a hierarchy of importance. Wilson and Junger felt that "of all the criteria that a screening test should fulfil, the ability to treat the condition adequately when discovered, is perhaps the most important." However, it is important not to be too restrictive in our thinking about what represents an "adequate" treatment. For example, with some cases of prenatal diagnosis, couples may benefit from genetic counselling such that they may "prepare themselves" for the birth of an affected child. In this circumstance, the counselling could be considered to be a form of treatment. An analogous question is whether termination of pregnancy can be considered a treatment and if so for whom. However, the debate on that question has been explored elsewhere.'8 The requirement for an acceptable treatment is compatible with our argument for a proper consideration of all costs and benefits. The advocates of screening have a duty to demonstrate potential benefit prior to the introduction of screening. Improved prognosis following the use of an "effective" treatment will usually be the most important of these benefits. The importance of availability of facilities for diagnosis and treatment is self-evident. Informing patients that they have a treatable disease but withholding any intervention is likely to cause more harm than not performing screening. Helpful Pointers The remaining "principles" are, in our view, helpful pointers as to which screening programmes are desirable and, if fulfilled, will describe situations where net utility is likely Wilson and Junger recognise the difficulty in defining "importance" and include conditions with high prevalence and those which have very serious

consequences for the individual or society as a whole. Screening for a condition which is not "important" is unlikely to result in net utility or accrue more utility than other interventions competing for scarce resources. However, it may be worth considering screening for relatively trivial conditions, providing screening is cheap, does not cause anxiety, and there is an effective treatment with few side-effects. Wilson and Junger do not, however, specify whether patients, a third party, or society as a whole, should prioritise importance. John Stuart Mill, in his essay On Liberty, argues that "there is no one so fit to conduct business, or to determine how or by whom it shall be conducted, as those who are personally interested in it."19 We have reservations about applying this principle of utilitarianism in the context of health-care resource allocation. It is difficult for an individual patient to be objective as to whether his/her health problem is more important than that of another patient or whether he/she deserves scarce resources in preference to others. Attempts have been made to involve society in the prioritisation process, for example, in Oregon, USA.2° However, such consultation was expensive and timeconsuming, and the complexity of the exercise meant that there were inconsistencies in the initial priority-list. For example, treatment for crooked teeth was ranked higher than treatment for Hodgkin's lymphoma. In view of these difficulties, we see little alternative to the existing decision-making process, provided that this is in consultation with experts and lay organisations. We recognise that this will be biased by the social values of those involved. The criteria for suitability and acceptability of a test for screening will vary from one condition to another. While the test should have high validity and repeatability, a compromise between sensitivity and specificity may be required. A test associated with high costs (both tangible and intangible) may be acceptable to an individual or society as a whole if the condition is severe and an effective treatment is available. For example, most women will endure the indignity and embarrassment associated with taking a cervical cytology smear if it means that carcinoma of the cervix may be prevented. The requirement for an understanding of a condition's natural history and the recognition of a latent stage are important in the development of techniques for diagnosis and treatment but in themselves are not essential. Screening for HIV is performed, yet the natural history of acquired immunodeficiency syndrome (AIDS) is still not "adequately understood."

III

442

II I NJ

8 Public Health

The balancing of economic costs resulting from case-finding with those due to possible expenditure as a whole is compatible with our modified utilitarian approach and with the principle of justice. If screening is available for one individual, then justice would expect screening to be offered for another individual who has a similar risk of having the disease and a similar potential to benefit from treatment. If a screening programme is considered to be desirable, why then should future patients not benefit from screening? Wilson and Junger argue that screening should be a continuous process, since the "start-up" costs associated with screening (providing accommodation, purchas:.ng equipment, training staff, etc.) are usually large in comparison with the marginal costs of each test performed. While these points are true, screening for prevalent (existing) cases requires different considerations from screening for incident (new) cases. Thus, the case-yield and the consequent utility may be high in the first wave of screening if the natural history of a condition is slow, and hence there may be a large number of undetected cases in a population. However, the number of new cases that develop during a screening cycle may be small, and hence the consequent utility for these few cases may not outweigh disutility or opportunity costs.

Wilson and Junger Criteria Endorsed

An assessment of net utility will assist in choosing service options which will result in the "most happiness" for the population. We recognise that assigning utility values within the cost/benefit equation is difficult. Obtaining a consensus as to the objective of a screening programme is an essential first step. For example, what is the purpose of genetic-carrier screening-empowerment of the individual in making reproductive choice or reducing the number of births of handicapped individuals? We endorse the Wilson and Junger criteria, since if the majority are fulfilled (especially those relating to effectiveness of treatment), then the conclusions will approximate to those of a utilitarian assessment. The principle of justice might suggest that access to health care is a moral right for all. However, scarcity of resources means that there must be prioritisation. There is no reason why philosophers should not be interested in the resolution of moral dilemmas in order to maximise utility from limited health care resources. Such choices have to be made, and it is better that such decisions are made with an explicit, ethical mechanism rather than them depending on the values of a few health service planners.

References

1. Standing Medical Advisory Committee. Blood cholesterol testing: the cost effectiveness of opportunistic cholesterol testing. London: HMSO, 1990. 2. Quoted in: Beauchamp T L, Childress J E Principles of biomedical ethics [3rd ed). New York Oxford University Press, 1989: 120. 3. See reference 2: 127-134. 4. Glover] C B. Causing death and saving lives. Harmondsworth: Penguin, 1977: 86-91. 5. Foot P The problem o f abcrtion and the doctrine o f the double effect. In: Rachels J, ed. Moral problems [2nd ed). New York: Harper & Row, 1975. 6. See reference 4: 92-116. 7. See reference 4 and also Harris J. Violence and responsibility. London: Routledge & Kegan Paul, 1980 passim. 8. See reference 7: Harris and ilso Harris J. The value of life: an introduction to medical ethics. London: Routledge and Kegan Paul, 1985: 28-33. 9. See reference 4: 100-102. 10. Frankena W. Ethics [2nd ed). Englewood Cliffs, NJ: Prentice-Hall, 1973: 47. 11. See reference 5: 59-70.

12. Singer P Practical ethics. Cambridge: Cambridge University Press, 1979: 169. 13. O'Brien B. "What are my chances doctor?"-a review of clinical risks. London: Office of Health Economics, 1986. 14. Fraser F C Evolution of a palatable multifactorial threshold model. American journal of human genetics 1980; 32: 796-813. 15. Skrabanek P Why is preventive medicine exempted from ethical constraints? Journal of medical ethics 1990; 16: 187-190. 16. Clarke A. Is non-directive genetic counselling possible? Lancet 1991; 338: 998-1001. 17. Wilson J M G, Junger G. The principles and practice of screening for disease. Public Health Papers, 34. Geneva: World Health Organization, 1968: 26-39. 18. Clarke A. Response to: What counts as success in genetic counselling? Journal of medical ethics 1993; 19: 47-49. 19. Mill J S. On liberty. In: Gray J. On liberty and other essays. Oxford: Oxford University Press, 1991: 121. 20. Dixon J, Welch H G. Priority setting: lessons from Oregon. Lancet 1991; 337: 891-894.

Small: Canada's Highest Court Unchains Injection Drug Users

8.4 Harm Reduction Programs

Canada's Highest Court Unchains Injection Drug Users; Implications for Harm Reduction as Standard of Healthcare Dan Small

The philosophers have only interpreted the world in various ways; the point, however, is to change it.' After a long legal and cultural battle, North America's only supervised injection facility, Insite, is finally safe from arbitrary political interference. This was a case where personal experience, activism, advocacy medicine, and science stood side-by-side to protect the rights of even the most marginalized members of the community to life-saving healthcare. The case began at the Supreme Court of British Columbia and eventually made its way to the Supreme Court of Canada (ScC). At the centre of the case were the personal stories of people who relied on supervised injection to stay alive together with testimony from scientists, physicians, healthcare officials, and the operators of Insite. On 30 September 2011, Supreme Court of Canada drew a legal line in the sand that highlights the rights of people with addictions to the security of their person under section 7 of the Charter of Rights and Freedoms (Charter of Rights)? The Charter enshrines the values of Canadian culture regarding the rights of individuals with respect to the provincial, federal, and territorial governments. The judges of Canada's highest court are appointed from a wide variety of political backgrounds. The decision was unanimous and reinforced the foundation of our understanding of addiction as a healthcare matter in Canada. As a caveat, I am not a distant academic examining supervised injection from the point of view of a removed observer. I am part of the senior management of the non-profit organization that founded and operates Insite, the PHS Community Services Society (PHS), and, as such, I have been intricately involved in the development, set-up, management, and advocacy for Insite. I am a participant observer, and so this commentary is written from the point of view of my personal experience.

The journey for Insite has been wrought with challenges because it confronts our inner web of belief about how to best approach addiction. Supervised inj ection exists in a moral minefield at the very heart of our culture. By culture, I am speaking about what we believe to be right and wrong, the implicit and explicit values that are the building blocks of our understanding, practice, and societal approaches to people with addictions.3 As such, it is my belief that this legal victory is about something much more fundamental than the legal, medical, or scientific issues that arise from it. It points the way towards a shift in our cultural orientation that allows for addiction to be constructed as a social issue best addressed, metaphorically, by the Chief of Medicine rather than the Chief of Police. This cultural change is best illustrated by a story from only a decade ago when harm reduction was not widely accepted or understood in Canada. Harm reduction innovations involve attempting to reduce harms associated with drugs, such as fatal overdoses, but to do so without necessary eliminating the use of drugs (abstinence). The Harm Reduction Journal provides the following definition: We define "harm reduction" as "policies and programs which aim to reduce the health, social, and economic costs of legal and illegal psychoactive drug use without necessarily reducing drug consumption."4 In the year preceding the opening of Insite in 2003, I was collaborating on a draft press release in response to local opposition to harm reduction. The press release simply stated that addiction is, primarily, a healthcare matter. At that time, the notion of publicly stating that addiction is a healthcare rather than criminal justice issue was so controversial that I could not convince anyone in the establishment to lend their name to the press release. In the end, two community activists agreed to sign their names to what was, at the time, a provocative public statement. Now, ten years later; this has become an established legal fact in Canada. Addiction is a healthcare matter. The PHS initiated the case at the Supreme Court of British Columbia in 2008 to protect the program from closure by a conservative federal government. We did

Source From Harm Reduction Journal 9, 34 (2012). Reprinted with permission.

II

all

III Bad

8 Public Health so at a time when there was no formal institutional sup-

port for a struggle to protect Insite in the courts. The only intervener was the British Columbia Civil Liberties Association. Despite their role in co-managing Insite, Vancouver Coastal Health (vcx) counseled our organization not to turn to the courts to protect the program and refused to provide any formal assistance for the PHS to legally defend it. In the initial case, neither [the] Vancouver [Coastal] Health Authority (vcx) nor the Attorney General of British Columbia sought intervener status. Fortunately, the PHS was able to obtain pro bono representation by three lawyers: Monique Pongrecic-Speier, E Andrew (Drew) Schroeder, and Joseph Arvay, all of whom nobly took on the case when it felt like Insite had been backed against the wall. A related and important case, entered by the Vancouver Area Network of Drug Users (Respondent/Appellant on cross appeal) was heard jointly. The PHS case focused on two overarching themes. The first theme related to the division of powers between the federal and provincial governments and essentially argued that the operation of Insite falls under the jurisdiction of the Province of British Columbia. The second argument pertained to the Charter. The Charter is the first part of the Constitution of Canada and contains a passage of particular relevance to Insite.2 This portion of the Charter, section 7, i s of central importance to Insite and states that: "Everyone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with :he principles of fundamental justice"' (p. 4). The PHS argued the federal Health Minister's withholding of an exemption from the Controlled Drugs and Substances Act (cDSA) that was required at that time for Insite's operation was prejudicial and arbitrary. More so, it jeopardized the life chances of people who need the facility to access life-saving healthcare. The case at the BC Supreme Court laid the groundwork for the facts that would later form the foundation for the landmark ruling at the Supreme Court of Canada. Given the case's importance, it is worth examining at a high level. There are, in my view, four key findings of fact in this first case. The first key fact pertains to the notion that addiction is a healthcare matter. The Government of Canada conceded this as an indisputable fact. The presiding judge at the BC Supreme Court, Justice Ian Pitfield, highlighted this absolutely critical cultural admission in his Reasons for Judgment when he stated: "drug addiction is an illness' (p. 20). The declaration of addiction as an illness allows for the devotion of healthcare resources to addressing it.

The second fact that was established was that drugs, in and of themselves, do not necessarily cause serious interruptions in health. Rather, it is the method, mechanism, and context within which drugs are ingested that brings about danger: Controlled substances such as heroin or cocaine that are introduced into the bloodstream by injection do not cause Hepatitis C or HIV/AIDS. Rather, the use of unsanitary equipment, techniques and procedures for injection permits the transmission of those infections, illnesses or diseases from one individual to another" (p. 33, para. 87). This is the foundation of supervised injection as an intervention. The point of intervention focuses on reducing the harms associated with drug use without forcing abstinence as a precondition for receiving healthcare. The idea that drugs, as substances, are not automatically intrinsically evil or dangerous has been a culturally controversial notion. There are, broadly speaking, two competing overarching narratives about addiction.3 The first narrative focuses, essentially, on drug use and the drugs themselves as intrinsically dangerous. The second does not centre on the drugs themselves but instead concentrates on the way drugs are administered (e.g., clean versus unclean syringes) and the psycho-socio-cultural context of their use (such as criminalization, poverty, and mental pain). This distinction can be traced to the work of psychologist Bruce K. Alexander who first discussed two orientations for constructing addiction 30 years ago.' The first orientation essentially constructs addiction as follows: first, a person takes a drug, then, eventually, the drug takes the person as a result of repeated exposure. Understanding addiction in terms of an exposure to drug concentrates on the eradication of drugs as the point of intervention. Following on from this reasoning, it is the reduction in drug exposures that will ultimately reduce or eliminate addiction. The rival narrative about addiction, described in terms of coping or adapting by Alexander and his colleagues, constructs the problem altogether differently. According to this alternative explanation, people take drugs; drugs do not take people. This perspective maintains that people misuse drugs due to impoverished conditions and psychosocial pain that require extraordinary coping strategies. In turning focus away from the dangers of the drugs themselves to the ways in which they are being used, the original court decision was aligned with this latter orientation.

Small: Canada's Highest Court Unchains Injection Drug Users

The third legal fact pertained to the notion that effective medical interventions are available to measurably reduce the harms of addiction. The primary healthcare intervention in Insite is the provision of sterile injection equipment and the supervision of injection: The risk of morbidity and mortality associated with addiction and injection is ameliorated by injection in the presence of qualified health professionals° (p. 33, para 87).

injection is sacrosanct. Without supervised injection, people might perish from fatal overdoses before realizing the opportunity to one day pursue detox, treatment, and abstinence. Instead, all that would be left would be a mortality statistic: a faint reminder that they ever lived. In summary, there were four legal facts that go to the heart of a particular cultural understanding of addiction:

1. Addiction is a healthcare matter. 2. Drugs do not cause deadly HIV, HCV, and fatal overdoses: unclean needles and unsupervised inThe finding of this fact, based on the scientific and medical evidence before the court, established that a j ection do. supervised injection facility helps to prevent disease 3. Supervised injection is effective at preventing and death. Contrary to its popular characterization morbidity and mortality. Harm reduction opens as an isolated program, Insite also offers detox and the door to a range of healthcare (e.g., detox, treatment). treatment on site. The final fact is perhaps the most culturally con- 4. Abstinence, though laudable, is not compulsory for effective healthcare interventions, with measurable troversial because it foregrounds the fact that effective healthcare interventions exist for addiction that do not outcomes (e.g., such as saving lives by intervening in otherwise fatal overdoses or preventing fifv) demand abstinence. Justice Pitfield understood that for addiction. The idea of condoning or enabling while Insite is not traditional treatment, it is, nonetheless, important healthcare: addiction with supervised injection takes second place to keeping people alive. While users do not use Insite directly to treat addiction, they receive services and assistance at The establishment of these four key facts, in my view, Insite which reduce the risk of overdose that is a laid the groundwork for both a legal and a cultural vicfeature of their illness, they avoid risk of being in- tory with respect to the notion of supervised injection. As the case advanced, it gathered cultural mofected or of infecting others by injection and they gain access to counselling and consultation that mentum as part of a growing acceptance of a particular may lead to abstinence and rehabilitation. All of understanding about addiction as described above. At this is healthcare° (p. 51, para. 136). the final stage, 14 interveners had joined the proceedings including the Vancouver Coastal Health Authority, This recognition of this fact feeds into the cultural anx- Canadian Nurses Association, British Columbia ieties about somehow enabling or encouraging addic- Nurses' Union, Registered Nurses' Association of tion by not outlawing it with vehemence. The ruling Ontario, Association of Registered Nurses of British also addressed this culturally notorious notion of over- Columbia, Canadian Medical Association, Canadian looking addiction: Civil Liberties Association, Canadian HIV/AIDS Legal Network, International Harm Reduction AssociaSociety cannot condone addiction, but in the face tion, CACTUS Montreal (a non-profit organization of its presence it cannot fail to manage it, hope- dedicated to providing non judgmental assistance fully with ultimate success reflected in the cure of and risk reduction for at risk individuals including the addicted individual and abstinence" (p. 54, those who use illicit drugs, street involved youth, sex para. 144). trade workers as well as transvestite and transsexual persons), Canadian Public Health Association, British The rival perspective, abstinence at all cost, would Columbia Civil Liberties Association, Attorney Genpresumably withhold supervised injection as a health- eral of Quebec and Dr. Peter AIDS Foundation. The care intervention even if it resulted in preventable fatal Attorney General of British Columbia was a respondent overdoses. This was the very reason the PHS entered with regard to the doctrine of inter jurisdictional the courts: we believe that harm reduction is a door- immunity that was won at the Appeal Court level and way into treatment, detox, and abstinence and that argued for localized control over Insite as a provincial the safeguarding of human life offered by supervised initiative.

MI Ali II

.1

II

.al

IIILA

8 Public Health

Sadly, there has always been a psychosocial phenomenon that is culturally expunged or relegated to the shadows due to the moral anxieties that it creates in the wider community. This has been true in the case in the present and past with issues or experiences that make people uncomfortable such as death, sexuality, mental illness, or addiction, which are sequestered in social life and institutional settings.8 This process of sequestering, or hiding away of the social phenomena that alarm or anger us, also takes its shape in the form of cultural erasures and silences, things that are unsaid which can, in actuality, be more influential than what is said.9 Addiction is just such a phenomenon; people with addictions have been sequestered, silenced, and erased from positive social life. Their personhood has been so dramatically undermined that their identities are sometimes socially spoiled,1° leaving them metaphorically chained. Throughout the education campaign to save Insite, we attempted to combat this identity erosion by highlighting the personhood of people living with addictions by echoing the idea that everyone living with addictions was someone's son or daughter. The field of mental health provides an interesting analogy with respect to the process of liberation of people with pain from societal shackles of stigma and incarceration. Psychiat.:y's approach to people with mental illnesses transitioned towards humanistic treatment in the 1780s under the leadership of more humanistic psychiatrists and reforms in the mental health field. Physician Vincenzo Chiarugi (1759-1820) led a movement espousing the humane treatment of the mentally ill, which took its first applied step, in practice, in 1788 with the opening of Hospital Bonifacio in Florence, Italy, where he served as the physician director.'' Chiarugi's approach was in keeping with the goals of Grand Duke Pietro Leopold of Tuscany, a socially conscious aristocrat who ordered the establishment of Bonifazio, and was predicated on respectful and humane treatment.12 He is a significant figure in the history of humanistic medicine and one of the fathers of compassionate psychiatry. Chiarugi is thought to be the first figure to forbid the use of chains to restrain the mentally ill (a policy which he established during his role as physician director of Santa Doretea hospital before 1793).12 Similarly, Jean-Baptiste Pussin (1746-1811) and Madame Marguerite Pussin (1754—?) helped to infuse compassion into the practice of mental health.12,13 After having been a patient at Bicare hospital in the suburbs of Paris, Jean-Baptiste Pussin went on to become the director of a psychiatric ward from 1785 to 1802 during which time he implemented a series of compassionate improvements in the treatment of mentally ill.14

Pussin outlawed the employment of chains to imprison the mentally ill in 1797 while serving as the governor of Bicetre.12 Pussin is an important forerunner in the history of humanistic mental health services. In popular culture, physician Phillippe Pinel is widely thought of as being the first individual to liberate the mentally ill from chains. The renowned painting Madwomen at the Salpetriere, painted by Tony-Robert Fleury, has helped to immortalize this legend. The painting shows Pussin removing the chains of psychiatric patients while Pinel looks on and symbolizes a transition towards more humanistic approaches to the mentally ill in the 18th century. Today the painting hangs in the entrance to the Library Charcot in the Salpetriere hospita1.14 In fact, it was Pussin who inspired Pinel to ban the use of chains for detaining people living with mental illnesses.12,15 After having worked at Bicetre between 1793 and 1795, Philippe Pinel was so inspired by the work of Pussin that he credited him with the emancipation of the mentally ill and the first actual application of humanistic psychiatric treatment. When Pinel moved to Salpetriere, the largest hospital in Paris, he established a post for Pussin, who took up the position there from 1801 until his death in 1811.14 At Salpetriere in Paris, Pussin and Pinel worked together to apply humanistic approaches to psychiatric treatment. There is some humanistic truth at the heart of legendary characterization of Pinel as the person that liberated the mentally ill from their chains. He did, in fact, show significant leadership by moving away from abandonment and brutal imprisonment to a therapeutic approach based on medical science and compassion.12 Similarly, the Insite legal case helps to make a similar transition from the cruelty of criminalization in addiction to a healthcare model where people with addiction have fundamental rights to life-saving healthcare. The mental health field transitioned from a model based on incarceration and neglect of the mentally ill to an approach based on compassion, science, and medical treatment in the 18th century. In the addiction field, this transition has taken another two centuries (21st century). The metaphor of chaining of the addicted goes beyond the symbolic. In 2009, the US incarcerated more than 400,000 individuals for nonviolent drug offenses (a greater number than those incarcerated for all offenses in the 27 nations of the European Union combined).16 The incarceration of the addicted has been s o dramatic that, when seen through the lens of epidemiology, it can be considered as a catastrophic event that has resulted in tremendous suffering and death. The epidemiological tool of years of life lost (YLL) is useful

Small: Canada's Highest Court Unchains Injection Drug Users

for examining the impact of a criminal justice approach to addiction. Drucker defines years of life lost as "the number of years between the victim's age at death and the age that his or her usual life expectancy would predict. Thus, for the average American with a life expectancy of 75 years, a child's death at age ten implies a loss of sixty-five potential years of life"16 (p. 69). Building on this logic, Druckernotes that 1,513 people died representing an estimated 47,000 YLL in the Titanic disaster, 2,819 deaths representing an estimated 104,303 YLL in the World Trade Center tragedy and reasons that over the past 35 years (since the introduction of the Rockefeller drug laws in New York), more than 7 million people have been incarcerated. This translates into an estimated 14 million YLL representing 350,000 deaths in a group of the same age.16 The Insite victory is emblematic of a different cultural understanding of addiction that is supplanting a traditional one. This newer approach assumes people living with addictions are in need of healthcare rather than punishment through the criminal justice system. By focusing on the federal Health Minister's refusal to provide an exemption for Insite under the existing regulatory framework, the scc did not have to make any alterations to existing provincial and federal jurisdictions over the program. Canada's federal government attempted to argue that the federal minister of health had, technically, never "not given" a permit for Vancouver's supervised injection site and therefore never formally jeopardized its operation. However, the court ruled that it was self-evident that the federal Health Minister had every intention to close Insite: The Minister of Health must be regarded as having made a decision whether to grant an exemption, since he considered the application before him and decided not to grant it. The Minister's decision, but for the trial judge's interim order, would have prevented injection drug users from accessing the health services offered by Insite, threatening their health and indeed their lives [emphasis added]2 (p. 9). The Scc concluded that the Minister's intention to shut Insite would have threatened the lives of the people who rely on the program. It also noted the program would not have remained open had it not been for protection provided by the Supreme Court of BC. National public policy with respect to supervised injection in Canada began with the establishment

of a localized response to a healthcare emergency. Supervised injection, in this case, began from the ground up rather from a top-down policy. The goals of Insite originated out of local need, inspired by the idea that people with addictions deserved something better than death from overdose. The goals and outcome measures were simple. The program aimed to provide a doorway to life, supported housing, physicians, healthcare services and supports. The intent of the program was to provide its interventions (e.g., clean syringes, supervised injection) in an accessible way without barriers (such as abstinence or onerous intake procedures). One thing that is clearly demonstrated by the case of Insite is that science is not enough, on its own, to change public policy, especially in stigmatized areas like addiction. The simple existence of a scientific evidence base does not automatically lead to changes in policies or practice. Policy-makers and elected officials need to pay attention to the evidence base. In some disappointing instances, as the case of Insite has shown, policy-makers need to be forced to pay attention to the established facts. Conversely, scientists need to take an active role in affecting public policy when the evidence indicates the need for change.'' The SIF, for instance, had more than three dozen peer-reviewed papers associated with its evaluation.18-47 Despite the fact that much of the evaluation was paid for by the Government of Canada, they chose to ignore the scientific findings. The evidence base generated by Canada's supervised injection trial should have earned it a medical exemption (the next stage of operational permit after the scientific authorization originally granted to Insite), but it was not provided. Uncompromising advocacy, including public protest and legal challenge, was required to obtain the permit. In this way, science in healthcare and applied research are not the same.' While science in healthcare is portrayed as "objective," applied research is seen as rooted in the context of the community needs. From my perspective, science in healthcare needs to move more towards clinical application. Yes, it needs to be sound and rigorous, but its main purpose should be to serve the patients, families, and the community. Scientific evaluation of the Insite may have been a necessary condition, but it was certainly not a sufficient condition to bring about a permanent change in public policy or a sustainable supervised injection facility. Sometimes scientists and bureaucrats spend too much time worrying about protecting objectivity at the expense of advocacy. Yet, there are, time and time

8 Public Health

again, instances where advocacy needs to be undertaken and undertaken vigorously. The supervised injection facility was one of those times. It demanded public advocacy. Despite the victory, on the day of court announcement, I felt relief more than elation. I couldn't help but wonder what would have happened to Insite if things had been different? If we had lost the court case, would the various stakeholders have chosen the ethically sound course of action by continuing to operate in spite of an unjust law prohibiting the facility's operation? There would have been a strong ethical case for breaking the law and keeping Insite open. There was no scientific uncertainty about the effectiveness of Insite at engaging a hard to reach population in healthcare and saving lives. The only equipoise was political. Had we lost, would we have been forced to live through a medical, legal and ethical disaster while people died of preventable overdoses because it was against the law? These are dark questions that haunted Insite right up until the very moment that the final decision was rendered. Thankfully, we never had to publically confront what might have been had we lost. In my view, there are three important cultural ramifications of this case. Firstly, this legal decision says a lot about what it is to be a person, to have personhood, in Canada. Personhood symbolizes our connection to the wider human family.49 What it is to be a person exists in the borderland of human relations where personal agency and meaning are psychosocially constructed as part of an inner and outer conversation. Socially devalued features of our selves such as addiction reduce our opportunities to be on the threshold of a successful life.10 The personhood of people with addiction is often undermined or threatened by policies, programs, and implicit or explicit exclusion (e.g., drug courts and therapeutic communities are typically founded on the principle that addicts must abdicate a portion of their self-determination). The personhood of people with addictions in Canada has been emphasized, and their constitutiona_ rights feature prominently in this legal ruling. The second consequence of this case is that other jurisdictions may, and should, establish supervised injection if the epidemiological variables demand it. On this point, I disagree with the overemphasis on obtaining consensus from an overly broad collection of stakeholders (e.g., the municipal government, the local police) who are typically consulted in order to obtain their blessing so that lives can be saved by supervised injectio a. A letter of support from the Chief of Police or Mayor of a city would not be required to establish a cancer treatment program.

Correspondingly, one shouldn't be required in the case of harm reduction programs. If the evidence base is there to support an intervention, then we should move past consensus-building when it comes to lifesaving healthcare. The SCC has ruled that the morality of an activity, such as drug addiction, isn't enough to ignore someone's rights to security of their person under section 7 of Canada's Charter of Rights and Freedoms (the Charter): Additionally, the morality of the activity the law regulates is irrelevant at the initial stage of determining whether the law engages a s. 7 right. Finally, the issue of illegal drug use and addiction is a complex one which attracts a variety of social, political, scientific and moral reactions. While it is for the relevant governments to make criminal and health policy, when a policy is translated into law or state action, those laws and actions are subject to scrutiny under the Charter"2 (p. 9). The suitability of supervised injection shouldn't be debated any longer as though it were on par with a discussion on a sports show about which sports team will win the championship. Supervised injection is healthcare, and whether it is required in a jurisdiction needs to be determined by evidence and not arbitrary opinions or fickle political stances in search of votes. If science and medicine have established the best course of action, then we shouldn't turn to opinion polls to determine the best healthcare. Finally, it is my opinion that this ruling provides further affirmation that many healthcare providers know. harm reduction needs to be an explicit part of the standard of care now: Every single health authority and region in Canada should have a proactive policy detailing best practices in harm reduction when it is epidemiologically indicated. It is simply not acceptable to pretend harm reduction doesn't exist or to let moral opposition rather than evidence-based analysis guide decisions in this area. Any jurisdiction that doesn't have a positive policy on harm reduction is misguided. An example is provided by the City of Abbotsford in British Columbia. In June 2005, Abbotsford amended their zoning (bylaw no. 1378-2004) in order to overtly exclude harm reduction: Prohibited uses would include safe injection sites, needle exchanges, mobile dispensing vans, methadone treatment facilities and other types or similar uses.

Small: Canada's Highest Court Unchains Injection Drug Users

The experience of Insite should be an important lesson for jurisdictions that ignore, or in the case of Abbotsford, outlaw, harm reduction as part of healthcare. They do so at their own peril legally, medically and ethically. This commentary is not meant to be a distant scientific paper but, instead, an experience-based and socially positioned interpretation of events that I have lived. As one of the creators of Insite, I had always imagined that establishing an injection site would be the most difficult task to accomplish. In fact, it seems to me that the protection of it, once established, has been an even larger challenge. The struggle to protect Insite has distracted us from many other important prevention, treatment, enforcement, and harm reduction initiatives that need to be established to comprehensively address addiction. The

fact that the program has survived is itself an indicator of social change, and I believe that this ruling signals that we have reached an important cultural milestone. We've gone so much further than hoping that addiction will one day be understood as a healthcare matter. The very fact that Insite can exist, with the permanence of a Supreme Court decision, supports harm reduction as part of the standard of care, with the sustainability it deserves. With this ruling, we've moved beyond hopefulness to a point in our history where people with addictions have been unchained. Canada's highest court has spoken. It's the law. If you have built castles in the air, your work need not be lost; that is where they should be. Now put the foundations under them" (p. 255).

References 1. Marx K: Selected Works. In Theses on Feuerbach. 1845 12. edition. Edited by Marx K, Engels F Moscow: Progress 13. Publishers; 1989:13-15. 2. Supreme Court of Canada: Canada (Attorney General) v. PHS Community Services Society. In SCC 44 33556. Edited by Canada SC. Ottawa: Supreme Court of 14. Canada; 2011. 3. Small D: Mental illness, addiction and the supervised injection facility. Visions: BC's Mental Health and Addic- 15. tions Journal 2004, 2(1):37-39. 4. Harm Reduction Journal. Definition of Harm Reduction; 2012. [cited 2012 24 April]; available from: http://www. 16. harmreductionjournal.com. 5. Solomon R, Green M: The First Century. The history of 17. nonmedical opiate use and control policies in Canada, 1870-1970. University of Western Ontario Law Review 1982, 20(2):307-337. 6. Pitfield, Justice I: PHS Community Services Society v. At- 18. torney General of Canada: BCSC 661: The Supreme Court of British Columbia; 2008. 7. Alexander BK, Hadaway PF: Opiate addiction: the case for an adaptive orientation. Psychol Bull 1982, 92(2):367-381. 19. 8. Giddens A: Modernity and Self-Identity: Self and Society in the Late Modern Age. Stanford: Stanford University Press. Cambridge: Polity Press; 1997. First pub. 1991, 20. Cambridge: Polity Press. 9. Saris AJ: Telling stories: life histories, illness narratives and institutional landscapes. Cult Med Psychiatry 1995, 19(1):39-72. 10. Goffman E: Stigma: Notes on the Management of Spoiled Iden- 21. tity. New York: Simon and Schuster, Inc.; 1986 (1963). 11. Mora G: Vincenzo Chiarugi (1759-1820) and his psychiatric reform in Florence in the late 18th century. J Hist Med 1959, 14(October):424-433.

Gerard DL: Chiarugi and Pinel considered: soul's brain/ person's mind. J Hist Behav Sci 1997, 33(4):381-403. Schuster J-P, Hoertel N, Limsin F: The man behind Philippe Pinel: Jean-Baptiste Pussin (1746-1811). Br J Psychiatry 2011, 198:241. Schuster J-P, HoertelN , LimosinF :The man behind Philipe Pinel: Jean-Baptiste Pussin (1746-1811)-psychiatry in pictures. Br J Psychiatry 2011, 198:241. Weiner DB: Philippe Pinel "Memoir of Madness" of December 11, 1794: a fundamental text of modern psychiatry. Am J Psychiatry 1992, 149(6):725-732. Drucker E: A Plague of Prisons: The Epidemiology of Mass Incarceration in America. New York: The New Press; 2011. Small D, Drucker E : Policy makers ignoring science and scientists ignoring policy: the medical ethical challenges of heroin treatment. Harm Reduction Journal 2006, 3 (16): 1-14 . DeBeck K, Wood E, Zhang R, Tyndall M, Montaner J, Kerr T: Police and public health partnerships: evidence from the evaluation of Vancouver's supervised injection facility. Substance Abuse Treatment, Prevention and Policy 2008, 3(1):1-5. Fairbairn N, Small W, Shannon K, Wood E, Kerr T: Women's experiences in North America's first medically supervised safer injection facility. Soc Sci Med 67(8):817-823. Kerr T, Kimber J, DeBeck K, Wood E: The role of safer injection facilities in the response to HIV/AIDS among injection drug users. Curr HIV/AIDS Rep 2007, 4(4): 158-164 . Kerr T, Small W, Moore D, Wood E: A micro-environmental intervention to reduce harms associated with drug-related overdose: evidence from the evaluation of Vancouver's safer injection facility. Int J Drug Policy 2007, 18:37-45.

8 Public Health

22. Kerr T, Stoltz J, Tyndall M Li K, Zhang R, Montaner J, 36. Wood E: Impact of a medically supervised safer injection facility on community drug use patterns: a before and after study. Br MedJ 2006, 332:220-222. 23. Kerr T, Tyndall MW, Lai C, MontanerJSG, Wood E: Drug- 37. related overdoses within a medically supervised safer injection facility. Intl Drug Policy 2006, 17:436-441. 24. Kerr T, Tyndall MW, Lai C, Montaner JSG, Wood E: Circumstances of first injection among illicit drug users 38. accessing a medically supervised safer injecting facility. Am J Public Health 2007, 97(7):1228. 25. Kerr T, Tyndall MW, Li K, Montaner JS, Wood E: Safer 39. injecting facility use and syringe sharing among injection drug users. Lancet 2005, 366:316-318. 26. Kerr T, Wood E, Small D, Palepu A, Tyndall MW: Potential use of safer injecting facilities among injection drug 40. users in Vancouver's Downtown Eastside. Can Med Assoc J 2003, 169(8):1-5. 27. McKnight I, Maas B, Wood E, Tyndall MW, Small W, Lai C, et al: Factors associated with public injecting among 41. users of Vancouver's supervised injection facility. Am J Drug Alcohol Abuse 2007, 33(2):319-325. 28. Milloy MJ, Wood E, Small W, Tyndall M, Lai C, Montaner J, Kerr T: Incarceration experiences in a cohort 42. of active injection drug users. Drug Alcohol Rev 2008, 27(6):1-7. 29. Petrar 5, Kerr T, Tyndall MW, Zhang R, McKnight B, Montaner JSG, Wood E: Injection drug users' perceptions regarding use of a medically supervised safer in- 43. jetting facility. Addict Behav 2007, 32:1088-1093. 30. Stoltz JA, Wood E, Small W Li K, Tyndall M, Montaner J, Kerr T: Changes in injecting practices associated with the use of a medically supervised injection facility. J Pub 44. Health (Oxford) 2007, 29(1):35-39. 31. Tyndall MW, Kerr T, Zhang R, King E, Montaner JG, Wood E: Attendance, drug use patterns, and referrals made from North America's first supervised injection facility. Drug Alcohol Depend 2005, 83(3):193-198. 45. 32. Tyndall MW, Wood E, Zhang R, Lai C, Montaner JS, Kerr T: HIV sero-prevalence among participants at a supervised injection facility in Vancouver, Canada: implications for prevention, care and treatment. Harm Reduction 46. Journal 2006, 3(36):1-5. 33. Wood E, Kerr T, Buchner C, Marsh D, Montaner JS, Tyndall MW: Methodology for evaluating Insite: Canada's first medically supervised safer injection facility 47. for injection drug users. Harm Reduction Journal 2004, 1(9):1-5. 34. Wood E, Kerr T, Montaner JS, Strathdee 5, Kerr T, Wodak A, Spittal P, Hankins C, Schechter MT, Tyndall M: Rationale for evaluating North America's first medic- 48. ally supervised injecting facility. Lancet Infect Dis 2004, 4(5):301-306. 49. 35. Wood E, Kerr T, Small W, Li K, Marsh D, Montaner JS, Tyndall MW: Changes in public order after the opening of a medically supervised safer injection facility for injec- 50. tion drug users. Can Med AssocJ 2004, 171(7):731-734.

Wood E, Kerr T, Stoltz J, Qui Z, Zhang R, Montaner JSG, Tyndall MW: Prevalence and correlates of hepatitis C among users of North America's first medically supervised safer injection facility. Public Health 2005, 119(12):1111-1115. Wood E, Kerr T, Tyndall MW, MontanerJSG: The Canadian government's treatment of scientific process and evidence: inside the evaluation of North AmericA first supervised injecting facility. Int J Drug Policy 2007, 19(3):220-225. Wood E, Montaner JS, Kerr T: Reflection and reaction: illicit drug addiction, infection disease spread, and the need for an evidence-based response. Lancet 2008, 8:142-143. Wood E, Tyndall MW, Lai C, Montaner JSG, Kerr T: Impact of a medically supervised safer injecting facility on drug dealing and other drug-related crime. Subst Abuse Treat Prey Policy 2006, 1(13):1-4. Wood E, Tyndall MW, Li K, Lloyd-Smith E, Small W, Montaner JSG, Kerr T Do supervised injecting facilities attract higher-risk injection drug users? Am J Prey Med 2005, 29(2):126-130. Wood E, Tyndall MW, Montaner JS, Kerr T: Summary of findings from the evaluation of a pilot medically supervised injecting facility. Can Med Assoc J 2006, 175(11):1399-1404. Wood E, Tyndall MW, Qui Z, Zhang R, Montaner JS, Kerr T: Service uptake and characteristics of injection drug users utilizing North America's first medically supervised safer injection facility. Am J Public Health 2006, 96(5):770-773. Wood E, Tyndall MW, Stoltz J, Small W, Lloyd-Smith E, Zhang R, Montaner JSG, Kerr T: Factors associated with syringe sharing among users of a medically supervised injecting facility. Am J Infect Dis 2005, 1(1):50-54. Wood E, Tyndall MW, Stoltz J, Small W, Zhang R, O'Connell J, Montaner JSG, Kerr T: Safer injecting education for HIV prevention within a medically supervised safer injecting facility. Int J Drug Policy 2005, 16:281-284. Wood E, Tyndall MW, Zhang R, Stoltz J, Lai C, Montaner JSG, Kerr T: Attendance at supervised injecting facilities and use of detoxification services. N Eng J Med 2006, 354(23):512-514. Wood E, Tyndall MW, Zhang R, Montaner JS, Kerr T: Rate of detoxification service use and its impact among a cohort of supervised injection facility users. Addiction 2007, 102:916-919. Wood RA, Wood E, Lai C, Tyndall MW, Montaner JSG, Kerr T: Nurse-delivered safer injection education among a cohort of injection drug users: evidence from the evaluation of Vancouver's supervised injection facility. Int J Drug Policy 2008, 19(3):183-188. Latour B: From the world of science to the world of research. Science 1998, 280(5361):208-209. Small D: Looking into the cultural mirror: addiction, secret lives and lost personhood. Visions 2005, 2(5):29-30. Thoreau HD: Walden. Vermont: Orion Publishing Group; 1995 (1910).

Christie/Groarke/Sweet Virtue Ethics as an Alternative to Deontological and Consequential Reasoning

Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate Timothy Christie, Louis Groarke, and William Sweet

Introduction

The purpose of this paper is to extract and analyze ethical theories that bear on two approaches to dealing with problematic substance use, i.e., abstinence-based approaches and harm reduction approaches. The primary issue to be investigated is how harm reduction— because it does not require individuals to either reduce their consumption of illicit substances or to abstain from illicit substance use---can be ethically justified. The harm reduction approach claims that many of the negative consequences associated with problematic substance use are avoidable through specific interventions, and these interventions can be effective regardless of whether substance use persists. The abstinence-based approach, however, claims that it is important to get individuals off drugs or at least to decrease consumption. This latter approach generally maintains that, since harm reduction interventions tolerate continued substance use, they send the wrong message. We employ three different ethical models to analyze the values conflict between the harm reduction and the abstinence-based approaches. Our methodology in this paper is different than that usually employed in addressing ethical issues in the applied ethics field (Beauchamp & Childress, 2001). Typically in applied ethics, the principles of autonomy, beneficence, nonmaleficence, and justice are first applied to specific situations. Then the analyst will use his or her moral intuitions to determine which principle or principles are most important given the particular facts of the situation. We approach the debate in a different way because depending on moral intuitions is highly problematic. Specifically, we will focus on the philosophical theories from which the above-mentioned principles have been abstracted. For example, the principle of autonomy is abstracted from Deontological ethics, beneficence and nonmaleficence from Utilitarianism, and the principle of justice was originally articulated by Aristotle who is a key source of contemporary Virtue Ethics. The advantages of our approach are, first, it directly addresses what are clearly

the three most influential models in contemporary ethical thought and, second, it is more robust than simply referring to abstract principles and then using intuitions to determine which principle is most important. Harm Reduction

To discuss harm reduction it is necessary to identify (1) its fundamental assumptions and definition, (2) specific kinds of harm reduction interventions, and (3) the implications of the political advocacy of many harm reduction proponents. The fundamental assumption of harm reduction, which is the primary focus of this paper, is that it is important to try to reduce drug-related harm (Erickson, Butters, & Walko, 2007). The harm reduction approach does not require individuals to reduce or abstain from drug use; rather, it attempts to mitigate the negative consequences of drug use. Abstinence could be an eventual outcome and is consistent with harm reduction; however, abstinence is not a condition of the harm reduction approach. There are numerous kinds of harm reduction interventions, and they apply in different ways depending on the particular substances and activities involved: illicit opiates, cocaine, ecstasy, alcohol, tobacco, etc. The specific interventions discussed in this paper are needle exchange programs and supervised injection sites. Not all harm reduction interventions have the same ethical justification. For example, needle exchange programs and supervised injection sites assist individuals who are already engaged in an activity. If an individual is going to inject drugs regardless of the harms involved, needle exchange gives that person clean needles, and a supervised injection site provides a medical setting in which the person can inject the drugs, but in neither case does the intervention involve supplying the substance. The ethical justification for other harm reduction interventions would be different. For instance, heroin prescription programs actually provide the substance to individuals, which is different than simply providing sterile equipment or supervising injections. Noting this difference is not to make the judgment that there may be something ethically problematic with heroin prescription; it simply recognizes that the situation is different and, therefore, the ethical justification may be different. For the sake of simplicity, we have chosen to focus on the generic philosophical aspects of harm reduction, with specific references to needle exchange programs and supervised injection sites, when necessary.

Source: From International Journal of Drug Policy 19, 1 (2008), 52-8. Reprinted with permission.

II

JII

III I.k.1

8 Public Health

Finally, as with any important development in public policy, there are elements of political advocacy within the harm reduction movement. Some advocates argue for complete legalization of illicit drugs and radical drug law reform (Hankins, 2000), while others argue for a medical model that is more cautious (Anderson, 2000). We address the advocacy issue peripherally in the sense that we think harm reduction makes ethical sense, is superior to alternative policy approaches to problematic substance use, and can surmount the standard ethical objections to it. Criticisms of Harm Reduction

The most important ethical concern with harm reduction is related to the "value-judgment" that it is more important to reduce the harms associated with drug use than it is to reduce or prDhibit drug use. The controversial character of this value judgment is amplified because the relevant drugs are illegal. Critics of harm reduction have argued that (:.) it encourages drug use, (2) it sends a mixed message, and (3) it fails to get people off drugs. There is a significant amount of the literature, however, supporting the claim that harm reduction interventions do not increase or encourage drug use. In fact, the evidence demonstrates the opposite. The experience of the Canadian supervised injection site shows that some patients, who would not have otherwise sought treatment, evenually seek treatment, including abstinence-based programs, as a result of using the site (Tyndall et al., 2005). This is consistent with what has been observed in needle exchange programs as well; participants frequently seek referrals for treatment (Strathdee, Celentano, & Shah, 1999; Hagan, McGough, & Thiede, 2000). Based on what is currently known, the claim that harm reduction programs encourage drug use is unsubstantiated. The second criticism of harm reduction is that it "sends a mixed signal." This objection was made in a 1996 letter to the Governor's Advisory Council on AIDS (Whitman, 1996). In this letter, Christine Whitman, the former Governor of New Jersey, acknowledges that the National Academy of Science and the Centers for Disease Control and Prevention endorse needle exchange programs as effective interventions for interrupting the spread of HIV. She argues, however, that needle exchange programs "send a mixed signal" and that governments should not "be in the business of facilitating illegal activity" These sentiments are echoed in the United States' National Drug Control Strategy, which states that (1) there should be no tolerance for substance use whatsoever (ONDP, 2007a), (2) it is

essential to deter substance use (ONDP, 2007b) and (3) it is imperative to disrupt illegal drug markets (ONDP, 2007c). The impetus for these objections is the belief that any policy that tolerates drug use sends a "mixed signal" and frustrates primary prevention efforts. It seems, however, that these arguments are simply a restatement of the first objection, namely, that harm reduction will encourage illegal drug use. For example, if the outcome of encouraging illegal drug use was not relevant to this argument, it is difficult to understand what "sending a mixed signal" would mean or why it would even be a relevant consideration. The significance of this objection is mitigated by the same evidence that invalidates the first objection. A third criticism of harm reduction is found in the following quotation from the (current) Canadian Federal Minister of Health, Tony Clement: Do safe injection sites contribute to lowering drug use and fighting addiction? Right now the only thing the research to date has proven conclusively is [that] drug addicts need more help to get off drugs. Given the need for more facts, I am unable to approve the current request to extend the Vancouver site for another three and a half years (Health Canada, 2006). Scientific evaluation of the Canadian supervised injection site revealed that the facility attracted drug users who were formerly most likely to inject in public and be at highest risk of HIV infection and death due to overdose (Wood, Kerr, & Montaner, 2004a; Wood, Kerr, & Small, 2005a; Wood, Tyndall, & Li, 2005b). The opening of the supervised injection site was followed by statistically significant reductions in public drug use and publicly discarded syringes (Wood et al., 2004b); was associated with reduced syringe sharing and other forms of HIV risk behaviour (Kerr, Tyndall, Li, Montaner, &Wood, 2005); and was a key referral mechanism to addiction treatment and other community resources, including abstinence-based programs (Tyndall et al., 2005). It was not associated with increased drug use or other drug-related problems (Kerr et al., 2006). The Minister's reasoning, however, suggests that it is more important to "contribute to lowering drug use [,] fighting addiction," and getting people "off drugs" than it is to achieve these outcomes. This belief is so strongly held that the Minister refused to renew the Section 56 exemption to the Controlled Drugs and Substances Act given to the supervised injection site, which will make it illegal for the site to operate after this exemption expires in June 2008.

Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning

This is a substantive objection to the theoretical underpinnings of harm reduction and requires a careful ethical analysis. If a program will get people off drugs, then the Minister would seem to approve of it; if it does not get people off of drugs, the Minister is definitely against it, even if the program otherwise avoids many of the negative consequences of illicit drug use. Embedded in this objection is the valuesbased preference for an abstinence-based approach to drug policy. The abstinence-based approach differs from harm reduction in a very important respect; it requires the individual either to abstain from drugs or attempt to abstain from drugs (Christie & Anderson, 2003; Tammi & Hurme, 2007). The harm reduction approach, however, does not require the individual to either abstain from or reduce consumption of illicit drugs. It simply seeks to prevent the negative consequences of illicit substance use. Ethical Analysis

In his recent article entitled "Ethics and Drug Policy," Alex Wodak argues that the scientific debate about harm reduction is over (Wodak, 2007a). Harm reduction works. The real conflict, which has largely been overlooked, is about ethics. More specifically, it is a revival of the debate between Utilitarian and Deontological approaches to ethics. Wodak claims that, implicit in the reasoning of harm reduction advocates, is a Utilitarian argument, which holds that the key guideline in ethics is that if negative consequences can be avoided, they should be avoided. Abstinence advocates, however, generally seem to employ a Deontological ethic, which maintains that the moral worth of one's actions has nothing to do with the consequences of those actions but, rather, is determined by the intention of the actors. Our analysis will examine this debate in more depth. We will present and interpret the ethical theories of the Consequentialist Utilitarian, John Stuart Mill, of the founder of Deontological ethics, Immanuel Kant, and Aristotle's Virtue Ethics, which many see as a remedy to the Consequentialist— Deontological dilemma. Utilitarianism (John Stuart Mill)

John Stuart Mill was a 19th century British philosopher who was seminal in the development of liberalism (Arnold, 2006) and is still recognized as one of the key proponents of Utilitarianism (Mill, 1863). His theory of liberty holds that individuals in society should have broad-ranging freedoms, such as liberty of thought, discussion, association, and the right to

pursue their own life goals as they see fit. The only justifiable limit to an individual's liberty occurs when that person's actions will harm others. This is called the "harm principle." (Mill did not consider merely giving of fence or causing inconvenience as harm. He thought if harm were defined to include these ways, individuals could be prevented from criticizing social conventions.) His contention was that by restraining the state's ability to interfere with the liberty rights of individuals, the long-term consequences for society will be positive. Many liberal theorists root their theories in some conception of "natural" or "human rights." They maintain that rights, such as the rights to life, liberty, and property, are natural, inalienable, and intrinsically valuable. However, Mill thinks that individual liberty rights are instrumental, i.e., good for the sake of achieving something else. This idea is rooted in his commitment to Utilitarianism. Utilitarianism is the theory which contends that the foundation of morality is the greatest happiness principle and holds that an action is right if it tends to promote the greatest happiness for the greatest number and wrong if it tends to produce the opposite of happiness. Happiness is defined as pleasure and the absence of pain, and unhappiness is defined as pain and the privation of pleasure. Mill contends that pleasure and freedom from pain are the only things that are intrinsically valuable and that everything else is instrumental to them. When determining what will produce happiness, Mill is not focusing just on the agent's own happiness but on that of the entire community. In determining this, each agent has one "vote." For the Utilitarian, the consequences for the community are of fundamental importance, and when calculating the consequences we must consider all relevant variables, including the difference between short-term and long-term consequences. For example, an action might produce much happiness in the short-term but, in the long-term, result in more harm than good; or the opposite might be true. With regard to harm, Mill's "liberal" view clearly makes allowances for individuals to harm themselves (e.g., to consume alcohol to excess). But if drunkenness leads an individual to neglect his or her responsibilities or to interfere with others, Mill would allow the state to intervene. The ethical issue is not the drinking or the harm to oneself but the individual fulfilling his or her responsibilities—although drinking is the cause of the failure to carry out one's responsibilities. If a person could drink and not harm society, Mill would certainly not allow the state to interfere with that person's liberty.

AI

II

:Al

1111:1.1

8 Public Health

This bears on the cun•ent discussion about drug use. Consider, for example, the harms of injectionrelated infections, overdoses, blood-borne disease, violence, property crime, involvement in the sex trade, and incarceration (Kerr & Palepu, 2001; Hunt, 2005) The cost for every untreated opiate user is estimated to be over $45,000 (Cdn) per year (Wall, Rehm, & Fischer, 2000), the lifetime cost of treating a person with HIV exceeds $250,000 (Cdn) (Kuyper, Hogg, Montnaer, Schecter, & Wood, 2004), a course of Hepatitis-C treatment can range from $10,000 to $30,000 (Cdn) per patient (Public Health Agency of Canada, 2003), and Emergency Department utilization is significantly higher among those with problematic substance use than those without (Palepu, Tyndall, & Leon, 2001; McGeary & French, 2000). Mills liberal view, then, would allow the state to interfere with a person's libe7ty to use such substances when that use leads to harmful consequences for society. However, his liberalism does not answer the question of what is the most appropriate state response: abstinencebased policies or harm reduct on policies? The answer to this question requires a Utilitarian analysis. Utilitarianism requires us to consider the current policy environment, the scientific evidence for each approach, and whether it is effective or not. This will give us guidance as to which approach best promotes the general happiness. As discussed above, the policy environment—at least in the United States and Canada—is predominately abstinence-based; it is also influenced by the criminal justice system in which prohibition is dominant. For example, in Canada, 73% of federal spending in the National Drug Strategy is spent on enforcement measures such as border control, police investigations, and federal prosecution expenses, whereas less than 3% is spent on harm reduction (deBeck, Wood, Montaner, & Kerr, 2006). This imbalance occurs despite the fact that there is significant evidence indicating that abstinence-based policies have had many (albeit unintended) negative consequences (Hankins, 2000; Oscapella, 1996). Moreover, these negative outcomes are largely preventable through harm reduction interventions, such as needle exchange programs and supervised injection sites (Wodak, 2007b; Stoltz, Wood, & Small, 2007). Since abstinence-based policies have many unintended negative consequences, Mill would conclude that they would likely not promote the greatest happiness for the greatest number. Moreover, he would note that the harm reduction approach will avoid many negative consequences. Thus, a Utilitarian would likely arrive at the conclusion that ethics demands a policy of harm reduction and not simply abstinence.

Deontological Ethics (Kant, 1785)

Immanuel Kant was an 18th century German philosopher who considered ethics an essential component of human life. He contrasted ethics with physics, explaining that physics describes the laws of nature according to which every material thing "acts" and ethics describes the laws of morality according to which every person "ought to act." He lamented the fact that frequently what "ought to happen" does not happen because, unlike purely material objects, human agents are free to choose how to act—i.e., whether to follow the ethical law Since human beings are rational creatures, Kant thought the laws of ethics could be derived from "reason" alone without having to appeal to specific circumstances or experience. On this theory, whether an action is right or wrong depends on the intrinsic nature of the action rather than on the situation or specific circumstances in which one finds him or herself. In other words, for Kant, ethics is primarily concerned with doing the right thing because it is the right thing to do, not because it is in the agent's self-interest, or because it will produce good consequences, or for any other "instrumental" reason. According to Kant, only actions motivated by a sense of duty have moral worth, and duty consists of acting out of respect for the moral law (or, as Kant calls it, the Categorical Imperative). The Categorical Imperative can be stated as follows: first, "act only on that maxim whereby thou canst at the same time will that it should become a universal law," or, secondly, "so act as to treat humanity, whether in thine own person or in that of any other, in every case as an end withal, never as a means only" For Kant, the Categorical Imperative is a test useful for determining duty. He formulated it in different ways but thought that each formulation stated essentially the same insight and thus would always lead to the same conclusion. The first test is whether the maxim—the principle of action—could be universalized without contradiction. Kant gives the example of lying. If an individual needs to borrow money to get out of a difficult situation and is aware that he or she will not be able to repay the money but is also aware that no one will lend him or her money unless a promise of repayment is made, should that individual lie? Suppose that individual were to say "yes." Kant would say that the universalized form of the principle of this action—i.e., "Making a false promise (i.e., lying) is a universal law"—would make the concept of a promise incoherent; there would be no sense in promising. Others could never know what a speaker was saying or committing him or herself to when he or she uttered the words, "I promise . . ."

Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning

Likewise, on the second formulation, lying would be using another person as a means to achieve one's own ends and thus not respecting that person as an end in him or herself. According to Kant's deontology consequences are irrelevant when considering the moral quality of an action. But the entire debate between harm reduction and abstinence is one that focuses on consequences. The major criticism of the abstinence-based approach is that it has greater negative consequences than does the harm reduction approach. Similarly, the major justification for harm reduction is that it works (i.e., has fewer negative consequences). However, on a Kantian model, an appeal to consequences is irrelevant when promoting or criticizing an approach; rather, one is restricted to focusing on the intention of the agent. An abstinence-based approach would almost certainly "pass" the Kantian test. An application of Kant's Categorical Imperative would be: can we universalize the maxim that everyone everywhere should refrain from engaging in illicit injection drug use? Clearly, we could. A harm reduction approach, however, could not be justified on a Kantian model because its "raison d'etre" is ultimately to avoid negative consequences. For example, needle exchange programs and supervised injection sites have value, not because illicit injection drug use is good but rather to avoid more serious negative consequences. Kant would argue that this type of instrumental reasoning does not meet the relevant ethical standard. For an action to pass the test of the Categorical Imperative it must be able to be rationally willed by everyone and be able to be acted upon by everyone, that is, it would have to be seen to be good in itself without appealing to consequences. The essence of harm reduction is to appeal to good consequences as a justification for specific actions, e.g., needle exchange and supervising injections. It is important to note, however, that this does not mean that Kant would find harm reduction immoral but only that it is not moral. On Kant's view, consequences are irrelevant to the morality of actions and so reflect neither virtues nor vices. Virtue Ethics

Virtue Ethics is most famously associated with Aristotle. It has enjoyed resurgence in contemporary moral philosophy as an alternative to the narrowness of the Deontological—Consequentialist dichotomy. Virtue Ethics does not focus on isolated acts but on the character of the agent—e.g., honesty, loyalty, courage, compassion, kindness, fairness, etc. Aristotle does not separate morality from politics. The point of politics is

to have a good society populated with citizens of good moral character. In the Nicomachean Ethics, Aristotle famously offers the mean as a heuristic device to determine whether a specific character trait is a virtue (Aristotle, 350 Bc). Simply put, we can have too much or too little of a character trait understood as a mid-point between two opposites. Consider, for example, the virtue of courage understood as the correct balance of fear to fearlessness. People who suffer from cowardice have too much fear and not enough fearlessness; those who suffer from rashness have too much fearlessness and not enough fear. Or consider the virtue of honesty understood as the correct balance between hiding and revealing the truth. The liar, the person who is dishonest, hides the truth too much; the blabbermouth, the person who lacks propriety, reveals the truth too much. Moreover, Aristotle insists that morality depends upon a number of contextual factors. We need to do the right thing, to the right people, at the right time, in the right way, for the right reasons. Thus, a Virtue Ethics model takes account of context and consequences without reducing ethics to simple matters of promoting pleasure, avoiding pain, or doing one's duty. In the context of the present discussion on harm reduction and problematic substance use, we need to ask what virtue would require of a policy-maker (and by implication, of a society): implementing harm reduction policies or implementing abstinence-based policies. The positive character trait driving those involved in harm reduction programs might be called "compassion." Compassion, defined in terms of the correct ratio between removing and not removing suffering or pain or misery from others, is clearly a virtue. When, however, it comes to helping those engaged in illegal or unhealthy and destructive behaviour, the issue becomes more complicated. What does it mean to be too compassionate or not to be compassionate enough? In finding the mean, we can better gauge, from the perspective of Virtue Ethics, the morality of harm reduction and abstinence-based programs. So how compassionate should a society be? Define the virtue of compassion as having the right measure of sensitivity or solicitude for suffering. Compassion is then a mean between two extremes: being "too hard" is not having enough sensitivity, and being "too soft" is having too much sensitivity. Thus abstinence advocates would argue that harm reduction is "too soft" and harm reduction advocates would accuse abstinence-based policies of being "too hard." Aristotle would caution that people with different roles in society should aspire to different degrees of compassion. A military drill sergeant who has to toughen up his charges to face hardship in battle

II

II I Nal

8 Public Health

should clearly be, so to speak, more hard than soft. The same could be said of a coach preparing athletes for grueling competition. One could even argue that parents should practice a certain degree of "hardness," that overly permissive parenting spoils the child. But these are cases where not relieving suffering (i.e., hardness) has positive results. In the case of harm reduction and abstinence-based programs, the evidence suggests that hardness (the abstinence-based approach) does not seem to have any positive effe:ts and might have significant (unintended) negative consequences. Doing away with harm reduction programs, like supervised injection sites and needle exchange programs, means that more individuals die, become sick, contract HIV, are involved in criminal activity, etc. As a result, one could argue that the rationale for "hardness" is eliminated. There is, however, an equally important second issue. It is virtuous to promote virtue; it is vicious to promote vice. By implementing harm reduction programs, are we helping people living with problematic substance use become virtmus persons, or are we aiding and abetting vicious behaviour? From the needle exchange literature and the literature on supervised injection sites, it is clear that harm reduction programs do not plausibly lead to an increase in drug use and, in at least some cases, lead to rehabilitation. Therefore, one could argue that compassion is not aiding and abetting and is the appropriate moral response. If harm reduction programs do not promote substance use but merely tolerate it—there is a difference between these features—and if, instead of making people ill, they relieve the suffering of the ill, they are morally valuable. Conclusion

This paper does not attempt to provide a critical review of the scientific evidence bearing on harm reduction and abstinence-based programs. Instead, we accept the view that there is extensive evidence that the harm reduction approach is effective and that current policy

approaches, which are largely abstinence-based, are ineffective. This does not, however, settle the issue of whether and how the harm reduction approach is ethically justifiable; this is the concern of the present paper. Consequently, we conducted an ethical analysis using three dominant ethical theories. While there are other important theoretical approaches, such as the "principle" approach of applied ethics, the feminist approach, pragmatism, and ethical theories from other cultures, the three that we discuss are clearly the most influential today and are the source of many of the principles referred to in contemporary applied ethics. In conclusion, the major ethical issue we have investigated is how the harm reduction tenet—that individuals need not be required to abstain or at least to attempt to abstain from substance use—might be ethically justified. This value judgment is clearly justified on Utilitarian grounds because, based on the evidence, harm reduction policies would produce the greatest good for the greatest number. However, Kant would not think that the values guiding harm reduction meet the appropriate ethical standard, because the "raison d'etre" of harm reduction is exclusively to prevent negative consequences. While he would likely embrace an abstinence-based approach, consequences would be irrelevant to this judgment. Virtue Ethics takes account of the social consequences but also the moral character of the agent. The virtue of compassion would provide a strong ethical foundation for at least some harm reduction policies without requiring recourse to Utilitarianism. Acknowledgments

We would like to thank Dr. Robert Larmer, Professor of Philosophy at the University of New Brunswick, for his comments on earlier versions of this paper. We would also like to thank the peer-reviewers of this manuscript and the editors at the International Journal of Drug Policy. Their critical comments and careful review have improved this paper significantly.

References

Anderson, J. (2000). Interpreting the relation between injection drug use and harm: A cautionary note. CMAJ, 162(12), 1695-1696. Aristotle. (350 BC). Nicomachean ethics (selections). In S. M. Cahn, & P Markie (Eds.), Ethics: History, theory and contemporary issues (3rd ed., pp. 124-178). Oxford: Oxford University Press. Arnold, M. (2006). Mill's On Liberty. Ezine articles. Retrieved 4 September, 2007, from http://ezinearticles.com/?MillsOn-Liberty&id= 181525.

Beauchamp, T. L., & Childress, J. F (2001). Principles of biomedical ethics (5th ed.). New York: Oxford University Press. Carpenter, C. (2007). Workplace drug testing and worker drug use. Health Services Research, 42(2), 795-810. Christie, T., & Anderson, J. (2003). Drug treatment courts are popular but do they work and are they ethical and appropriate for Canada? Health Law in Canada, 23(4), 70-78. deBeck, K., Wood, E., Montaner, J., & Kerr, T. (2006). Canada's 2003 renewed drug strategy—an evidence-based review. HIV AIDS Policy Law Review, 11(2/3), 1,5-12.

Christie/Groarke/Sweet: Virtue Ethics as an Alternative to Deontological and Consequential Reasoning 457

Erickson, P., Butters, J., Walko, K., Butterill, D., Reggie, C., Fischer, B., et al. (2007). CAMH and harm reduction: A

background paper on its meaning and application for substance use issues. Centre for Addiction and Mental Health. Retrieved 19 November 2007 from http://www.camh.net/ Public policy/Public policy papers/harmreductionbackground.html. Hagan, H., McGough, J. P, Thiede, H., et al. (2000). Reduced injection frequency and increased entry and retention in drug treatment associated with needle-exchange participation in Seattle drug injectors. Journal of Substance Abuse Treatment, 19,247-252. Hankins, C. (2000). Substance use: Time for drug law reform. CMAJ, 162(12), 1693-1694. Health Canada, (2006). No new injection sites for addicts until questions answered says Minister Clement. News Release 2006-85 1 September 2006. Retrieved 19 November 2007 from http://www.hc-sc.gc.ca/ahcasc/media/nrcp/2006/2006 85e.html. Hunt, N. (2005). A review of the evidence-base for harm reduction. Forward Thinking on Drugs, 55. Kant, I. (1785). Groundwork of the metaphysic of morals. In S. M. Cahn & P Markie (Eds.), Ethics: History, theory and contemporary issues (3rd ed., pp. 270-308). Oxford: Oxford University Press. Kerr, T., & Palepu, A. (2001). Safe injection facilities in Canada: Is it time? CMAJ, 165(4), 436-437. Kerr, T., Stoltz, J. A., Tyndall, M., Li, K., Zhang, R., Montaner, J., et al. (2006). Impact of a medically supervised safer injection facility on community drug use patterns: A before and after study. 58 BMJ, 332(7535), 220-222. Kerr, T., Tyndall, M., Li, K., Montaner, J., & Wood, E. (2005). Safer injection facility use and syringe sharing in injection drug users. Lancet, 366(9482), 316-318. Kuyper, L. M., Hogg, R. S., Montnaer, J. S., Schecter, M. T., & Wood, E (2004). The cost of inaction on HIv transmission among injection drug users and the potential for effective interventions. Journal of Urban Health, 81(4), 655-660. McGeary, K. A., & French, M. T. (2000). Illicit drug use and emergency room utilization. Health Service Research, 35(1 Pt 0,153-169. Mill, J. S. (1863). Utilitarianism. In S. M. Cahn & P. Markie (Eds.), Ethics: History, theory and contemporary issues (3rd ed., pp. 317-351). Oxford: Oxford University Press. ONDP (Office of National Drug Control Policy United States Government). (2007a). Disrupting drug markets. National drug control strategy 2006. Retrieved 19 November 2007 from http://www.whitehousedrugpolicy.gov. ONDP (Office of National Drug Control Policy United States Government). (2007b). Healing America's drug users. National drug control strategy 2006. Retrieved 19 November 2007 from http://www.whitehousedrug policy.gov. ONDP (Office of National Drug Control Policy United States Government). (2007c). Stopping drug use before it starts: Education and community action. National drug control

strategy 2006. Retrieved 19 November 2007 from http:// www.whitehousedrugpoficy.goy Oscapella, E. (1996). Illegal drugs policy, AIDS and hepatitis: From prohibition to harm reduction. Med. Law, 15, 391-398. Palepu, A., Tyndall, M. W. , Leon, H., Muller, J., O'Shaughnessy, M. V, Schecter, M. T., et al. (2001). Hospital utilization and costs in a cohort of injection drug users. CMAJ, 164(4), 415-420. Public Health Agency of Canada, (2003). Hepatitis C prevention, support and research program. Retrieved 19 November 2007 from http://www. phac-aspc.gc.ca/hepc/pubs/ psrpmideval-ppsrevalinter/i_problem_e. html. Stoltz, J., Wood, E., Small, W, Li, K., Tyndall, M., Montaner, J., et al. (2007). Changes in injecting practices associated with the use of a medically supervised safer injection facility. Journal of Public Health (Oxford), 29(1), 35-39. Strathdee, S. A., Celentano, D. D., Shah, N., Lyles, C., Strambolis, V. A., Macalino, G., et al. (1999). Needleexchange attendance and health care utilization promote entry into detoxification. Journal of Urban Health, 76(4), 448-460. Tammi, T., & Hurme, T. (2007). How the harm reduction movement contrasts itself against punitive prohibition. International Journal of Drug Policy, 18, 84-87. Tyndall, M. W , Kerr, T., Zhang, R., King, E., Montaner, J. G., & Wood, E. (2005). Attendance, drug use patterns, and referrals made from North America's first supervised injection facility. Drug Alcohol Dependence, 6(238), 51-54. Wall, R., Rehm, J., Fischer, B., Brands, B., Fliksman, L., Stewart, J., et al. (2000). Social costs of untreated opioid dependence. Journal of Urban Health, 77(4), 688-722. Whitman, C. (1996). Letter to Mr. David Troast, Chairperson of Governor's Advisory Council on AIDS. Trenton, NJ. Wodak, A. (2007a). Ethics and drug policy. Psychiatry, 6(2), 59-62. Wodak, A. (2007b). Health exchange and prevention of lily: The evidence for effectiveness is beyond dispute. Addiction, 102,161-163. Wood, E., Kerr, T., Montaner, J. S., Strathdes, S. A., Wodak, A., Hankins, C. A., et al. (2004). Rationale for evaluating North America's first medically supervised safer-injecting facility. Lancet Infectious Disease, 4(5), 301-306. Wood, E., Kerr, T., Small, W, et al. (2005). Changes in public order after the opening of a medically supervised safer injecting facility for illicit injection drug users. CMAJ, 2, 126-130. Wood, E., Kerr, T., Small, W, Li, K., Marsh, D. C., Montaner, J. S., et al. (2004). Changes in public order after the opening of a medically supervised safer injecting facility for illicit injection drug users. CMAJ, 171(7), 731-734. Wood, E., Tyndall, M. W, Li, K., Lloyd-Smith, E., Small, W, Montaner, J. S., et al. (2005). Do supervised injecting facilities attract higher risk injection drug users? American Journal of Preventive Medicine, 2,126-130.

al

458

I I dal

MLLE

8 Public Health

8.5 Inequality in Health

When Are Health Inequalities Unjust?—The Social Determinants of Health Norman Daniels

When is an inequality in health status between different socioeconomic groups unjust? More generally, when is an inequality in health status between different demographic groups unjust? F. . .1 We have known for over 150 years that an individual's chances of life and death are patterned by social class: The more affluent and better educated people are, the longer and healthier their lives (Villerme 1840, cited in Lynch et al. 1998). When these inequalities are present in poor, developing countries, it is tempting to think that poverty and deprivation adequately explain them. We should avoid the temptation. Some rather poor countries pursue policies that produce excellent aggregate health outcomes while moderating significantly the degree of health inequality. In addition, these effects of socioeconomic status (SES) are present as well in rich, developed countries, where the sources of inequality are not deprivation and deep poverty. More striking, these patterns persist when there is universal access to health care—a fact quite surprising to those who think that financial access to medical services is the primary determinant of health status. In addition to socioeconomic inequalities' that are correlated with health inequalities, we have extensive evidence about patterns of racial and ethnic disparities in many countries. Many of these inequalities are due to the way in which social exclusion and patterns of discrimination push ethnic mino::ities (or majorities, as in some former colonies, like South Africa) into the lowest socioeconomic groups. Still, there is often a significant residual effect of race, as in the United States, even when correlations are controlled for income, education, and insurance levels. There are also striking patterns of gender inequality in health in some countries—leading Sen (1990) to ask where the millions of "missing" women are in some Asian countries, as compared to gender ratios in developed countries. F. . .1 Of course, we cannot infer causation from correlation, and correlations alone will not support judgments

about injustice. To know whether one inequality is a cause or determinant of another and then to ascribe some judgment about inequity to it, we have to know something about the mechanisms at work. We need to know, for example, that although poor health or disability can lead to lower income, which can be a significant consequence in many settings, such "health selection," as it is called, explains only part of the correlation between low income and poor health in general,' and we have good reason to suppose the existence of mechanisms that work in the other direction, from socioeconomic inequality to health differences. Being less well educated, for example, can lead to engaging in some risky patterns of behavior, and these risks increase health inequalities.3 But even this mechanism fails to explain the general effect of socioeconomic inequality on health (Marmot 2004), and researchers are exploring more promising mechanisms. But even if more work is needed to clarify the exact mechanisms, it is not unreasonable to talk here about the social "determinants" of health (Marmot 1999). To the extent that these social determinants are socially controllable, we clearly face questions of distributive justice. In some ways, gender and race or ethnic inequalities are easier to address than class inequalities. Since these kinds of inequalities are the result of social exclusion and other unjust practices aimed at vulnerable groups, we are generally and justifiably inclined to view them as inequities. Identifying them, however, does not make them easy to remedy. Socioeconomic inequalities pose harder questions, since few people are radical egalitarians opposing all forms of such inequality. Many who are not at all troubled by significant inequalities in income, wealth, or opportunities for a higher quality of life are particularly troubled by health inequalities. They believe that a socioeconomic inequality that otherwise seems just becomes unjust if it contributes to health inequalities. Is every health inequality that results from unequally distributed social goods unjust? If there is an irreducible health gradient across socioeconomic groups, does that make the very existence of those inequalities unjust? Does a concern about health inequalities drive us to even more radical forms of social and economic egalitarianism than we had imagined we were committed to?

Source: From "Justice, Health, and Health Care," in Medicine and Social Justice: Essays on the Distribution of Health Care, ed R. Rhodes, M. Battin, and A. Silvers (Oxford University Press, 2012), 10-14. Reprinted with permission.

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health 459

Alternatively, are some health inequalities the result of acceptable trade-offs? Perhaps they are simply unfortunate by-products of inequalities that work in other ways to help worse-off groups. For example, it is often claimed that permitting inequality provides incentives to work harder, thereby stimulating growth that will ultimately benefit the poorest groups. To whom must these trade-offs be acceptable if we are to consider them just? Are they acceptable only if they are part of a strategy aimed at making the situation more just? In our judgments about justice, does it matter exactly how social determinants produce inequalities in health status? Unfortunately, these questions have been almost totally ignored within the field of bioethics, as well as within ethics and political philosophy more generally.4 With some significant exceptions, bioethics has not looked "upstream" from the point of delivery of medical services to the role of the health-care system in improving population health. Bioethics has even more rarely looked further upstream to the distribution of social goods that determine the health of societies (cf. Marchand et al. 1998). This omission is quite striking, for bioethics here lags behind both social science and policy concerns. The social science literature on social determinants of health has grown impressively for a quarter of a century. More important, a concern about health equity and its social determinants has emerged as an important consideration in the policies of several European countries over the past two decades (Benzeval et al. 1995; Ostlin and Diderichsen 2001). WHO has devoted growing attention to inequalities in health status and the policies that cause or mitigate them. So have research initiatives, such as the Global Health Equity Initiative, funded by the Swedish International Development Agency and the Rockefeller Foundation (cf. Evans et al. 2001). In the past decade, there has emerged a growing social science and policy concern about health disparities in the United States, although its focus has largely been on issues of race, not class, with a few important exceptions. To fill this bioethical gap, I shall argue that we must supplement a widely cited intuitive analysis of what health inequalities are unjust—those that are avoidable, unnecessary, and unfair (Dahlgren and Whitehead 1991; Whitehead 1992)—with a more systematic account. f. . .1 I turn f. . .1 to Rawls's theory of justice as fairness [. . .1 to illustrate what an account of the just distribution of the determinants of health might look like. My contention is that—quite unintentionally—Rawls's theory provides a defensible account of how to distribute the social determinants of health fairly and thus tells us something useful about

when health inequalities are unjust. Specifically, each of his principles of justice as fairness governs a cluster of key social determinants of health. General conformance with these principles will consequently flatten the socioeconomic gradient of health as much as we can reasonably demand (and, arguably, considerably more than we observe even in wealthy nations with developed social welfare protections). In effect, social justice in general is good for population health and its fair distribution. There will remain a problem of residual health inequalities produced by otherwise justifiable inequalities, and we shall consider the extent to which justice as fairness gives any guidance about them. f. . .1 First, however, I shall describe some of the findings from recent social epidemiological work and point to their serendipitous connection to Rawls's principles of justice. Social Determinants of Health: Some Basic Findings

Four central findings in the literature on the social determinants of health have implications for an account of justice and health inequalities. First, if we look across nations, the national income/health gradients we observe are not the result of some fixed or determinate laws of economic development but are influenced by social policy choices. The wealth of nations matters up to a relatively modest point, but policies always matter, and arguably they matter more in poor countries. Second, if we look within societies, the individual SES/health gradients we observe are not just the result of the deprivation of the poorest groups. Rather, a gradient in health operates across the whole socioeconomic spectrum within societies even where universal medical coverage and other welfare programs are in place. Third, although there is broad agreement that the degree of social inequality contributes to the steepness of a society's health gradient, there is conflicting evidence and a complex debate about whether the steepness of the gradient is affected by the degree of income inequality in a society. Fourth, there are reasonable hypotheses identifying some social and psychosocial pathways through which inequality affects health. These causal pathways can be molded or changed by specific policy choices that should be guided by considerations of justice, and they point to institutions that are appropriately governed by Rawls's principle of justice as fairness. Cross-National Evidence on Health Inequalities

The finding that prosperity is related to health, whether measured at the level of nations or individuals, might

II all

1111.1.1

8 Public Health

lead one to the conclusion that these income/health gradients are inevitable. Bt .t evidence suggests otherwise. Figure 1 shows the relationship between the wealth and health of nations as measured by per capita gross domestic product (GDPpc) and life expectancy. There is a clear association between GDPpc and life expectancy, but only up to a point. The relationship levels off beyond about $8,000 GDPpc, with virtually no further gains in life expectancy. This leveling effect is most apparent among the advanced industrial economies that largely account for the upper tail of the curve in Figure 1. Closer inspection of this figure points out some startling discrepancies. Though Cuba and Iraq (before the second Iraq war) are equally poor (GDPpcs about $3,100), life expectancy in Cuba exceeds that in Iraq by 17.2 years in 1995 (and no doubt by more now). And while the difference between the GDPpc for Costa Rica and the United States is enormous (about $21,000), Costa Rica's life expectancy is nearly the same as that of the United States. In fact, despite being the richest nation on earth, Lie United States performs rather poorly on health indicators (i.e., forty-fourth in life expectancy at birth).5 Taken together, these observations support the notions that the relationship between economic development and health is nor. fixed and that national health achievement is mediated by factors other than

aggregate wealth. Culture, social organization, and government policies, among others, are important in determining population health and thus in explaining the differences in health outcomes among nations. Variations in these socially controllable factors are thus important from the perspective of social justice. Individual SES and Health

At the individual level, numerous studies have documented the "socioeconomic gradient." On this gradient, each increment up the socioeconomic hierarchy is associated with improved health outcomes over the rung below (Black et al. 1988; Davey-Smith et al. 1990; Pappas et al. 1993; Adler et al. 1994). Note that this relationship is not simply a contrast between the health of the rich and the poor but is observed across all levels of SES (remember that not all the components of SES—income, education, social class—behave identically; see footnote 2). Though the effects of income or wealth work across the SES spectrum, the gradient is steeper at lower income levels, with considerable flattening out at the highest income levels. This "concavity" of the gradient means that transfers of resources from the best-off to the worst-off SES groups would improve aggregate health and would have little negative effect, if any, on the best-off groups. (I return to this point in my later discussion of Rawls's difference principle.)

90 —

Japan 80 —

Cuba Costa Rica Georgia , • 0,

• •

• • • ea

340•ci• ° • • o. 0 .5 a • • . • ,eZertl. • • • -

.t 70 — co

•

•

Luxembourg

Sc. o •-w-• d • • •

•

•

• USA

•

• • Brunei

Qatar

• Libya •

A • 60

• ••40 Iraq

ai

ti

•

di • •

•el .0.••

X 11.1

•,..

50

•

Botswana

40 — •

Sierra Leone

30 0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

Per Capita Gross Domestic Product (1995 USS purchasing power parities)

Figure 1 Relationship between Country Wealth and Life Expectancy Source: United Nations Human Development Report Statistics 1998.

40,000

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health

What is particularly notable about the SES gradient is that it does not appear to be explained by differences in access to health care. Steep gradients have been observed even among groups of individuals, such as British civil servants, with adequate access to health care, housing, and transport (Davey-Smith et al. 1990; Marmot et al. 1998). The Whitehall studies of British civil servants provide strong support for such a gradient, not only for mortality rates and life expectancy but also for morbidity rates across a range of diseases. Strikingly some of these studies show that civil service rank explains more of the risk of premature heart disease among ranks than such standard risk factors as smoking, serum cholesterol level, or blood pressure (Marmot et al. 1984). Importantly, the steepness of the gradient varies substantially across societies. Some societies show a relatively shallow gradient in mortality across SES groups. Others, with comparable or even higher levels of economic development, show steeper gradients in mortality rates across the socioeconomic hierarchy.

Relative Income and Health

A lively debate exists in the empirical literature about whether income inequality, as opposed to absolute levels of income, helps to determine the steepness of the gradient. Were this true, we might find that middle income groups in a country with high income inequality have lower health status than comparable or even poorer groups in a country with low income inequality. There is some evidence for this pattern in the United States if variations among states in inequality are examined (Kennedy et al. 1998; Lynch et al. 1998; Subramanian and Kawachi 2006), but there are conflicting study results as well (Mellor and Milyo 2002; Deaton and Lubotsky 2003). This effect, if it can be established, is apparent in Figure 2, where the prevalence of selfreported fair/poor health is higher for almost every income group (and the gradient is steeper) for those living in states with the highest income inequality (Kennedy et al. 1998). This effect of shifting the curve, if it is supported by the empirical findings, would suggest that income inequality has an effect over and above what is implied by the concavity of the SES gradient itself.

35 —

30 —

Percent reporting fair/poor health

---- Low Inequality —o— HighInequality

25 —

20 —

15 —

10 —

The income-poor health curve is shifted up in the states with the highest income inequality for almost the entire population

5—

0 0P C1

0PC1 •scl' N.4r

r,ce

0PC1 _11 ;1.D' tir‘

0PCI cl•

_,,,,,;r‘ _se t2V

sir t'Sr

Household income

Figure 2 Self-Rated Health and Individual Household Income

II

.111

111111

8 Public Health

Leaving aside the contentious results of these studies,6 it is worth noting why the income relativity thesis has attracted such interest. If it is true, it is not just the size of the economic pie but how the pie is shared that matters for population health. It is not the absolute deprivation associated with low economic development (lack of the basic conditions necessary for health, such as clean water, adequate nutrition and housing, and general sanitary living conditions) that explains health differences among developed nations but the degree of relative deprivation within them. Relative deprivation refers not to a lack of the goods that are basic to survival but rather to a lack of sources of self-respect that are deemed essential for full participation in society. (These "recognitional" features of just arrangements are emphasized in Rawls account, as we shall see.) If the income relativity thesis can be established, it would then require elaboration of the causal mechanisms that underlie it.7 The income relativity thesis is interesting and worth further investigation, but it has only a modest bearing on the claims I pursue [. . .1. As we shall see, there is strong evidence that health inequalities are reduced if we concentrate first on making the worst-off groups socially and economically as well off as possible, paying special attention to improvements in human capital and investments in public health. If true, the income relativity thesis would strengthen support for giving priority to making the worst cff as well off as possible, for it would make an even stronger claim that it works to the advantage of all. In addition, it might point us to other mechanisms that we may have to address. Pathways Linking Social Inequalities to Health Inequalities

To address key issues about sccial policy and distributive justice, we must better understand the mechanisms underlying the correlations between health outcomes and social inequalities in income, education, and other factors. Unless we know something about these underlying causal mechanisms, we cannot focus properly on the socially controllable factors that a just policy must regulate. Fortunately, there are plausible and researchable pathways through which social inequalities produce inequalities in health. The first empirical result I highlighted earlier was the claim that social policy more than the wealth of nations, explains variations in health outcomes across countries. Developing countries that invest heavily in human capital, for example in education, have better health outcomes. Indeed, one of the strongest predictors of life expectancy among developing countries is adult literacy, particularly the disparity between male

and female adult literacy, which explains much of the variation in health achievement among these countries after accounting for GDPpc. Similarly, in the United States, differences between the states in women's status—measured in terms of their economic autonomy and political participation—are strongly correlated with female mortality rates (Kawachi and Kennedy 1997; Kawachi et al. 1999). Even if we do not know all the underlying mechanisms through which improved education, including female literacy, improves health outcomes, we have good evidence that investment in this socially controllable factor has measurable health effects. The second empirical thesis I highlighted concerned the strong effect of a socioeconomic gradient of health. This gradient is demonstrated even where socioeconomic inequality does not mean absolute deprivation in material goods or access to health care—as in the Whitehall studies. Indeed, if we focus on the British civil servants, one important variable in their job situations concerns their degree of control over work (Marmot 2004). Such control may affect the kinds of stress people are exposed to, and this in turn may affect basic immune mechanisms or may directly impact the course of certain diseases. Uncovering mechanisms here is a complex research program that may require work on both animals and humans, including some carefully selected natural experiments. How readily we can convert this understanding of mechanisms into modifications of social policy, including, for example, the organization of the workplace, remains to be seen. Our third thesis concerned the controversial evidence about the degree of income inequality and its effects on health. It may be premature to look for mechanisms, since the basic association is controversial. Still, we may better understand some of the mechanisms through which income inequality might work if we note some of the related studies. In the United States, the states with the most unequal income distribution invest less in public education, have more uninsured persons, and spend less on social safety nets (Kaplan et al. 1996; Kawachi and Kennedy 1997). Differences in human capital investment are particularly striking. These are demonstrated for educational spending and, more importantly, for outcomes; even when controlling for median income, income inequality explains about 40 percent of the between-state variation in the percentage of fourthgrade children who are below the basic reading level. Similarly strong associations are seen for high school dropout rates. These data clearly show that educational opportunities for children in states with high income inequality are quite different from those in

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health

states. Furthermore, these early effects not only have an immediate impact on health, increasing the likelihood of premature death during childhood and adolescence (as evidenced by the much higher death rates for infants and children in the highinequality states) but also have lasting effects showing up later in life as part of the SES gradient in health (Bartley et al. 1997; Davey-Smith et al. 1990). These societal mechanisms are tightly linked to the political processes that influence government policy. For one thing, income inequality appears to affect health by undermining civil society. Income inequality erodes social cohesion, increasing social mistrust and reducing participation in civic organizations—two features of civil society (Kawachi and Kennedy 1997; Kawachi et al. 1997). Lack of social cohesion, in turn, is reflected in significantly less political activity (e.g., voting, serving in local government, volunteering for political campaigns), thus undermining the responsiveness of government institutions in addressing the needs of the worst-off groups. This is demonstrated not only by the human capital investment data presented earlier but also by the lack of investment in human security. States with the highest income inequality, and thus the lowest levels of social capital and political participation, provide far less generous social safety nets. more egalitarian

An Intuitive Analysis of Health Inequities

In the public health literature (Braveman 1999), there is an influential intuitive answer to [our starting question about justice,] namely, Whitehead's (1992) and Dahlgren and Whitehead's (1991) claim that health inequalities count as inequities when they are avoidable, unnecessary, and unfair. How do these conditions fit together? Presumably, inequalities that are avoidable and unnecessary but not unfair do not count as inequitable. For example, if lung cancer rates differed between smokers and nonsmokers, and if smoking rates were not themselves strongly influenced by SES, ethnicity, or gender but seemed completely a matter of individual, voluntary, informed choice, then we might not think the cancer rate inequality unfair and thus not inequitable. Similarly, an inequality may be unfair but also unavoidable or even necessary, given our inability to change certain conditions. For example, suppose we learned that the mortality rate of certain workers, say truck drivers, means that their life expectancy is lower than that of other workers and many believe (intuitively) that this is unfair. We may be able to reduce that higher mortality rate through road and equipment improvements and through better training of drivers

and enforcement of traffic laws, but suppose there remains an elevated mortality rate that seems unavoidable, or perhaps necessary given the importance of trucking. Some may insist that the inequality is still unfair, if unavoidable, even though they would conclude that the inequality is not an inequity. If we can agree on what is avoidable, unnecessary, and unfair, and if this analysis is correct, then we can agree on which inequalities are inequitable. The Whitehead-Dahlgren analysis is deliberately broad. It not only includes socioeconomic differences but others as well. Age, gender, race, and ethnic differences in health status exist that are independent of socioeconomic differences, and they raise distinct questions about equity or justice. For example, should we view the lower life expectancy of men compared to women in developed countries—as much as six years' difference in some observations—as an inequity?8 If it is rooted in biological differences that we do not know how to overcome, then, according to this analysis, it is not avoidable and therefore not an inequity. Perhaps not all of it is unavoidable—higher smoking rates for men or higher workplace risks may explain some of the difference. But how much? This is not an idle controversy: Taking average rather than gender-differentiated life expectancy in developed countries as a benchmark or goal will yield different estimates of the inequity women face in some developing countries.9 In any case, the analysis of inequity here is only as good as our understanding of what is avoidable or unnecessary. The same point applies to judgments about fairness. Is the poorer health status of a social class or ethnic group that engages in heavy drug and alcohol use unfair? We may be inclined to say it is not unfair, provided that the use or avoidance of drugs and alcohol is truly voluntary. But if many people in an ethnic group or class behave similarly there may also be factors at work that reduce the autonomous nature of their behavior and how much responsibility they should bear (Wikler 1978; Roemer 1995). The analysis thus leaves us with judgments about responsibility that are complex and unresolved; as a result, we have continuing disagreements about fairness (or avoidability). The poor in many countries have no clean water, sanitation, adequate shelter, basic education, vaccinations, and prenatal and maternal care. As a result of some or all of these factors, there are infant mortality differences between them and richer groups. Since social policies could supply the missing determinants of infant health, the inequalities are avoidable. Are these inequalities also unfair? Most of us would immediately think that they are, perhaps because we believe that international (Pogge 2002) or domestic

II

II 11.1,1

8 Public Health

policies that create and sustain poverty are unjust. We also believe that social policies that compound poverty with lack of access to the determinants of health are doubly unfair. Of course, libertarians would disagree. They would insist that what is merely unfortunate is not unfair; on. their view, we have no obligation of justice, as opposed to charity, to provide the poor with what they are missing. Many of us might reject the libertarian view as itself unjust because of this dramatic conflict with our beliefs about poverty and our social obligations to meet peopleb basic needs. The problem becomes more complicated, however, when we remember one of the basic findings from the literature on social determinants: We cannot eliminate health inequalities sim?ly by eliminating poverty. Health inequalities persist even in societies where poor people have access to all the determinants of health noted previously. Furthermore, these health inequities persist as a gradient of health throughout all SES classes, not just between the very poorest groups and those immediately above them. Some of the mechanisms may be difficult to avoid and may depend on deep structural features of an economy. Thus, educational and skill differences that steer people into hard manual labor rather than nonmanual labor significantly affect the rate at which poor health is self-reported (Case and Deaton 2003). Whether we consider these mechanisms and their effects unfair may depend in part on our views about the fairness of overall social inequality. At this point, many of us are forced to re-examine what we believe about the justice of the remaining socioeconomic inequalities. Unless we believe that all socioeconomic inequalities (or at least all inequalities we did not choose) are unjust—and very few do so—we must consider more carefully the problem created by the health gradient and the fact that it becomes steeper as inequality increases. Our judgments about the intuitive notion of fairness in the DahlgrenWhitehead account provide less guidance in thinking about the broader issue of t he social determinants of health inequalities even if we, rightly or wrongly felt confident in appealing to them when rejecting the libertarian position. Indeed, we may even believe that some socioeconomic inequality is unavoidable or even necessary and therefore is no: unjust. A Rawlsian Analysis of Health Inequities

One reason we develop general ethical theories, including theories of justice, is to provide a framework for settling important disputes about conflicting moral beliefs or intuitions of the sort facing the WhiteheadDahlgren analysis. For example, in A Theory of Justice,

Rawls sought to leverage our relatively broad liberal agreement on principles guaranteeing certain equal basic liberties into an agreement on a principle limiting socioeconomic inequalities, a matter on which liberals considerably disagree (J. Cohen 1989). His strategy was to show that a social contract designed to be fair to free and equal people ("justice as [procedural] fairness") would not only justify the choice of those equal basic liberties but would also justify the choice of principles guaranteeing equal opportunity and limiting inequalities to those that work to make the worst-off groups fare as well as possible. Rawls's account, though developed to answer this general question about social justice, also provides principles for the just distribution of the social determinants of health, unexpectedly adding to its scope and power as a theory. The extra power of the theory is a surprise, since [. . .] Rawls deliberately simplified the construction of his theory by assuming that his contractors are fully functional over a normal life span. Examining the social determinants of health inequalities from the perspective of Rawls's theory is particularly appealing because justice as fairness is egalitarian in orientation (Daniels 2003) and yet justifies certain inequalities, such as those in income and wealth, that contribute to health inequalities. In addition, my extension of Rawls links the protection of health to the protection of equality of opportunity, again setting up the potential for internal conflict. To see whether this combination of features leads to contradictions in the theory or to insight into the problem, we must examine the issue in more detail. Rawls on Allowable Inequalities

How does Rawls justify socioeconomic inequalities? Why wouldn't free and equal contractors simply insist on strictly egalitarian distributions of all social goods, just as they insist on equal basic liberties and equal opportunity? Rawls's answer is that it is irrational for contractors to insist on equality if doing so would worsen their lifetime prospects. If incentives to develop skills and take risks increase social productivity, then the social pie that gets divided may be increased so much that less than equal shares for those who are worst off are still absolutely larger than equal shares of a smaller social pie. Specifically, Rawls argues that contractors would choose the difference principle, which [. . .] permits inequalities provided that they make the worst-off groups in society as well off as possible.rn The argument for the difference principle appears to suggest that relative inequality is less important than absolute well-being, a suggestion that is in tension with other aspects of

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health

Rawls's view. The tension is illustrated by Rawls's insistence that inequalities allowed by the difference principle should not undermine the value of political liberty or the requirements of fair equality of opportunity. The priority given these other principles over the difference principle thus limits the inference that Rawls has no concern about relative inequality. Specifically, as we shall see, these principles work together to constrain inequality and to preserve the social bases of self-respect for all. Two points will help avoid misunderstanding of the difference principle and its justification. First [. . .] it is not a mere trickle-down principle but one that requires maximal flow to help the worst-off groups. The worst off, and then the next worst off, and so on (Rawls [1971] calls this "chain connectedness") must be made as well off as possible, not merely somewhat better off, as a trickle-down principle implies. The difference principle is thus much more demanding than a principle that would permit any degree of inequality provided that there was some trickle of benefits to the worst off." Indeed, it is more egalitarian than alternative principles that merely assure the worst off a "decent" or "adequate" minimum. Part of the rationale for the more demanding principle is that it would produce less strain of commitment, less sense of being unfairly left out, at least for those who are worst off, than principles that allow more inequality (Cohen 1989). Indeed, from what we have learned about the social determinants of health, the more demanding difference principle would also produce less health inequality than any other principles that allow inequalities. By flattening the health gradient, it also benefits middle-income groups, not simply the poorest. In this regard, its benefits are important beyond the level where we have helped the worst off to achieve "sufficiency." This point provides a reply to those who suggest that the difference principle has no appeal once the worst off are sufficiently provided for (Gutmann and Thompson 1996). Second, when Rawlsian contractors evaluate how well off the principles they choose will make them, they are to judge their well-being by an index of "primary social goods" [. . .]. The primary social goods, which Rawls thinks of as the "needs of citizens," include liberty, powers, opportunities, income and wealth, and the social bases of self-respect. (These objective measures of well-being should be contrasted with measures of happiness or desire satisfaction that are familiar from utilitarian and welfare economic perspectives.) In his exposition of the difference principle, Rawls illustrates how it will work by asking us to consider only the

simpler case of income inequality. In doing so, he assumes that the level of income will correlate with the level of other social goods on the index. This simplification can be misleading. In crucial cases—including those involving health—the correlation may not occur, and it will be essential to use the whole index, rather than just the income component, to measure how well off people are.0 For example, let us suppose that having "democratic" control over one's workplace is crucial to self-realization and the promotion of self-esteem (Cohen 2001). Marmot (2004) documents how hierarchical workplaces may create health inequalities through the effect of status on stress. Suppose further that hierarchical workplaces are more efficient than democratic ones, providing higher incomes for the worst-off workers than democratic workplaces. Then the difference principle does not clearly tell us whether the hierarchical workplace contains allowable inequalities since the worst off are better off in some ways (economic) but worse off in others (health and thus opportunity and selfrespect). Without knowing the weighting of items in the index, we cannot use it to say clearly what inequalities are permitted. When we evaluate which income inequalities are allowable by asking which ones work to make the worst-off groups as well off as possible, we must judge how well off groups are by reference to the whole index of primary goods and not simply the resulting income. This point is of particular importance in the current discussion. My extension of Rawls [. . .] treats health status as a determinant of the opportunity range. Since opportunity is included in the index of primary social goods, the effects of health inequalities are thereby included as well. I also noted earlier that the social bases of self-respect are among the primary social goods on the index. From Marmot's (2004) work, we may infer that hierarchical workplaces not only undermine health, and thus opportunity, but do so because there is inadequate protection of selfrespect. Clearly, we must use the whole index, not the simplifying assumption that income correlates with all other items on it. Unfortunately, Rawls says very little about how items in the index are to be weighted. This is one of the crucial points on which the theory says less than we might have wished. Therefore, we have little guidance about how these primary goods are to be traded off against each other in constructing the index. This silence pertains not only to the use of the index in the contract situation but also to its use by a legislature trying to apply the principles of justice in a context where many features of a society are known.

Ii JI

111111

8 Public Health

Flattening the SES Gradient: of Health

Now that we have seen why Rawls believes that inequalities are allowable under some conditions, we can consider how those general inequalities affect the distribution of health and contribute to health inequalities. If allowable general inequalities magnify health inequalities, we might conclude that justice, as Rawls conceives it, is not good for our health. Instead, we shall see why social justice, as described by Rawls's principles, is good for our health and promotes its fair distribution.° 1. . .1 To understand this claim, let us start with the ideal case, a society governed by Rawls's principles of justice as fairness that seeks to achieve what Rawls calls "democratic equality" (Rawls 1971; Daniels 2003). Consider what the society requires in distributing the social determinants of health. In such a society all are guaranteed equal basic liberties (Rawls 1982a), including effective exercise of (or worth of) political participation (Daniels 1975; cf. Rawls 1995). In addition, there are institutional safeguards aimed at assuring all, richer and poorer alike, the value of effectively exercisable political participation rights. Without such protection, basic capabilities of citizens cannot develop. Recognizing that all citizens have these capabilities protected is critical to preserving self-esteem, on Rawls's view. In requiring institutional support for political participation rights, Rawls (1995) rejects the claim that freedom of speech of the rich is unfairly restricted by limiting their personal expenditures on their own political campaigns 1. . .1. After all, the limitation does not unduly burden the rich compared to others. There is evidence that political participation is itself a social determinant of health. Therefore, the Rawlsian ideal assures institutional protections that counter the usual effects of socioeconomic inequalities on participation and thus on health. The Rawlsian ideal of democratic equality also involves conformity with a principle guaranteeing fair equality of opportunity 1. . .1. Not only is discrimination prohibited by the principle, but it requires strong measures to mitigate the effects of socioeconomic inequalities and other social contingencies on opportunity. In addition to equitable public education, such measures would include developmentally appropriate day care and early childhood interventions to promote the development of capabilities independently of the advantages of family background. Such measures match or go beyond the best models of such interventions we see in European efforts to provide day care and early childhood education. We also note that the strategic importance of education for protecting equal opportunity has implications or all levels of education,

including access to graduate and professional education and continuing adult education in a world where work skills must be continuously upgraded. The equal opportunity principle also requires extensive public health, medical, and social support services aimed at promoting normal functioning for all 1. . .1. It even provides a rationale for the social costs of reasonable accommodation of persons with incurable disabilities, 1. . .1. Because the principle aims at promoting normal functioning for all as a way of protecting opportunity for all, it aims at both improving population health and reducing health inequalities. Obviously, this requires universal comprehensive health care, including public health, primary health care, and medical and social support services. To act justly in health policy, we must understand how socioeconomic (and other) inequalities work to produce differential health outcomes. Suppose that, as Marmot (2004) argues, structural and organizational features of the workplace that induce stress and loss of control tend to promote health inequalities. If this is true, then those features should be modified to reduce their negative effects on health as a public health requirement of the equal opportunity approach; this is on a par with the requirement to reduce exposure to toxins in the workplace [. . .1. Moreover, modifying such features would be favored by Rawls's emphasis on securing the social bases of self-respect. Finally in the ideal Rawlsian society, the difference principle significantly reduces allowable inequalities in income and wealth and, more generally, in the index of primary social goods.14 The inequalities allowed by this principle (together with the principles assuring equal opportunity and the value of political participation) are probably more constrained than those found in even the most industrialized societies. If so, then the inequalities that conform to the difference principle would produce a flatter gradient of health inequality than we currently observe in even the more extensive welfare systems of Northern Europe. Earlier, I noted the lively ongoing debate about the effects of income inequality on health inequalities. I believe my claim that the difference principle would flatten gradients more than principles calling for weaker constraints on inequality is true even if the income relativity thesis turns out not to be supported by the evidence. The difference principle is about how well off the worst off are in absolute terms, and though applying it may reduce relative inequality it would affect mainly health inequalities even if the income relativity thesis does not hold. Those who oppose the income relativity thesis in any event endorse the basic facts about a socioeconomic gradient of health. Pulling the bottom

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health

up in absolute terms—coupled with other social policies of the sort assured by the other principles of justice as fairness—would therefore flatten the gradients. If income relativity does have a further effect on health inequalities, then my claim is strengthened further. In short, Rawls's principles of justice as fairness regulate the distribution of the key social determinants of health, including the social bases of self-respect. Nothing about the theory, or my extension of it E. . .1, should make us focus narrowly on medical services. Properly understood, justice as fairness tells us what justice requires in distributing all social determinants of health. Nor, I should add, does the fact that we should focus on the distribution of goods in addition to health care mean that health care is no longer of special moral importance. Not only does health care make its own significant contribution to population health, but even if all the other socially controllable factors were properly distributed, some pathology would remain and we would have to respond to its impact on opportunity. Health care may not be the only good of special moral importance, given the moral importance of health, but it remains a good of special moral importance. Residual Inequalities: Unjust or Not?

We still face an interesting theoretical issue, as well as many practical issues in our nonideal or partially just world. First, the theoretical issue. Even if the Rawlsian distribution of the determinants of health flattens health gradients further than those observed in the most egalitarian developed countries, we must still expect some health inequalities. In part, this may happen because we may not understand all the relevant causes or the interventions for modifying them. The theoretical issue is whether the theory requires us to reduce further those otherwise justifiable inequalities because of the inequalities in health status they create. We should not further reduce those inequalities if doing so reduces productivity so much that we can no longer support the institutional measures we already employ to promote health and reduce health inequality. Our commitment to reducing health inequality should not require steps that threaten to make health worse off for those with less-than-equal health status. So, the theoretical issue reduces to this: Would it ever be reasonable and rational for contractors to accept a tradeoff in which some health inequality is allowed in order to produce some nonhealth benefits for those with the worst health prospects? We know that in real life people routinely trade health risks for other benefits. They do so when they commute longer distances for a better job or take a ski vacation. Some such trades raise questions of fairness. For example, when is hazard pay

a benefit workers gain only because their opportunities are unfairly restricted, and when is it an appropriate exercise of their autonomy E. . .1? Many such trades are ones we think it unjustifiably paternalistic to restrict; others we see as unfair. Rawlsian contractors, however, cannot base such trades on specific knowledge of their own values. They cannot decide that their enjoyment of skiing makes it worth the risks to their knees or necks. To make the contract fair to all participants, and to achieve impartiality, Rawls imposes a thick "veil of ignorance" that blinds them to all knowledge about themselves, including their views of the good life. Instead, they must judge their well-being by reference to the index of primary social goods (noted earlier) that includes a weighted measure of rights, opportunities, powers, income and wealth, and the social bases of self-respect. But our theoretical question about residual health inequalities reminds us that the theory says too little about the construction of the index to provide us with a clear answer. One of Rawls's (1971) central arguments for singling out a principle protecting equal basic liberties and giving it (lexical) priority over his other principles of justice is his claim that once people achieve some threshold level of material well-being, they will not trade away the fundamental importance of liberty for other goods. Making such a trade might deny them the liberty to pursue their most cherished ideals, including their religious beliefs, whatever they turn out to be. Can we make the same argument about trading health for other goods? There is some plausibility to the claim that rational people should not trade health for other goods. Loss of health may prevent us from pursuing what we most value in life. We do, after all, see people willing to trade almost anything to regain health once they lose it. If we take this argument seriously we might conclude that Rawls should give opportunity including the effects of health status, a heavier weighting in the construction of the index than income alone." Such a weighting would mean that absolute increases in income for the worst off that might otherwise have justified increasing relative income inequality, according to the difference principle, now fail to justify those inequalities because of the negative effects on opportunity. Although the income of the worst off would increase, they are not better off according to the whole (weighted) index of primary social goods, and so the greater inequality is not permitted. Rawls's simplifying assumption that income correlates with other goods fails in this case (as it did in the hypothetical example of workplace democracy cited earlier).

II 411

IIILIJ

8 Public Health

Nevertheless, there is also strong reason to think that the priority given to health, and thus opportunity, is not as clear-cut as the previous argument implies, especially where the trade is between a risk to health and other goods that peop: e highly value. Refusing to allow any (ex ante) trades of health risks for other goods, even when the background conditions are otherwise fair, may seem unjustifiably paternalistic, perhaps in a way that refusal to allow trades of basic liberties is not. I propose a pragmatic route around this problem, one that has a precedent elsewhere in Rawls. Fair equality of opportunity, Rawls admits, is only approximated even in an ideally just system, because we can only mitigate, not eliminate, the effects of family and other social contingencies (Fishkin 1983). For example, only if we were willing to violate widely respected parental liberties could we intrude in family life and "rescue" children from parental values that arguably interfere with equal opportunity. Similarly, though we give general priority to equal opportunity over the difference principle, we cannot achieve complete equality in health any more than we can achieve completely equal opportunity. Even ideal theory does not produce perfect justice. Justice is always rough around the edges. Specifically, if we had good reason to think that "democratic equality" had flattened inequalities in accord with the principles of justice, then we might think we had done as much as was reasonable to make health inequalities fair to all. The residual inequalities that

emerge with conformance to the principles are not a compromise with what justice ideally requires; they are acceptable as just. So far, we have been considering whether the decision about such a trade-off can be resolved from the perspective of individual contractors. Instead, suppose that the decision is to be made by the legislature in a society that conforms to Rawls's principles. Because those principles require effective political participation by all socioeconomic groups, we can suppose that the groups most directly affected by any trade-off decision have a voice in the decision. Since there is a residual health gradient, groups affected by the trade-off include not only the worst off but those in the middle as well. Developing a democratic process that involved deliberation about the trade-off and its effects might be the best we could do to answer the theoretical question E. . .1. In contrast, where the fair value of political participation is not adequately assured—and we doubt that it is so assured in even the most democratic societies—we have much less confidence in the fairness of a democratic decision about how to trade health against other goods. It is much more likely that those who benefit most from the inequalities, that is, those who are better off, also wield disproportionate political power and will influence decisions about trade-offs to serve their interests. It may be that the use of a democratic process in nonideal conditions is the fairest resolution we can practically achieve, but it still falls well short of what an ideally just democratic process would involve. E. . .1

Notes 1.

2.

3.

4.

Deaton (personal communicE.tion, July 13, 2006) cautions against thinking that the various components of SES (income, education, social class) all work the same way and that the convenience of combining them may be outweighed by the inaccuracy that can result. See also Deaton (2002a,b). Deaton (personal communication, July 13, 2006) and Case and Deaton (2003) poir t to the impact of ill health on early retirement and thus on income reduction; Case and Deaton argue that the significantly worse self-reported health of those in the lowest income quartile compared to those in the highest is primarily driven by the ill health of those out of work, presumably because they are ill. Health selection may have a greater effect on income than class. Cf. Marmot (1994). Behavior risk factors nevertheless contribute significantly to inequality: Jha et al. (2006) show that smoking contributes nearly half of the male mortality in the lowest social class in four countries. Sen (2004) is an important exception.

I.. .1

5.

6.

CIA Factbook, https://www.cia.gov/cia/publications/ factbookkankorder/2102rank.html, accessed July 18, 2006. Sparked by Wilkinson's (1992) paper showing a correlation between national measures of income inequality and the gradient of health inequality among a select group of Organisation of Economic Cooperation and Development (OECD) countries, researchers in the United States and elsewhere found conflicting results during the subsequent decade. The initial positive results from the OECD countries disappear when a full rather than a selected set of them is studied. Negative results are also found primarily in countries more egalitarian than the United States (Sweden, Japan, Canada, Denmark, New Zealand), as noted by Subramanian et al. (2003) and Subramanian and Kawachi (2004). Some negative results are also found in the United States, especially if units of aggregation are smaller, such as municipalities rather than states. A significant number of US studies support the claim, especially at the state level. Positive results are also found in a recent study in Chile, a country with

Daniels: When Are Health Inequalities Unjust?-The Social Determinants of Health

7.

8.

9.

more income inequality than exists in the United States (Subranamian et al. 2003). This pattern of seemingly conflicting results might be explained if income inequality affects health inequalities only above some threshold level of inequality, a threshold crossed in the United States and more unequal countries but not in more egalitarian countries (Subramanian and Kawachi 2004). Alternatively, the effect of income inequality might be countered by more egalitarian social policies, such as the more redistributive health and welfare policies in other OECD countries. This latter alternative highlights an important point about the claim generally made by supporters of the income relativity thesis: Income inequality is a causal factor that works through various mechanisms, political, social, and possibly individual (e.g., stress), and is not claimed to be "by itself" or "per se" a cause of health inequalities. In the United States, in those studies that support the thesis, the strength of this relationship between social inequality and health inequality is not trivial. One study of US metropolitan areas, rather than states, found that areas with high income inequality had an excess of death compared to areas of low inequality that was equivalent in magnitude to all deaths due to heart disease (Lynch et al. 1998). Sen (2004) points to the gender inequality and denies that it is an inequity, not because we could not do something about it but because what we could do, e.g., give men more health care than women, would be discriminatory, violating "process" constraints on fair treatment. Murray's reduction of the male-female gender gap to 2.5 years from an observed 6 years, based on extrapolation from some cases and assumptions about greater exposure to risk, was criticized as possible gender bias in the construction of DALYs. See Anand and Hanson (1998). 1. . .1

10. A careful discussion of Rawls's argument for the difference principle and the extensive critical literature it has generated is beyond the scope of this chapter. It is important, however, to distinguish Rawls's own social contract argument from the many informal and intuitive reformulations of it. See Barry (1989) and G. A. Cohen (1992; 1995) Some controversy is fed by Rawls's (1971: Sect. 13) different formulations of the difference principle; some comes from the complexity of other assumptions involved in his argument, such as "chain connection" and "close-knittedness"; some follows from disagreement with giving such strong priority to the worst off, regardless of the benefits lost to others and regardless of how well off the worst off are. For a useful discussion, see Williams (1995). 11. What was said previously applies to the ideal case, where a society is in general conformance with the principles of justice. In a nonideal setting, where we may be making a system more just, Rawls is willing to say that an improvement is not unjust as long as it moves the worseoff groups closer-even if not maximally closer-to being as well off as possible. 12. In thinking about the impact of income inequality on health, Deaton (personal communication, July 2006) makes a related point when he cautions that we should think more broadly about social inequality and not income alone. 13. Sen (2004) too makes the point that social justice broadly understood is crucial to identifying and addressing health inequity. 14. G A Cohen (1992) has argued that a strict interpretation of the difference principle would allow few incentive-based inequalities; for a more permissive view, see Daniels (2003). 15. Rawls (1971: 93) does suggest that since fair equality of opportunity is given priority over the difference principle, within the index we can assume that opportunity has a heavier weighting.

References Adler, N. E., Thomas, B., Chesney, M. A., Cohen, S., Folkman, S., Kahn, R. L., and Syme, S. L. 1994. Socioeconomic status and health: The challenge of the gradient. American Psychologist 49: 15-24. Anand, S., and Hanson, K. 1998. DALYs: Efficiency versus equity. World Development 26: 307-10. Barry, B. 1989. Theories of Justice. London: Harvester Wheatsheaf, pp. 213-34. Bartley, M., Blane, D., and Montgomery, S. 1997. Socioeconomic determinants of health: Health and the life course: Why safety nets matter. British Medical Journal 314: 1194. Benzeval, M., Judge, K., and Whitehead, M. (eds.). 1995.

Tackling Inequalities in Health: An Agenda for Action. London: Kings Fund. Black, D., Morris, J. N., Smith, C., Townsend, P. and Whitehead, M. 1988. Inequalities in Health: The Black Report; The Health Divide. London: Penguin Group.

Braveman, P 1999. Monitoring Equity in Health: A Policy-

Oriented Approach in Low- and Middle-Income Countries. Geneva: World Health Organization. Case, A., and Deaton, A. 2003. Broken Down by Work and Sex: How Our Health Declines. Cambridge, MA NBER Working Paper No. 9821. Cohen, G. A. 1989. On the currency of egalitarian justice. Ethics 99: 906-44. . 1992. Incentives, inequality, and community, in G. Petersen (ed.), The Tanner Lectures on Human Values, vol. 13. Salt Lake City: University of Utah Press, pp. 262-329. . 1995. The Pareto argument for inequality. Social Philosophy and Policy 12 (Winter): 160-85. Cohen, J. 2001. Taking people as they are? Philosophy and Public Affairs 30(4): 363-86. Dahlgren, G., and Whitehead, M. 1991. Policies and Strategies to Promote Social Equity in Health. Stockholm: Institute of Future Studies.

II au

!ULU

8 Public Health Daniels, N. 1975. Equal liberty and unequal worth of liberty. In N. Daniels (ed.), Recding Rawls. New York Basic Books, pp. 253-81. . 2003. Democratic equality: Rawls's complex egalitarianism. In S. Freeman (ed.), Companion to Rawls. Oxford: Blackwell, pp. 241-76. Daniels, N., Kennedy, B., and Kawachi, L 1999. Why justice is good for our health: The social determinants of health inequalities. Daedalus 128 (4): 215-51. 2000. Is Inequality Bad for Our Health? Boston: Beacon Press. Daniels, N., Light, D., and Caplan, R. 1996. Benchmarks of Fairness for Health Care Reform. New York: Oxford University Press. Davey-Smith, G., Shipley, M. J., and Rose, G. 1990. Magnitude and causes of socioecor.omic differentials in mortality. Further evidence from the Whitehall Study. Journal of Epidemiology and Community Health 44: 265-70. Deaton, A. 2002a. Policy implications of the gradient of health and wealth. Health Affairs 21 (2): 13-29. . 2002b. The convoluted story of international studies of inequality and health. International Journal of Epidemiology 31 (3): 546-9. Deaton, A., and Lubotsky, D. 2003. Mortality, inequality and race in American cities and states. Social Science and Medicine 56: 1139-53. Fishkin, J. 1983. Justice, Equal Opportunity, and the Family. New Haven, CT: Yale University Press. Gutmann, A., and Thompson, D. 1996. Democracy and Disagreement. Cambridge, MA: Harvard University Press. Jha, P., Peto, R., Zatonski, W , Boreham, J., Jarvis, M. J., and Lopez, A. D. 2006. Social inequalities in male mortality, and in male mortality frorr smoking: Indirect estimation from national death rates in England and Wales, Poland, and North America. Lancetonline D01:10.1016/ So140-6736 (06)68975-7. Kaplan, G. A., Pamuk, E. R., Lyr.ch, J. W, Cohen, R. D., and Balfour, L. 1996. Inequality in income and mortality in the United States: Analysis of mortality and potential pathways. British Medical Journal 3(12): 999-1003. Kawachi, I., Daniels, N., and Robinson, D. 2005. Health disparities by race and class: Why both matter. Health Affairs 24(2): 343-4. Kawachi, I., and Kennedy, B. P. 1997. Health and social cohesion: Why care about income inequality? British Medical Journal 314: 1037-40. Kawachi, I., Kennedy, B. P., Lochner, K., and Prothrow-Stith, D. 1997. Social capital, income inequality and mortality. American Journal of Public Health 87: 1491-8. Kawachi, I., Kennedy, B. P., Prod- row-Stith, D., and Gupta, V 1999. Women's status and the health of women: A view from the states. Social Science and Medicine 48: 21-32. Kennedy, B. P., Kawachi, I., Glass, R., and Prothrow-Stith, D. 1998. Income distribution, socioeconomic status, and self-rated health: A U.S. multi-level analysis. British Medical Journal 317: 917-21. Lynch, W , Kaplan, G. A., Pamuk, E. R., Cohen, R. D., Heck., K. E., Balfour, J. L., and Yen, I. H. 1998. Income inequality

and mortality in metropolitan areas of the United States. American Journal of Public Health 88: 1074-80. Marchand, S., Wikler, D., and Landesman, B. 1998. Class, health, and justice. Milbank Quarterly 76: 449-68. Marmot, M. 1994. Social differentials in health within and between populations. Daedalus 123: 197-216. . 1999. Social causes of social inequalities in health. Cambridge, MA: Harvard Center for Population and Development Studies, Working Paper Series 99.01. . 2004. The Status Syndrome: How Social Standing Affects Our Health and Longevity. New York Henry Holt, Times Books. Marmot, M. G., Fuhrer, R., Ettner, S. L., Marks, N. F., Bumpass, L. L., and Ryff, C. D. 1998. Contribution of psychosocial factors to socioeconomic differences in health. Milbank Quarterly 76: 403-48. Marmot, M. G., Hemingway, B. H., Brunner, E., and Stanfield, S. 1997. Contribution of job control and other risk factors to social variations in coronary heart disease incidence. The Lancet 350: 235-9. Marmot, M., Shipley, M. J., and Rose, G. 1984. Inequalities in death-specific explanations of a general pattern. The Lancet 5: 1003-6. Mellor, J., and Milyo, J. 2002. Is exposure to income inequality a public health concern? Lagged effects of income inequality in individual and population health. Health Services Research 38: 1. Ostlin, P., and Diderichsen, E 2001. Equity Oriented National Health Strategy for Public Health in Sweden. Policy Learning Curve Series No. 1. Brussels: European Center for Health Policy. Pappas, G., Queen, S., Hadden, W , and Fisher, G. 1993. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. New England Journal of Medicine 329(2): 103-9. Pogge, T. W. 2002. World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms. Cambridge: Blackwell. Potts, M., and Walsh, J. 2003. Tackling Indian HIV epidemic: Lessons from Africa. British Medical Journal 326: 1389-92. Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. 1982a. The basic liberties and their priorities, in The Tanner Lectures on Human Values, vol. III. Salt Lake City: University of Utah Press. . 1993. Political Liberalism. New York: Columbia University Press. . 1995. Political Liberalism (paperback ed.). New York: Columbia University Press. Roemer, J. 1995. Equality of opportunity, Boston Review, April-May. Available at http://www.bostonreview.net/ BR20.2/Roemer.html. Sen, A. K. 1990a. Justice: Means versus freedoms. Philosophy & Public Affairs 19(2): 111-21. . 1990b. More than 100 million women are missing. New York Review of Books 37: 30 . 2004. Why health equity? In S. Anand, E Peter, and A. Sen (eds.) Public Health, Ethics, and Equity. New York: Oxford University Press, pp. 21-34.

Daniels: When Are Health Inequalities Unjust?—The Social Determinants of Health

Subramanian, S. V, Delgado, I., Jadue, L., Vega, J., and Kawachi, I. 2003. Income inequality and health: Multilevel analysis of Chilean communities. Journal of Epidemiology and Community Health 57: 844-8. Subramanian, S. V., and Kawachi, I. 2004. Income inequality and health: What have we learned so far? Epidemiologic Reviews 26: 8-91. . 2006. Whose health is affected by income inequality? A multilevel interaction analysis of contemporaneous and lagged effects of state income inequality on individual self-rated health in the United States. Health and Place 12(2): 141-56. Villerme, L. 1840. Tableau d'Etat Physique et Moral des Ouvriers, vol. 2. Paris: Renouard. Cited in B. G. Link, M.

E Northridgei, C. Phelan, and M. L Ganz, Social epidemiology and the fundamental cause concept: On the structuring of effective cancer screens by socioeconomic status. Milbank Quarterly 76 (1998): 375-402. Whitehead, M. 1992. The concepts and principles of equity and health. International Journal of Health Services 22: 429-45. Wikler, D. 1978. Persuasion and coercion for health: Issues in government efforts to change lifestyle. Milbank Memorial Fund Quarterly: Health and Society 56(3): 303-38. Wilkinson, R. G. 1992. Income distribution and life expectancy. British Medical Journal 304: 165-8. Williams, A. (Andrew). 1995. The revisionist difference principle. Canadian Journal of Philosophy 25: 257-82.

8.6 Cases

Case 1

Health Care Workers and Flu Shots Vancouver Coastal Health (VCH) is the local health au-

thority for Vancouver, BC. It employs thousands of health care workers at many levels in many kinds of health care facilities.' They include doctors and nurses in hospitals and occupational therapists and assistants in long-term residential care facilities and everything in between. VCH employees see hundreds of thousands of patients every year with many different kinds of health conditions. Since 2012, VCH has required all of these health care workers to get flu vaccine shots or wear a protective mask at all times when in contact with patients and clients during winter flu seasons—at least if they want to keep their jobs.2 Why did VCH make this policy mandatory? VCH says that health care workers have a moral responsibility not to harm their patients and clients.2•3 This is a special moral duty health care workers have that is greater than the one ordinary people have not to harm others. If a health care worker at any level had the flu, they could easily infect a patient. Doing this would harm the client, and in some cases it could even lead to his or her death. VCH says the best way not to harm patients in this way is to get a preventative flu vaccine

shot or wear a protective mask. VCH explained that it had tried to persuade and encourage health care workers to be vaccinated or wear a mask, but this effort had been a failure.2•3.4 Before the rule became mandatory, many health care workers were already getting vaccinated, but many others at every level still weren't. The only alternative, VCH says, was to make flu vaccination or the wearing of an appropriate mask mandatory. Not everyone is happy about this policy. Alan Cassells, a respected journalist who writes about health matters, has said the scientific research does not prove that having all or a very high percentage of health care workers vaccinated against the flu actually results in fewer infections in patients.5 The BC Nurses Union, however, has said scientific evidence shows that nurses and other health care workers should choose to be vaccinated because this stops infections and so prevents harms to patients.° But the BC Nurses Union also believes that health care workers should not be required as a condition of continued employment to have the flu vaccine or wear a mask. Such a requirement wrongly violates an individual's moral right to decide for themselves whether the scientific evidence shows that they should either be vaccinated or wear a protective mask.3.6

Notes 1. Vancouver Coastal Health, 'Quick Facts," http://www.vch.ca/ about-us/quick-facts. 2. Vancouver Coastal Health, "Health Care Worker Influenza Control Program—Questions and answers," https://www.vch.ca/media/Staff-Flu-Shot-QA.pdf.

3.

4.

E Galankis et al., "Ethics of Mandatory Vaccination for Health Care Workers," Eum Surveillance 18 (2013), article 20627. Kumanan Wilson, "Case Discussion in Response to Mandatory Immunization of Local Health Authorities,"

471

8 Public Health

5.

in Population and Public Heali h Ethics (Toronto: University of Toronto Joint Centre for Bioethics, 2012), 118-21. Alan Cassells, "Will a Flu Shot Keep You Healthy?' FOCUSonline October 2012, http://wwwfocusonline.caficrnode/447.

Case 2

Students and Study-(Enhancing Drugs Because of intense competition for awards and admission to graduate programs, students at Canadian colleges and universities are using cognition-enhancing drugs like Adderall, Ritalin, Concerta, and Modafinil to increase their concentration during study sessions and help them complete assignments on time.' •2 An article in the Canadian Medical Association Journal in 2011 estimated that 8 to 35 per cent of Canadian university students are taking these kinds of drugs for non-medicinal purposes.3 Their use may even be more common among pre-med students.4 The drugs are prescribed to treat things like attention-deficit hyperactivity or the sleeping disorder narcolepsy. Possessing and using them without a prescription is illegal in Canada, but this hasn't stopped students not diagnosed with these issues from getting and using the drugs. According to a CTV News report from February 2013, these drugs can easily be purchased on websites like Craigslist and from sellers who visit campus libraries or from students who have a prescription.5 (Their prices go up at exam time!) The popularity of these drugs has raised a variety of questions. Do they really work in ways users hope? Modafinil has been demonstrated to improve memory by 10 per cent.° A third-year arts and economics student at McGill University who used Ritalin said, "You sit down and concentrate for eight hours or so without feeling the need for breaks or wanting to stop." But an assignment won't get done if, rather than helping you to concentrate on the assignment, using Aderall means you become focused on Facebook for hours.

6.

BC Nurses Union, "Position Statement Influenza Control Policy," https://www.bcnu.org/AboutBcnu/ Documents/position-statement-influenzacontrol.pdf.

And according to Cynthia Forlini, a research fellow at the University of Queensland in Australia, "There seems to kind of be a ceiling of enhancement." Students who are already doing well benefit less. There are also bad side-effects.'" Most users reported an unpleasant "come-down," and some said their sleep was disrupted. Experts say other effects include loss of appetite and dehydration. Adderall can cause a change in the rhythm of the heart and even a heart attack.'" In addition, some wonder if using these drugs is cheating."9 Students who use these drugs have a competitive advantage only when most other students aren't using them. If most or all students used them, then no one would have gained an edge, and everyone would be worse off because of the bad health side effects.m In 2011, Duke University in North Carolina made "the unauthorized use of prescription medication to enhance academic performance" an offence under its rules on academic dishonesty. n It's a form of cheating. '2 This was a policy supported by the student newspaper. In September 2016, its editors wrote that "Prescription drug abuse on campus should be treated as a public health . . . issue." 13 Britain's Academy of Medical Sciences compared the use of these drugs to steroid use in athletes.° It has called on universities there to consider a policy like Duke's. Many students disagree. It's not cheating "because it's still your work, . . . you're just doing it with an increased mental capacity," as a secondyear University of British Columbia student put it.2 According to UBC neuroscientist Peter Reiner, "There is a widely held perception that taking the drugs is [ethically] bad whereas coffee is OK. In principle, there's no real difference between the two, except all the evidence is that caffeine is a little more effective than Adderall."2

Notes 1.

2. 3.

Rhys Edwards, "This Is Your Brain o n Adderall," The Ubyssey 7 Apr 2013; Vassilena Sharlandjieva and Samantha McCabe, "The High in Higher Education," The Ubyssey 24 February 2016. James Bradshaw, "Students Reaching for ADHD Drugs to Deal with Academic Stress," The Globe and Mail 18 October 2013. Daniel Rosenfield et al., "Time to Address Stimulant Abuse on Our Campuses," Canadian Medical Association Journal 163 (6 September 2011), 1345.

4.

5. 6.

Robyn M. Emanuel et al., "Cognitive Enhancement Drug Use among Future Physicians," The Journal of General Internal Medicine 28 (2012), 1028-34. ovNews.ca Staff, "Canadian Students Abusing Adderall to Get Edge in Studying," 4 February 2013. R.M. Battleday and A.-K. Brem, "Modafinil for Cognitive Neuroenhancement in Healthy Non-SleepDeprived Subjects: A Systematic Review, European Neuropsychopharmacology 25 (November 2015), 1865-81.

Cases 1 473 7.

8.

9.

Cynthia Forlini, Serge Gauthier, and Eric Racine, "Should Physicians Prescribe Cognitive Enhancers to Healthy Individuals? Canadian Medical Association Journal 185 (3 September 2013), 1047-50. Man Lamkin, "A Ban on Brain-Boosting Drugs Is Not the Answer," The Chronicle of Higher Education 27 February 2011. Barbara Sahakian and Jamie Nicole LaBuzetta, "Professor's Little Helper—The Ethics of Enhanced Cognition," in Bad Moves (Oxford University Press, 2013), ch. 5.

Case 3

Breast Cancer Screening If you're a woman over 50 in Ontario, you will start to receive letters from Cancer Care Ontario (a provincial agency responsible for improving cancer services) encouraging you to have a breast cancer screening mammogram.' The letters will tell you that doing so is part of taking care of your health. They will continue to arrive until the agency learns that you have had a mammogram. Breast cancer is the most common cancer among Canadian women.2 About one in nine Canadian women will end up with it during their lifetime, and on average 14 Canadian women will die each day from breast cancer.2.3 A mammogram screening test is an x-ray that allows a doctor to inspect breast tissue for suspicious areas. It is not a test that determines the presence of cancer. Additional testing is needed to tell this. These tests will include a biopsy examination, meaning that some of the lump has to be removed for laboratory testing.3 The idea behind this screening test is to catch a cancer early in its history and especially before it has spread to the rest of the body. Once cancer has spread, it is often fatal. Does the screening test save lives? Starting in the 1970s and until a few years ago, it was believed obvious that mammogram screening saved lives for women of any age whether their family had a history

10. Veljko Dubljevid, "Cognitive Enhancement, Rational Choice and Justification," Neuroethics 6 (April 2013), 179-87. 11. Jenna Johnson, "College Administrators Worry That Use of Prescription Stimulants Is Increasing," Washington Post 27 November 2011. 12. Duke University, "Academic Dishonesty," haps:// studentaffairs.duke.edu/conduct/z-policies/ academic-dishonesty. 13. Editorial Board, "Duke and Dope," Duke Chronicle 22 September 2016.

of breast cancer or not.3.4.5 But a Canadian randomized control trial reported in 2014 demonstrated that there is no difference between those who had mammograms and those who didn't in the breast cancer death rate in women aged 40-49 without a family history of breast cancer.3,8 A summary of all the evidence in 2012 said that if 2000 women over 50 with no family history of breast cancer had breast cancer screening for 10 years or more, the life of one of these women would be saved by the screening who would otherwise have died.' The summary also pointed out that there were some downsides or harms to the screening.' Of those 2000 women who were screened, about 100 would have a lump or mass show up in the mammogram x-ray that turned out on a biopsy examination not to be a cancer. These are false positives. Ten women in the 2000 would have a cancer tumour that was treated with surgery and perhaps chemotherapy or radiation but unnecessarily. It would be a tumor that is growing so slowly that it would not have been a problem during their lifetime. These are overdiagnosis and overtreatment cases. But as science technology and society professor Alan Richardson points out, these facts and numbers by themselves won't decide that a breast cancer screening program is being oversold.8 According to Richardson, determining whether a screening program like the one in Ontario is worthwhile requires considering social and moral values.

Notes 1.

2.

3.

Cathy Risdon, Breast Cancer Screening: Making Sense of the Evidence, McMaster University Health Forum Presentation, 17 April 2015. Canadian Cancer Society, Breast Cancer in Canada 2016, http://www.cbcf.org/ontario/AboutBreastCancerMain/ FactsStats/Pages/Breast-Cancer-Canada.aspx. Jonathan Sussman, Breast Cancer Screening: Making Sense of the Evidence, McMaster University Health Forum Presentation, 17 April 2015.

4. 5.

6.

Siddhartha Mukherjee, The Emperor of All Maladies: A Biography of Cancer (New York: Scribner, 2010), p. 395. Canadian Task Force on Preventive Health Care, "Recommendations on Screening for Breast Cancer in Average-Risk Women Aged 40-74 Years," Canadian Medical Association Journal 183 (November 2011), 1991-2001. Anthony B. Miller et al., "Twenty Five Year Follow-up for Breast Cancer Incidence and Mortality of the Canadian National Breast Screening Study: Randomised Screening Trial," British Medical Journal 348 (February 2014), 348-58.

II

uw I

All

111111

8 Public Health 7.

P.C. Gotzsche and K.J. Jorgenson, "Screening for Breast Cancer," Cochrane Database of Systematic Reviews 2013, issue 6, article no. CD001877.

Case 4

Water in First Nations Communities Most Canadians expect that they will be able to get water from their kitchen and bathroom taps and that it will be safe to drink, wash food with, and bathe and shower in. For the most part this happens.' But in some parts of Canada—in particular, on First Nations reserves—water isn't so easily available or, if it is available, it isn't safe to drink or bathe in.'•2'3.4 While some other Canadians sometimes have the same kinds of problems with the availability or safety of their water, the rate at which these issues happen in First Nations communities is disproportionately greater.25 According to environmental justice lawyer Kaitlyn Mitchell, this difference "contributes to the perpetuation of social, political, and legal disadvantage" for people who live in these communities.5 It is, she says, "an ongoing human rights violation that serves to compound current and historic injustices First Nations face." Some Indigenous communities, for example, are without any running drinking water. They rely on delivery of water by truck, sometimes in big heavy jugs. Some rely on untreated water from contaminated wells or lakes. But even when they receive water treated in water plants, it can still be unsafe. The water can cause gastrointestinal illnesses and increase the risk of cancer if drunk or used to wash food. If people bathe or shower in it, skin infections like eczema and psoriasis can develop or worsen.'•5 By August 2016, for example, there were 15.3 drinking water safety warnings in 114 First Nations reserves across Canada.4

8.

Alan Richardson, "The Breast Cancer Screening's Debate Is about Values and Resources," The Globe and Mail 29 November 2011.

According to a 2016 investigation by The Globe and Mail newspaper, one third of Indigenous people living on reserves—just over 150,000 people—used drinking water from treatment systems that threatened their health.3 An earlier investigation by CBC News said that two-thirds of First Nations communities had at least one drinking water advisory from Health Canada between 2004 and 2014.2 For non-Indigenous communities, centralized water systems are governed by local town or city or municipal authorities.3 These authorities can find the financial resources needed to construct and adequately maintain and staff water systems through local taxes or utility fees.4 First Nations water systems, however, are mostly the responsibility of the Indigenous and Northern Affairs Department of the federal government.3.4 According to a recent report by the David Suzuki Foundation, while federal money is often there for the initial construction of water treatment facilities, the problem is that funding resources are not enough for the continued proper maintenance of these facilities and the training and salaries of the staff that run them.4 One of the department's current goals is to improve the availability and safety of fresh water in these communities. Recent federal budgets have promised to devote more money to the issue.4•' But it has also been decided that the best way to go about achieving the goal is to allocate whatever financial resources there are—whether they are increased or not—to improving the worst centralized water systems in Indigenous communities.3,6 For example, reserves that have had boil-water health notifications for a long time will have greater priority over those where the water issues are less significant.

Notes 1. 2.

3. 4.

5.

News, "Clean Running Water Still a Luxury o n Many Native Reserves," 12 April 20:L4. CBC News, "Bad Water: 'Third World' Conditions on First Nations in Canada," 15 Octoter 2015. Matthew McCleam, "Water Systems at Risk," The Globe and Mail 26 August 2016. David Suzuki Foundation and the Council of Canadians, Glass Half Empty? David Suzuki Foundation, February 2017; David Suzuki, "Government Must Do More to Address First Nations' Water Woes," The Georgia Straight 14 February 2017. Kaitlyn Mitchell, "Drinking Water Crisis in First Nations Communities Violates Human Rights" Ecojustice 23 February 2017. CBC

6.

7.

Indigenous and Northern Affairs Canada, "Department's Response to the National Assessment of First Nations Water and Wastewater Systems," https://www.aadnc-aandc.gc.ca/eng/1314383237592/ 1314383364994. Water Canada, "Federal Budget 2016 Promises Dedicated Funds for Water Infrastructure, Stewardship, and Science," http://watercanada.net/2016/canadas-budget2016-promises-dedicated-funds-for-water-infrastructurestewardship-and-science.

Study Questions

8.7 Study Questions 1. What is the difference between public health and clinical health topics? Consider some examples like abortion services. Do these belong to public health or clinical health? 2. What are collective action problems in health care? What is the difference between someone being a free rider and someone being an overconsumer? Is it always morally wrong to be a free rider or an overconsumer? 3. One writer in an article not included here has suggested that public health interventions are just those that deal with collective action problems in health (Johnny Anomaly, "Public health ethics and public goods," Public Health Ethics vol 4 (2011), 251-259). This is certainly a different definition than the one provided in the introduction to this chapter . Is it a better definition? 4. What is the difference between a screening test and a diagnostic test for an illness? Could one test be both a screening and a diagnostic test? 5. What is the difference between a false positive result from a screening test and an overdiagnosis result? Will overdiagnosis always lead to overtreatment? 6. When several different screening tests are available for an illness, they usually differ in their false positive and their false negative rates. Sometimes both are worse on one test, but often one is worse and the other better. What's worse in a population screening test: a higher false positive rate or a higher false negative rate? 7. Are there moral problems with harm reduction programs? What are they? Are they ones that mean that these kinds of programs should not exist? 8. Are all average differences in health between population groups unjust or unfair? What makes them unjust or unfair? What about differences in health between individuals? Are these always unjust or unfair? Are the differences between population groups unjust or unfair only when those differences between the individuals in one group and the other are unjust or unfair?

8.8 Suggested Further Reading Armstrong, Pat, and Ann Pederson, eds. 2015. Women's Health, 2nd edn. Toronto: Women's Press. Bernheim, Ruth, James Childress, Richard Bonnie, and Alan Melnick. 2015. Essentials of Public Health Ethics. Burlington, MA: Jones & Bartlett Learning. Bognar, Greg, and Iwao Hirose. 2014. The Ethics of Health Care Rationing. New York: Routledge. Dawson, Angus, ed. 2011. Public Health Ethics: Key Concepts and Issues in Policy and Practice. Cambridge: Cambridge University Press. and Marcel Verweij, eds. 2009. Ethics, Prevention, and Public Health. Oxford: Oxford University Press. Greenwood, Margo, Saraj de Leeuw, Nicole Lindsay, Charlotte Reading, eds. 2015. Determinants of Indigenous Peoples' Health in Canada. Toronto: Canadian Scholars' Press. Holland, Stephen. 2014. Public Health Ethics, 2nd edn. Cambridge: Polity Press. Holmes, Morgan, Linda Mooney, David Knox, and Caroline Schact. 2013. Understanding Social Problems, 5th Canadian edn. Toronto: Nelson Education. Mankiw, Gregory. 2009. "Public Goods and Common Resources," ch. 11 of Microeconomics, 5th edn. Mason, OH: South-Western Cengage Learning. Mikkonen, Juha, and Dennis Raphael. 2010. Social Determinants of Health: The Canadian Facts. Toronto: York University School of Health Policy and Management. Peckham, Stephen, and Alison Hann, eds. 2009. Public Health Ethics and Practice. Bristol: Policy Press. Rossiter, Kate, and Kate Reeve. 2004. The Last Straw! A Board Game on the Social Determinants of Health. Toronto: Brunswick Books.

l'iessavch vitih Humans 9.1 Introduction Health research is essential to advancing clinical patient care and human well-being generally. It is also needed to test "common-sense" and accepted medical practice, which can be in error, including in cases of invasive or risky treatments. For example, landmark studies from the 1980s found—contrary to widely held medical belief—that mastectomy is not supericr to lumpectomy in treating some breast cancers.' Similarly, more recent health research has been essential in establishing that arthroscopic knee surgery in osteoarthritis patients is not preferable to a combination of non-invasive physical and medical therapy2 But while scientific evaluation of new and current treatments is vital to advancing human health and well-being, it poses distinct, sometimes more complex, ethical challenges from those encountered in day-to-day patient care. Because of the distinctive nature of health research, we begin with an account of the differences between health research and health care to make sure those terms are understood at the outset. We then review the history of research scandals involving human research participants (sometimes referred to as "the dark history" of research) and describe regulatory responses in Canada and elsewhere. These preliminaries sketch out a range of ethical issues and problems raised by health research. Discussion of the main readings addressing general and specific ethical issues in health research is set against this background.

The Challenges of Conducting Ethical Health Research Therapeutic Misconception

therapeutic misconception the failure to understand that health research is aimed primarily at producing knowledge and is not conducted specifically for the sake of research participants' health.

Understanding the distinct ethical challenges posed by health research begins with an appreciation of the ways in which health research differs from health care and how they are sometimes confused. Health research, like clinical care, aims to improve human health. It is understandable then that the testing of new therapies and collection of observational health data may often be thought by research participants as intended to make a direct difference to their care. This is often referred to as the therapeutic misconception. This error has conceptual and empirical aspects and leads to underestimating risks and overestimating the prospects of direct personal benefits of research participation. The conceptual therapeutic misconception consists of not recognizing that the specific purpose of health research is to advance knowledge. Thus, unlike every treatment in a clinical setting, research testing of new drugs, medical devices, new ways of managing illness, the collection of health care data, and so on is not done for the sake of the individuals being studied but rather to advance knowledge of human illness and injury and therapeutic

The Challenges of Conducting Ethical Health Research

and other means for restoring and fostering human health. This knowledge is intended foremost to help future patients. Often, there may be no, or no reasonable, expectation that participants in a research study will ever benefit directly. The empirical therapeutic misconception consists in failing to recognize the often remote chance of actually receiving any direct therapeutic benefit from an experimental therapy or collection of personal health information for research purposes. Recent reviews of research involving investigational (that is, new) drugs report that only about 10 per cent are ever licensed as therapies.3 A large majority of research participants in clinical studies of new drugs is therefore exposed to research-related risks and never benefit directly from those experimental therapies. Even the 10 per cent figure is likely to mislead. Only a much smaller overall percentage of research participants is likely to receive any direct benefit from experimental drugs that are later licensed. Some phases of experimental research do not look principally, or at all, for therapeutic benefits; sometimes research is about a range of dosages for drugs, not all of which are likely to be both safe and effective. Further, in most experimental research there is normally up to a 50 per cent prospect of being randomized into a study arm where research participants will receive no new treatment at all—for example, they receive a placebo or a treatment available outside the study against which the experimental treatment is compared (referred to as "the control arm"). Research participants are not the only individuals who can fall prey to therapeutic misconception. Researchers can, too. This is partly because many of them are also health care providers and may see research as an extension of clinical care. We shall also see below that the therapeutic misconception may also be subtly present in how researchers conceptualize the design of studies as meeting duties of care to research participants.

study arm participants in clinical trials are randomly assigned to different groups (arms) receiving different interventions to compare the effectiveness and safety of those interventions.

Informed Consent We have seen that it is not the primary aim of health research to provide therapeutic benefits to research participants and that this will not be statistically likely in most cases. Failure to recognize this undermines a research participant's understanding of a study and their role in it and so is inconsistent with informed consent to participate in research. An obvious solution is to recognize that these misconceptions are based on informational deficiencies that should be addressed as part of the process of obtaining informed consent from potential research participants. But addressing therapeutic misconceptions is by no means the only special challenge of obtaining informed consent in health research. Indeed, health care research poses other burdens on informed consent that may be even more difficult to address. To make informed choices, research participants have to understand clearly the differences between research and clinical care. They also have to be informed about their clinical options, how the research differs from those options, and what the related potential risks and benefits of each are. Since less is typically known about the risks and benefits of experimental interventions, potential participants often have to make decisions under a greater burden of uncertainty about potential outcomes than in clinical care. They will also need to understand the research methodology in order to know what will happen to them during the research. This is often no small matter. Trying to understand what is involved in a randomized, double-blind, placebo-controlled trial and how a participant will fit into this methodology can be difficult for many. Potential research participants also need to be informed about research-related tests, which can be extensive, what their risks are, and how they differ from tests that are part of standard care. They should also be informed of special issues around confidentiality. In clinical care, access to patient information is limited to a patient's physician and some members of the patient's circle of care, whereas research data may be accessed by sponsors, government regulatory bodies, research ethics boards, and sometimes other researchers. Sometimes research participants' data, biological

randomized, double-blind, placebo-controlled trial a study in which research participants are randomly assigned to either an experimental treatment or a placebo and neither they nor the researchers know what they are receiving.

II ill

II IGLI

9 Research with Humans

Hawthorne Effect the alteration of research participants' behaviour as a result of knowing that they are being observed by researchers, typically by improving their behaviour.

conflicts of interest situations in which a person (for example, a researcher) has an obligation to a person or an institution that is in tension with another interest or obligation and could corrupt the decision-making of that person (that is, the researcher).

observational research attempts to advance knowledge by collecting data from research participants without making any change to their care, treatment, or lives in relation to what is being studied.

experimental (or interventional) research attempts to advance knowledge by studying the effects of research interventions on research participants.

samples, and personal information must be stored or analyzed in other countries where there are different standards for protecting confidentiality. The requirements of informed consent in health care research are often more extensive and demanding of research participants than those for patients receiving clinical care. There is also the further problem that informed consent cannot be obtained for some types of research. For example, research in emergencies must often be done without consent because participants may lack capacity and there may be no time to obtain consent from substitute decision-makers. In other cases, the consent process may undermine research for other reasons. For example, disclosing to nurses that they are being observed to determine how well they implement a new or experimental protocol for, say, triaging patients in emergency rooms may bias study results. If the nurses knew they were being observed, they could alter their behaviour to better comply with a new protocol, and so an accurate picture of its workability and usefulness would not emerge. (This is sometimes referred to as the Hawthorne Effect.) Sometimes obtaining consent is impractical for other reasons. For example, obtaining consent for collecting clinical chart data accumulated over many years may be impractical because of difficulties in contacting patients or because consent may introduce bias into the research because only healthy or living patients will respond.

Conflicts of Interest Potential participants should also be aware that the aims of research and health care can conflict and that the research enterprise itself contains a variety of conflicts of interest. They need to be informed about those conflicts and how they will be managed. Suppose researchers are trying to answer a scientific question about the safety and effectiveness of a therapeutic intervention. If a participant is not doing well in a study, he or she may wish to withdraw. But that will sometimes limit information about side effects, a clearer picture of which may emerge if the participant remains in the study. There is thus a potential conflict between the scientific needs of a study and the needs of a participant. The conflict can become more complicated if the researcher is also the participant's caregiver (e.g., personal physician or nurse), since the researcher will have conflicting roles and duties. Sometimes researchers also have financial or career interests in the outcomes of studies. Special processes are needed, then, to manage conflicts of interest. One way to address them, at least partially, is to disclose them to potential participants and emphasize the voluntariness of research participation, including that participants can withdraw at any time without having to give reasons. As well, researchers should often play a limited role in recruiting participants and obtaining informed consent, particularly when researchers are also caregivers. Researchers with financial interests in the interventions being studied should play a limited, if any, part in the collection and analysis of the data. Designing Ethical Studies Recognizing and responding to therapeutic misconception, the special challenges of seeking informed consent, and conflicts of interest do not nearly exhaust the ethical problems involved in. research. Arguably, the overarching ethical issue in clinical research is how to design experimental and observational studies to advance knowledge of human health and at the same time respect the rights and well-being of research participants. This is the main conflict of values in health research. How it is resolved defines the ethical character of the research. A main challenge for designing ethical research is for researchers to create scientific studies that appropriately minimize risks to participants—that is, valid research that obtains reliable, useful results and at the same time observes principles of non-maleficence and harm prevention. For observational research, the main ethical requirement is to protect the confidentiality of the research data that is collected. Experimental research must

The Challenges of Conducting Ethical Health Research

protect confidentiality but faces a further range of ethical problems related to minimizing risks of research interventions. An intuitively attractive solution for experimental studies is to accept research design as ethical where there is clinical equipoise between the study arms. Clinical equipoise means that an experiment on humans is only justified if there is genuine uncertainty among the relevant expert community as to which arm in a study is better.4 Clinical equipoise has seemed to many to answer important questions about the ethical design of experimental health research. In particular, it seems to meet a duty of care researchers have to participants. If experts cannot say which arm of a study is superior, then administering the treatments does not involve researchers knowingly exposing anyone to unnecessary harm. It also implies that there can be an acknowledged prospect of a direct benefit in the experimental arm. This appears to provide an important rule, then, for when a multi-arm study design can be ethically justified. It also appears to explain when a study must stop, since if evidence emerges during a study that shows one arm to be clearly inferior to another, clinical equipoise has been disturbed, and the study must stop in order not to expose some participants to a known inferior intervention. The problem with clinical equipoise, however, is that it is extremely difficult to know when it exists, particularly when untested treatments are compared against standard treatments with known effectiveness. Moreover, there are many accepted examples of experimental research that simply do not seem to meet standards of clinical equipoise (discussed below). And a general, often unacknowledged issue is how to determine when there is genuine uncertainty within the relevant expert community as to which arm in a proposed study is better. It will not be when all researchers are indifferent about competing arms, since researchers often strongly believe one arm to be superior to others. Must a majority of relevant experts be indifferent, then? Or is the idea of indifference irrelevant and to be replaced by the idea of respectable disagreement among relevant experts? If so, what constitutes respectable disagreement, and when is it overturned? One of the ethical principles that underlies clinical equipoise, that of not exposing research participants to known unnecessary risks, has further important implications for the design of ethically sound clinical research. It can apparently provide ethical guidance about when use of placebos and standard treatments as controls is justified. It implies that if there is a standard treatment, it would be ethically wrong to use a placebo control, since that would knowingly expose some research participants to a known inferior therapy, thus violating a duty of care. This principle also explains why health care research must satisfy high standards for scientific integrity, particularly for research that exposes participants to more than minimal (or everyday) risk. If research lacks scientific value, its results will not help to answer questions related to health care, and the participants will have been exposed to risk without any prospect that the research will contribute to knowledge about human health. Bad science is always unethical in health research. It wastes scarce research resources, wastes the time and abuses the trust of participants (who mistakenly think they are contributing to science), and exposes participants to risk without any prospect that the research will contribute anything of scientific or social value.

The Dark History of Research An International Problem The need to take account o f the special ethical requirements of conducting health research was brought to a head after the discovery of Nazi medical research atrocities during the Second World War. The subsequent trial and conviction of many Nazi doctors for war crimes led to the creation of the first internationally recognized code of research ethics, the Nuremberg Code (later replaced by the World Medical Association's Declaration of Helsinki).' The Nazis were not alone in participating in ethical violations in health

clinical equipoise the state of genuine uncertainty within the expert community about which study arm is therapeutically superior in a multi-arm study of an experimental or existing therapy.

iI

JII

1111

9 Research with Humans

research. The Japanese conducted similar experiments in Manchuria during the Second World War.° Developed modern democracies also engaged in unethical medical experiments before, during, and after the Second World War, including after the Nuremberg Code and Declaration of Helsinki were adopted. The most famous of these research scandals is probably the Tuskegee Syphilis Study, which observed the natural progress of syphilis in a population of poor, mostly illiterate, black agricultural workers in Alabama between 1932 and 1972.7 In order to conduct the study, participants were not told they had syphilis and were denied access to penicillin once it became a recognized effective treatment for the disease. They were enticed to participate with promises of free medical care, meals, and burial insurance. The Tuskegee study was not particularly exceptional. In 1966, American researcher Henry K. Beecher reported 22 examples of research that failed to meet adequate ethical standards, often involving vulnerable individuals, including children and the mentally ill.8 Neither the Tuskegee study nor the studies Beecher described were hidden from public view. They were known to medical researchers, and their results were often published in leading medical journals. Further evidence of US research scandals emerged through the 1960s. In 1972, public outrage as a result of news reports finally stopped the Tuskegee study. Government hearings that were held into the scandal led to The Belmont Report,' the first systematic attempt to summarize ethical principles and provide guidance for human research. The Belmont Report led to adoption in 1981 of a set of US federal regulations now in place for conducting clinical trials known as "the Common Rule."10 The Common Rule was then, and arguably remains today, the leading and most comprehensive set of research ethics regulations. Canada, too, has its own dark history of medical experimentation, part of which is reviewed in this chapter. Colin A. Ross's article "Dr Ewen Cameron" (see page 485) describes how a prominent Canadian psychiatrist engaged in "unethical, unscientific, and inhumane" brainwashing and later mind-control research in Canadian mental health institutions from the 1930s to the 1960s, including at McGill University. Dr Cameron was not alone among Canadian doctors in conducting this research, much of which was also published in medical journals. Some of the research was funded by the US military and Central Intelligence Agency. More recently, historians have begun to report controlled nutritional and other experiments undertaken without consent in First Nations communities and residential schools."

The Canadian Response Research scandals have not been as prominently discussed in Canada as in the United States, and they may not have been as extensive. That may be due more to lack of funding and opportunity than to any greater ethical sensitivity on the part of Canadian physicians and researchers. However, it is very unlikely that these types of research could be conducted today in Canada. Although there is no specific Canadian law governing health research in Canada, there is, in effect, a set of national standards established in The Di-Council Policy Statement: Ethical Research Involving Humans (second edition, usually referred to as "TcPS2").'2 TCPS2 establishes ethical rules and guidelines for regulating research along with processes for ethical review of research by independent "research ethics boards." The Tri-Council is a federally supported group of research funding councils that includes the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council, and the Social Sciences and Humanities Research Council. TCPS2 operates much as a de facto Common Rule in Canada in that to be eligible for Tri-Council funding, Canadian universities (including hospitals with university affiliations) and colleges are required to adhere to TCPS2 in all their human research, whether or not it is Tri-Council-funded. This encompasses most health research that is conducted in Canada. (Students who are interested in receiving a certificate of accreditation for completing a course on research ethics can take the TCPS2 online course. Links to TCPS2 and the tutorial are in Suggested Further Reading at the end of this

The Dark History of Research 481

chapter.) TCPS2 is notably unique among research ethics guidelines and codes in that it combines statements of standards with extensive discussion of their justification and meaning as a way of assisting interpretation and application of the standards to specific research contexts. Often, research ethics regulations from other countries have to be taken into account as well. Research originating in other countries must also often satisfy those countries' ethical requirements. For example, US research that is conducted abroad, including in Canada, must nevertheless satisfy the standards of the Common Rule. Physician researchers are also expected to observe the standards in the World Medical Association's Declaration of Helsinki in virtue of Canada's membership in the WMA. Often, these different codes and regulations do not fully agree, posing further challenges for ethical review and compliance. Problems Still Not Over

Despite the more careful regulation now of the ethical conduct of research studies, ethical scandals still arise from time to time. In "Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis" (see page 489), Canadian bioethicist Arthur Schafer highlights how commercial pharmaceutical funding of research can present financial conflicts of interest that can interfere with the ethical treatment of human research participants. The leading recent example of this is the case of Dr Nancy Olivieri, who was testing an experimental drug developed by the pharmaceutical company Apotex at the Toronto Hospital for Sick Kids. Olivieri found evidence in 1996 that the treatment could cause harm and regarded it as her duty to inform participants of the potential risks. Her contract with Apotex, however, stated that she could only release information about her findings with Apotex's consent, which was refused. Dr Olivieri disclosed the information anyway. Instead of being supported by the hospital, she was fired from her position as head of the Hemoglobin Research Program and was the general subject of a smear campaign. Olivieri argued that she had obligations to disclose this information to her participants, the hospital research ethics board, and other researchers. For a long time, she received no effective support from the University of Toronto where she held a related appointment. Commentators generally view this case as a reflection of the power large pharmaceutical companies have to pressure hospitals and universities to tolerate unethical conduct through their prerogative to withdraw research funding. Olivieri was later vindicated in a variety of reviews of the case. The obligation to disclose to research participants evidence of harm without the consent of drug sponsors is now accepted. Nevertheless, many commentators are still doubtful that issues regarding industry-sponsored research have been fully addressed."

Designing Ethical Research Studies In "An Ethical Framework for Biomedical Research" (see page 495), Ezekiel J. Emanuel, David Wendler, and Christine Grady identify eight ethical principles of health care research that they claim "delineate a systematic and comprehensive way of assessing the ethics of particular clinical research'" These principles are an attempt to provide a general ethical framework to address the challenges just reviewed and to make sure that the dark history of research is not repeated. The eight principles are: (1) The research must be a collaborative partnership between researchers and the communities in which the research is done. (2) The research must have social value. (3) The research must have scientific validity without denying generally available health services or requiring access to health services that are not practically available. (4) The selection of research participants must be fair. (5) There must be

9 Research with Humans

a favourable risk-benefit ratio. (6) The research must receive independent ethical review. (7) Informed consent must be appropriately obtained. (8) Research participants must be treated with respect throughout their participation in research and after they have finished their participation. Emanuel et al.'s discussion is an excellent starting place for thinking about the ethical requirements of health research. But while their eight principles recognize and address many of the ethical challenges facing research mentioned earlier, putting them into effect is frequently not straightforward. For example, it can be difficult to determine whether research will contribute to improvements in health care, whether the science is valid, whether important populations of interest will be excluded, what counts as an appropriate collaborative partnership between researchers and the communities they are investigating, what determines an appropriate balance of risks and benefits, whether ethical review has been independent and sufficient, whether the elements of informed consent have been properly fulfilled or waived on appropriate grounds, and what measures should be taken to respect participants after their involvement in research has finished. Some of Emanuel et al.'s discussed requirements are controversial. It might be asked whether requiring fulfillment of all these standards in the detail provided can be too demanding and unduly discourage or prevent needed health research. Also, Emanuel et al. do not directly address important controversies about designing ethical research. For example, there is no mention of recognizing the role of clinical equipoise in designing ethical research. In this respect, they may have been influenced by Franklin G. Miller and Howard Brody who have provided an acute challenge to the role of clinical equipoise in health research in "A Critique of Clinical Equipoise: Therapeutic Misconception in the Ethics of Clinical Trials" (see page 509) and critiqued an earlier defense of clinical equipoise by Emanuel et al. Miller and Brody point out that the recent history of clinical research has been mistakenly dominated by what they call "the similarity position"— namely, that the ethics of clinical equipoise requires upholding duties of care similar to those of clinical care, or generally putting care of research participants ahead of research considerations where they conflict. Miller and Brody argue in response that the ethics of research is fundamentally different from clinical care and that the notion of clinical equipoise should be abandoned. They say their position is reflected by research ethics boards' and regulators' recognition of the methodological importance of sometimes having placebo-controlled studies even when standard therapies are available for use as controls. They also point to the extra-therapeutic risks involved in research-related tests and other research procedures as being inconsistent with the similarity position, including invasive tests like lumbar punctures and "wash out periods" (when active treatments for the condition under study are discontinued to establish a non-treatment baseline). Miller and Brody argue that failure to recognize the fundamental differences between research and clinical care is in effect an extension of the therapeutic misconception mentioned earlier. In this context, however, it is the researchers who accept the idea of equipoise who have a misconception that research requires no known risks over and above those of standard clinical care. Miller and Brody's alternative is that scientific research should be driven by considerations of what methodologies will best contribute to scientific knowledge, unless they pose unacceptable levels of risk to participants One important practical shortcoming of their discussion is that they do not say what unacceptable levels of risk are. If researchers do not owe research participants a duty of care that is comparable to clinical duties of care, what can participants reasonably be asked to consent to? It is important to ask, however, whether questions about acceptable levels of risk raise objections to their view or whether they highlight another difficult area of research ethics deliberation, supporting, in effect, robust ethical review of research via independent research ethics boards. If Miller and Brody are right that the notion of clinical equipoise is inappropriate and often rightly ignored, a reconsideration of the nature and extent of a duty of care to research

Designing Ethical Research Studies

participants will be necessary to clarify the boundaries of ethically acceptable research. For many ethicists, Miller and Brody's article poses a strong challenge to conventional wisdom regarding the ethics of clinical research. It is fair to say, however, that their arguments are controversial and have not yet significantly influenced official attitudes toward clinical studies, which retain today a strong similarity position in research ethics codes and regulations. The use of clinical equipoise as a principle for guiding clinical studies is retained, for example, in Canadian research ethics guidelines in TCPS2. Obviously, scientifically and ethically well-designed studies as envisaged by Emanuel et al. are meaningless if research is not reported with integrity. In "Repairing Research Integrity" (see page 523), Sandra L. Titus, James A. Wells, and Lawrence J. Rhoades discuss the prevalence of researcher misconduct involving misrepresentation of research results and what measures can be taken to combat it. See also Case 1 in this chapter for a famous Canadian example of researcher misconduct (see page 547).

Research on Special Populations One important issue addressed b y Emanuel et al. that is now beginning to find its way into research ethics guidelines is to what extent interests of communities should be taken into account in clinical research. This has been a pressing issue in Canada because of the practical realities and legal status of First Nations peoples. They have interests in knowing that research will serve community interests, that they will not be exploited by researchers as they have been in the past, and that dissemination of research results will not promote discrimination against their members. Researchers, by contrast, are often interested in studying particular communities because they afford useful opportunities to understand the etiology of disease, especially genetic determinants, by looking at groups where prevalence of certain diseases is known to be high. The article by Charles Weijer and Ezekiel J. Emanuel, "Protecting Communities in Biomedical Research" (see page 518), is an attempt to develop a consultative framework for resolving these often competing interests. Weijer and Emanuel's ideas are broadly reflected in Chapter 9 of TCPS2, which is addressed primarily to research involving First Nations, Inuit, and Metis communities. Weijer and Emanuel are clear, however, that respect for communities should be extended further. This reflects the position taken by Emanuel et al., discussed earlier. In many respects, these authors' proposals are now frequently reflected in work researchers do with certain communities that share illness or disease, for example, HIv patients or drug addicts. However, consultations with such groups pose different sorts of challenges because of difficulties defining what constitutes a community and the frequent absence of geographically defined communities with identifiable leadership or political institutions. The chapter ends with two articles that discuss the challenges of doing research on vulnerable persons and populations. There is general consensus that the vulnerable have claims to special protection in clinical research, though there has been no agreement in research ethics guidelines as to what this means and, therefore, who it covers. The article by Samia A. Hurst, "Vulnerability in Research and Health Care; Describing the Elephant in the Room?" (see page 526), identifies the limitations of efforts to clarify what makes a person or population vulnerable. Hurst argues that vulnerability consists of "an increased likelihood of incurring additional or greater wrong"15 (that is, moral mistreatment) and that this definition should guide the application of research ethics guidelines. The definition can also inform the conduct of clinical care involving vulnerable persons or populations and so is relevant for other chapters in this textbook. Hurst's article can be usefully compared with the discussions of vulnerability in TCPS2.16 In "Best Practices for Health Research Involving Children and Adolescents" (see page 537), D. Avard et al. discuss ethical treatment of young persons in clinical research. These best

483

II Wu

IIILL.I

9 Research with Humans

a right that incompetent persons can have to accept or refuse treatment or participation in research. It is to be contrasted with a competent person's right to consent. assent

practices were intended to assist with the application of TCPS2 to pediatric research and are particularly notable for their presentation of the notion of assent. Research ethics guidelines now universally recognize that incompetent patients, including minors, have certain rights to refuse to participate in research even if substitute decision-makers—for example parents—are prepared to consent on their behalf. A right to assent/dissent is thr.s to be contrasted with rights to consent. But as Avard et al. acknowledge, the extent of a child's right to overrule his or her parent is controversial, and there is no accepted international standard in this respect. TCPS2 takes an unusually strong position, however, which is also reflected in Avard et al.'s discussion: namely, if the child lacks capacity for consent but has some ability to understand the research and dissents to participation, that decision must be respected despite a substitute decision-maker's willingness to consent on the child's behalf.

Notes 1. See Case 1 in this chapter. 2. Carly Weeks, 2008, "Athroscopic Knee Surgery Ineffective: Study," The Globe and Mail 11 September, ht:piJwwwtheglobeandmail.com/life/arthroscopic-knee-surgery-ineffective-study/article 18 137799. 3. Biotechnology Innovation Organization, "Clinical Development Success Rates 2006-2015" (2016), hups://www.bio.org/sites/default/files/Clinical%20Development%20Success%20Rates%2020062015%20-%20BIO,%20Biomedtracker,%20Amplion%202016.pdf. 4. Benjamin Freedman, "Equipoise and the Ethics of Clinical Research," New England Journal of Me dicine 317 (1987), 141-5. 5. The Nuremberg Military Tribunal, The Nuremberg Code, from US v. Karl Brandt et al. (1947), https:// history.nih.gov/research/downloads/nuremberg.pdf; World Medical Association, 2013, Declaration of Helsinki (2013), http://www.wma.net/en/30publications/lOpolicies/b3. 6. Sheldon H. Harris, Factories of Death: Japanese Biological Warfare 1932-45 and the American Cover Up (New York: Routledge, 1994). 7. Gregory E. Pence, "The Tuskegee Study," in Classic Cases in Medical Ethics, 2nd edn (New York: McGraw Hill, 1995). 8. Henry K. Beecher, "Ethics and Clinical Research," New England Journal of Medicine 274 (1966), 1354-60.

9. Naiional Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, Department of Health, Education and Welfare, The Belmont Report: Ethical Principles and Guidelines for 'he Protection of Human Subjects of Research (Washington, DC: United States Government Printing Office, (1978). 10. US Department of Health and Human Services, National Institutes of Health, and Office of Human Research Protections (revised), The Common Rule, Title 45 (Public Welfare), Code of Federal Regulations, Pan: 46 (Protection of Human Subjects), (2009). 11. Ian Mosbey, "Administering Colonial Science: Nutrition Research and Human Biomedical Experimentation in Aboriginal Communities and Residential Schools, 1942-1952." Social Histoiy 46 (2013) no. 91,145-72. 12. Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, and Social Sciences and Humanities Research Council, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2014), referred to as `TCPS2". 13. Jocelyn Downie, "The Power of Money: Commercialisation of Research Conducted in Public Institutions," University of Otago Law Review 11 (2006) no. 2,305-25. 14. Ezekiel J. Emanuel, David Wendler, and Christine Grady, "An Ethical Framework for Biomedical Research," in The Oxford Textbook of Clinical Research Ethics, edited by Ezekiel, J. Emanuel, Christine C Grady, Robert A. Crouch, Reider K. Lie, Franklin G. Miller, and David D. Wendler (New York: Oxford University Press, 2011) p.131. 15. Sam: a A. Hurst, "Vulnerability in Research and Health Care', Describing the Elephant in the Room?" Bioethics, (2008) 22(4): p. 195. 16. TCPS2, page 8 and 54.

Ross: Dr Ewen Cameron

9.2 The Dark History of Human Research in Canada

Dr Ewen Cameron Colin A. Ross Born in Bridge of Allan, Scotland on 24 December 1901, [Ewen] Cameron immigrated to Canada in 1929 to take a job as a psychiatrist at Brandon Mental Hospital in Brandon, Manitoba. He was recruited by Dr Thomas Pincock; one of the buildings in the Department of Psychiatry at the University of Manitoba in Winnipeg in the 1980s was the Pincock Building. Dr George Sisler and Dr John Matas, both of whom referred patients to Dr Cameron when he was at the Allan Memorial Institute in Montreal, taught at the University of Manitoba into the 1980s, as did Dr Gordon Lambert, who treated one of Dr Cameron's mind control victims on her return to Winnipeg. Despite these historical connections, I heard no conversation about Dr Ewen Cameron or CIA mind control while a resident and then a staff psychiatrist in the Department of Psychiatry in Winnipeg from 1981 to 1991, despite the fact that plaintiffs, including Val Orlikow from Winnipeg, settled a suit with the CIA in 1988. Mrs Orlikow's husband, David Orlikow, had been a prominent Member of Parliament from Winnipeg for many years. There was silence in psychiatry about CIA mind control, but no conspiracy of silence. No one was told to be quiet. From the perspective of academic psychiatry, mind control experimentation didn't exist, so there was no need to cover it up. Throughout the twentieth century, academic psychiatry provided no public commentary, ethical guidance, peer review, or moral oversight of any kind concerning mind control experimentation, despite the fact that the leading psychiatrists and medical schools were well funded by the CIA and military for mind control research. Mental patients, cancer patients, prisoners, and unwitting citizens were experimented on by mind control doctors at Yale, Harvard, McGill, Stanford, UCLA, and the other major universities. These human guinea pigs were never told that they were subjects in military and CIA mind control experiments, and they never gave informed consent. They received no systematic follow-up to document the harm done to them. The welfare of the "human subjects"

was not a relevant variable in the academic equation. What counted for the psychiatrists, I think, was money, power, perks, academic advancement, and the thrill of being a spy doctor. Despite the code of silence, and despite later claims by the Canadian Psychiatric Association that Dr Cameron was unaware he was working for the CIA, unwitting investigator status for Cameron is implausible for several reasons. He was far too politically connected to be unwitting. At various times, Dr Cameron was president of the Quebec, Canadian, American and World Psychiatric Associations, the Society of Biological Psychiatry, and the American Geriatrics Society. Dr Cameron was one of four co-founders of the World Psychiatric Association; another was Dr William Sargant,1 the foremost British authority on brainwashing. Many Board Members and Presidents of the Society of Biological Psychiatry were LSD researchers, funded by the military or otherwise in the mind control network. A letter from the CIA to Senator Pete Wilson dated 11 December 1985 states that the CIA contacted Dr Cameron directly. On page 4, the correspondent says: First, the CIA did not instigate this research, create the protocol, or supervise the work. Rather CIA contacted a prominent and highly respected Canadian psychiatrist, Dr Ewen Cameron, who was conducting research into treatment of mental illness with drugs such as LSD, and the CIA provided minimal and partial funding for a short time period. In return, the CIA received periodic reports on his research into behavioural modification through a process which he termed "psychic driving." Dr Cameron was eulogized in obituaries in the Canadian Psychiatric Association Journal,2 the Canadian Medical Association Journal,3 the American Journal of Psychiatry,4 and Recent Advances in Biological Psychiatry,5 the latter written by Hudson Hoaglund, PHD, who was personally referred to J. Edgar Hoover by G.H. Estabrooks. Dr Cameron received many awards including the Adolph Meyer Award, the Samuel Rubin Award, and the Montreal Mental Hygiene Institute Award, given to "a scientist who has made an outstanding

Source From The CIA Doctors: Human Rights Violations by American Psychiatrists (Richardson, TX: Manitou Communications, 2006). Reprinted by permission of the author.

II

ail

111111

9 Research with Humans

contribution to the mental health of the Canadian people." In a n article entitled "McGill University Department of Psychiatry 50th Anniversary" Pinard and Young6 echo the sentiments of the eulogist: Since the department's inception in 1943, research has been a preponderant part of its mission; this was stated in the very first reports to the university by the department's founder, Ewen Cameron. . . . The departmentb record has not been one of unblemished success. Cameron's drive led to the foundation and growth of the department, but also led him to perform much publicized experiments of doubtful ethical or scientific value in which patients received multiple courses of ECT or doses of LSD. In his obituary in the Canadian Psychiatric Association Journal2 Cameron is eulogized as follows: As a diligent seeker after new knowledge, a gifted author, a renowned administrator, and inspiring teacher he brought, not only to his professional colleagues but also to the community at large, a wider and deeper understanding of the importance and significance of the emotional life of man. Dr Cameron began conducting unethical, unscientific, and inhumane brainwashing experiments at Brandon Mental Hospital in :he 1930s. He continued this work into the 1960s. In one paper' Dr Cameron describes treating schizophrenics with red light produced by filtering light from 15 200-watt lamps through an inch of running water and a layer of sodium salt of ditolyl disazo-bis-napthylamine and sulphuric acid impregnated into cellophane. The colour red was chosen because it is the colour of blood. In these experiments, schizophrenic patients were forced to lie naked in red light for eight hours a day for periods as long as eight months. Another experiment involved overheating patients in an electric cage until their body temperatures reached 102 degrees F. After leaving Brandon Mental Hospital in 1936, Cameron took a job at Worcester State Hospital in Massachusetts. The Worcester Foundation for Experimental Biology received CIA money through MKULTRA Subproject 8, and was tie professional home of Dr Cameron's eulogist, Hudson Hoaglund.8 At Worcester State Hospital, Dr Cameron massively over-utilized insulin coma therapy by putting patients in coma for two to five hours per day for up to 50 days in a row. In a paper published in the American Journal of Psychiatry entitled "Psychic Driving," Dr Cameron9

describes his brainwashing techniques and says, "Analogous to this is the breakdown of the individual under continuous interrogation." Psychic driving was a procedure carried out in two stages; in the first stage, patients were depatterned, which meant they were reduced to a vegetable state through a combination of massive amounts of electroconvulsive shock, drug-induced sleep, and sensory isolation and deprivation. When fully depatterned, patients were incontinent of urine and feces, unable to feed themselves, and unable to state their name, age, location, or the current date. In the second stage, psychic driving was introduced. This consisted of hundreds of hours of tape loops being played to the patient through earphones, special helmets or speakers in the sensory isolation room. The tape loops repeated statements of supposed psychological significance. If such procedures were carried out under third world dictators, they would be denounced as human rights violations by American and Canadian psychiatry, and would be called brainwashing. There is a further reason to conclude that Ewen Cameron had a security clearance and was witting of CIA funding of his research; Dr Cameron definitely had a security clearance with the US government. In 1945 he was part of an American team that did psychiatric assessments of German War criminals including Rudolph Hess, who was examined at the request of the Military Tribunal in Nuremberg. Dr Cameron must have heard about the mescaline research done in the death camps by Nazi psychiatrists. He himself instituted similar work at McGill when he began experimenting with LSD. Rather than being the object of suspicion and investigation in the 1950s, Dr Cameron was well regarded in the Canadian media. Favourable articles about him were entitled "Canadian Psychiatrists Develop Beneficial Brainwashing,"10 "New 'Personalities' Made to Order,"u and "Two-Month Sleep, Shock New Schizophrenic Cure."12 Similarly as recently as 6 June 1987, the official position of the Canadian Psychiatric Association on Dr Cameron's brainwashing experiments was far from negative: . . . the fact that Dr Cameron's research would not be accepted by today's standards of ethical and scientific inquiry cannot be used as a retrospective critique of his work. What has to be recognized clearly is that in the intervening 20 to 30 years there has been a continuing progression of scientific and ethical research standards that included much more sophisticated peer review and ethical approval review now in place as part of standard practice. This represents the evolution of concern and control for all

Ross: Dr Ewen Cameron

medical research using human subjects deriving in part out of concerns experienced in several fields of medicine. Such experiments would not be permitted in today's research climate. The position on Dr Cameron taken by the Canadian Psychiatric Association is mistaken for several reasons. Dr Cameron received a grant from Canada's Department of Health and Welfare for $57,750.00 for the years 1961 to 1964 for "A Study of Factors Which Promote or Retard Personality Change in Individuals Exposed to Prolonged Repetition of Verbal Signals." The Helsinki Declaration governing ethical rules for medical research was adopted in 1964; Dr Cameron's brainwashing experiments clearly violated the principles of informed consent and protection of the patient from undue harm contained in the Helsinki Declaration. Dr Cameron's experiments alsoviolated the informed consent provisions of the Nuremberg Code, which arose out of the war crime trials of the Nazi doctors, in which Dr Cameron participated as a member of the American psychiatric team. He thus had direct knowledge of the medical atrocities the Nuremberg code was designed to prevent. The Canadian Psychiatric Association's position that Dr Cameron's research would "not be permitted in today's research climate" is correct, but ignores the fact that the rules of ethical conduct in medical research have not changed since Nuremberg. The fact that medical schools were routinely lax in ensuring that prevailing ethical codes were adhered to in the 1950s and 1960s is a condemnation of the medical schools, not a vindication of Dr Cameron. I consider the Canadian Psychiatric Association's official position on the mind control experiments conducted by Dr Ewen Cameron to be a violation of the Hippocratic Oath. Lies and silence concerning psychiatric mind control experimentation are a betrayal of the physician's ethical duty The fact that Dr Cameron's unethical, inhumane, and grossly damaging experiments were published in the psychiatric literature is a condemnation of the editorial standards of the journals, not a vindication of Dr Cameron. The only argument protective of the psychiatric journals is the fact that Dr Cameron whitewashed the experiments for publication. Dr Cameron's brainwashing experiments stopped in 1964, whereas the Tuskegee Syphilis Study continued until 1972. The continuation of the Tuskegee Syphilis Study under the auspices of the Center for Disease Control until 1972 does not provide vindication for Dr Cameron, rather it provides further grounds for criticism of organized medicine. The US Government has officially apologized to and financially compensated the victims of the

radiation experiments and the Tuskegee Syphilis Study, and the Canadian Government has established a fund that compensates victims of unethical experiments by Dr Cameron at the Allan Memorial Institute; compensation of $100,000.00 can be activated by documented victims by calling a toll-free number provided by the Canadian Government. Given the positions taken by two federal governments on such medical experiments, the position of the Canadian Psychiatric Association on Dr Ewen Cameron requires revision. Dr Cameron was not the only researcher at McGill funded by the CIA and the military. Another psychiatrist at McGill, Dr Raymond Prince,13 was funded through MKULTRA Subproject 121. Dr Prince was an unwitting investigator and is the only psychiatrist to have written about DA mind control in the peer-reviewed medical literature. He is the only MKULTRA contractor to have publicly identified himself to date. No other MKULTRA contractor has engaged in any public discussion of psychiatric participation in CIA and military mind control. Dr Hassan Azima was a young McGill psychiatrist who was being groomed as a military mind control contractor prior to his death from cancer in his early forties. A colleague, Dr Sarwer-Foner,'4 gave the Hassan Azima Memorial Lecture at a meeting of the Society of Biological Psychiatry; Dr Cameron was a Past President of the Society. Dr Azima15,16 worked at the Allan Memorial Institute, where he gave psilocybin to patients; psilocybin is the active ingredient of "magic mushrooms." He also attended LSD symposia and performed sensory isolation experiments17 that caused damage to patients. Two patients with "obsessional neuroses manifested acute psychotic episodes. They were treated with electric shock, which resulted in improvement in both paranoid and obsessional features." Azima and Cramer18 write: Contrary to the above case, a hebephreniccatatonic girl who remained in isolation for six days showed no perceptual alteration. Behaviourally, she manifested overt hostility, became quite talkative and self-assertive. Her F.D. [figure drawings] revealed gradual, but definite emergence of aggressive tendencies. She also experienced several spontaneous orgasms, and verbalized memories of her "sexual adventures." Another patient in the series is described as follows: Another case of obsession neurosis, suffering severe motor compulsions, who had not responded to any form of treatment, was put in isolation with the explicit aim of provoking a psychotic disorganization.

Al

II AEI

IIILLI

9 Research with Humans

He remained five days in isolation, began to manifest signs of depersonalization on the second day, and showed several acute psychotic episodes, lasting about three hours on the fourth and fifth days. The disorganization manifested itself, in part, as a marked disinhibition. He experienced many spontaneous orgasms, and manifested overt erotic behaviour toward the nurses. His eating habits deteriorated, and his behaviour was like that of a very hungry child during the feeding periods. In the postisolation period he showed some reorganization and lost some of his motor compulsions. But because of the appearance of some paranoid tendencies, he was put on electric shock therapy, which resulted in considerable improvement and subsequent discharge. Dr Azima19 published a paper with Dr Eric Wittkower, who worked at the Transcultural Psychiatry Institute at McGill, where Dr Prince was employed. Dr Wittkower founded and. edited The Transcultural Psychiatric Research Review. The Review was funded by CIA cutout the Society for the Investigation of Human Ecology, which lists a payment to Dr Wittkower of $7,500.00 in its 1961 Annual Report. The Board of Advisors for the Review inc:_uded Dr Ewen Cameron and Margaret Mead, who received CIA money for her anthropology research, and who was married to Gregory Bateson. Bateson took LSD supplied to him by a psychiatrist, and both Bateson and Mead were members of the Cybernetics Group, which was funded by CIA cutout The Josiah Macy, J r. Foundation.

Another McGill psychiatrist, Dr James Tyhurst, worked at the Allan Memorial Institute and received funding from Canada's Defence Research Board for studies of individual reactions to community disasters.2° Disaster studies were also the subject of investigation in MKULTRA Subproject 126, which was approved by the CIA in 1960. Dr Tyhurst attended a meeting with CIA personnel in 1951 in Montreal devoted to oversight of BLUEBIRD and ARTICHOKE.21 He also worked at Hollywood Hospital in Vancouver, where hundreds of patients were treated with LSD.22 Dr Donald Hebb, Head of the Department of Psychology at McGill during the 1950s, received funding from Canada's Defence Research Board for experiments on sensory isolation.23 The network of doctors with CIA and military funding at McGill included Dr Cameron, Dr Hebb, Dr Tyhurst, Dr Wittkower, and Dr Prince, and in addition Dr Azima was firmly established in the mind control network and using many of the same experimental procedures. LSD research was also done at McGill and Montreal General Hospital by Dr J.H. Quastel.24 Any claim that Dr Cameron's CIA funding was an anomaly or isolated incident is therefore incorrect. Medical experimentation by the Department of Psychiatry at McGill resulted in death, psychosis, vegetable states, organic brain damage, and permanent loss of memory among other damages. It resulted in the creation of amnesia, identity disturbance, and depersonalization among other dissociative symptoms. Dr Ewen Cameron was the main figure in these activities.

Notes 1. Sargant, W. 1957. Battle For the Mind. Doubleday: Garden City, NY. 2. Anonymous. 1967. "In Memoriam. Donald Ewen Cameron: 1901-1967," in Canadian Psychiatric Journal 12: 475. 3. Cleghorn, R.A. 1967. "D. Ewen Cameron, MD FRCP [C]," in Canadian Medical Association journal 97: 985. 4. B., E J. 1967. "D. Ewen Cameron: 1901-1967," in American Journal of Psychiatry 124: 168-9. 5. Hoagland, H. 1967. "Donald Ewen Cameron: 1901-1967," in Recent Advances in Biological Psychiatry 10: 321-2. 6. Pinard, G., and S.N. Young. 1993. "McGill University, Department of Psychiatry 50th Anniversary," in Journal of Psychiatric Neuroscience 4: 141-2. 7. Cameron, D.E. 1936. "Red Light Therapy in Schizophrenia," in British Journal of Physical Medicine 10: 11. 8. Hoagland, H., and Donald Ewen Cameron: 1901-1967. 1967. Recent Advances in Biological Psychiatry 10: 321-2. 9. Cameron, D.E. 1956. "Psychic Driving," in American Journal of Psychiatry 112: 502-9. 10. Moore, J. 1955. "Canadian Psychiatrists Develop Beneficial Brain Washing," in Weekend Magazine 5: 40.

11. Cahill, B. 1956. "New 'Personalities' Made to Order," in Montreal Gazette (18 June). 12. Cahill, B. 1957. "Two Month Sleep, Shock New Schizophrenic Cure," in Montreal Gazette (2 Sept.). 13. Prince, R. 1995. "The Central Intelligence Agency and the Origins of Transcultural Psychiatry at McGill University," in Annals of the Royal College of Physicians and Surgeons of Canada 28 (7): 407-13. 14. Sarwer-Foner, G. J. 1963. "On the Mechanisms of Action of Neuroleptic Drugs: A Theoretical Psychodynamic Explanation. The Hassan Azima Memorial Lecture," in Recent Advances in Biological Psychiatry 6: 244-57. 15. Azima, H. 1958. "Sleep Treatment in Mental Disorders," in Diseases of the Nervous System 19: 523-30. 16. Azima, H. 1962. "Psilocybin Disorganization," in Recent Advances in Biological Psychiatry 5: 184-98. 17. Azima, H., and E J. Cramer. 1956. "Effects of Partial Perceptual Isolation in Mentally Disturbed Individuals," in Diseases of the Nervous System 17: 117-22. 18. Ibid.

Ross: Dr Ewen Cameron

19. Azima, H., E.D. Wittkower, and J. LaTendresse. 1958. "Object Relations Therapy in Schizophrenic States," in American Journal of Psychiatry 115: 60-2. 20. Tyhurst, J.S. 1951. "Individual Reactions to Community Disaster," in American Journal of Psychiatry 107: 764-9. 21. Gillmor, D. 1987. I Swear By Apollo. Dr Ewen Cameron and the CIA Brainwashing Experiments.: Eden Press: Montreal.

22. MacLean, J.R., MacDonald, D.C., Ogden, F, and E. Wilby. "LSD 25 and Mescaline as Therapeutic Adjuvants," in The Use of LSD in Psychotherapy and Alcoholism, ed. H. Abramson. Bobbs-Merrill Company: New York, 407-29. 23. See note 21. 24. Author. 1959. "Clues to Biochemistry o f Schizophrenia: May Lead to Rational Therapy of Disease," in Factor 8-9 (Dec.).

Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis—Learning from the Cases of Nancy Olivieri and David Healy

•

Arthur Schafer

• The leading individual roles in this diptych are taken by two internationally eminent medical researchers, hematologist Nancy Olivieri and psychiatrist David Healy. The institutional players include one research-intensive university (the University of Toronto) and two affiliated research-intensive teaching hospitals (the Hospital for Sick Children, referred to as "Sick Kids" or "Sick Kids' Hospital" and the Centre for Addiction and Mental Health, referred to as "CAMH"). The cast of supporting characters is large. On one side are senior hospital administrators and medical faculty deans, together with hospital and university presidents and boards of directors. On the other side is to be found a small group of medical scientists, supported primarily by the Canadian Association of University Teachers (CAUT). Not coincidentally, the Olivieri and Healy scandals share in common a number of key elements •

•

• •

Wealthy and powerful drug companies hover in the background of both, and sometimes occupy a good deal of the foreground, as well: Apotex in the case of Olivieri, Eli Lilly in the case of Healy. These drug companies not only fund university and hospital researchers, they are also major donors to the institutions within which researchers carry out their clinical studies. Neither Apotex nor Eli Lilly was happy to have adverse information about their drugs publicized. Both Olivieri and Healy personally experienced serious negative consequences from their willingness to speak publicly about potential dangers to patients.

•

•

•

•

Each of them appealed for assistance, unavailingly, to the senior administrators of the University of Toronto and its Faculty of Medicine. Although there had been a changeover of university presidents and medical faculty deans in the interval between these two scandals, personnel changes made very little difference to the university's official response. In both scandals, university and hospital officials failed to recognize that there had been a fundamental violation of the principle of academic freedom at the affiliated hospitals. In both cases, the whistle-blowing physicians found themselves removed from their positions: Olivieri was fired from her position as director of the Hemoglobinopathy Research Program at Sick Kids' Hospital; Healy's employment contract with both CAMH and the University of Toronto's Department of Psychiatry was terminated. Both hospitals and the university denied strenuously that these "firings" were in any way related to the whistle-blowing Damaging rumours were circulated among Olivieri's colleagues, including allegations that she was scientifically incompetent, guilty of stealing money from her research grants, unethical in her patient care, and sleeping with some of the scientists who looked favourably on her research findings;' damaging rumours were circulated about Healy that he was a bad clinician, and both a racist and a member of a cult known as Scientology. A journalist who telephoned me for an interview at the height of the Healy controversy asked whether I knew that Healy was a prominent Scientologist. Her previous interviewee had been a hospital spokesperson who was circulating that piece of disinformation among the media, presumably in an effort to discredit Dr Healy. The perpetrators of these false but damaging accusations against Olivieri and Healy mostly preferred to remain anonymous. . . .

Source: From Journal of Medical Ethics 30 (2004) 1, 8-24. Copyright © BMJ Publishing Group Ltd with permission from BMJ Publishing Group Ltd. Reprinted with permission.

II aIi

111111

9 Research with Humans

Nancy Olivieri and the Hospital for Sick Children

. . . Once Dr Olivieri came to believe, based on scientifically credible preliminary evidence, that the experimental treatment she was administering might cause unanticipated harm to some of her patients/research subjects, she was duty-bound to disclose those risks. (The risks of harm were discovered by serial liver biopsy but actual harm, were it to occur, would be expected to occur very gradually over a period of many years.) Olivieri's university and her hospital had a corollary duty to support her request for assistance in this exercise of academic freedom and in the performance of her obligations as a physician and a researcher. Their failure to provide this support in an effective manner raises important questions about the way in which society funds biomedical research institutions and biomedical research. Every version of the Hippocratic Oath, from ancient times down to the present day, has had, as its leading principle, some version of the maxim that "the life and health of my patient will be my first consideration." Thus, whether or not Dr Olivieri is ultimately proven to have been correct in her negative interpretation of the preliminary scier tific data, once her data indicated the possibility of unanticipated harm, she was morally obliged to inform her patients of this risk. Writing in the New England Journal of Medicine, two blood science researchers, David G. Nathan, of Harvard's Dana-Farber Cancer Institute and David J. Weatherall, o f Oxford University, comment that: "as of this writing, the safety and efficacy of deferiprone have not been established." They suggest that it takes years of careful monitoring before the effectiveness of any iron chelator is clinically established; but they then go on to remark that: "Suffice it to say, when the dispute began, Olivieri had good reason to believe that deferiprone was neither safe nor effective."2 (Since Apotex discontinued the clinical trial prematurely, no conclusive scientific evidence exists, and the scientific aspect of the controversy remains unresolved.) One important qualification should be appended to this claim. Given that the stakes were high, both for the patients/research subjects and for the drug company, Olivieri had an obligation to exercise due diligence by consulting qualified colleagues about her interpretation. This she did, and they supported her concerns. (Apotex, however, most definitely did not agree with Olivieri's interpretation of her data, and the company was supported in its favourable interpretation by a number of scientists receiving financial support from them.) It would then be the responsibility of patients to weigh the hoped-for benefits against the

possible risks of harm. Respect for the value of patient autonomy clearly requires that those patients who are also research subjects be given all materially relevant information in order to enable them to decide whether they wish to continue participating in a clinical trial. It should go without saying that the information to which patients are morally and legally entitled includes information about risks of harm which comes to light during the course of a clinical trial. Olivieri also had ancillary obligations to report any newly discovered risks to the research ethics board of her hospital and to share her findings with other researchers, both at scientific meetings and in peer reviewed journals. Only in this way could her colleagues, worldwide, test and assess her conclusions and properly inform their own thalassemia patients of newly discovered potential risks. In every case, Dr Olivieri behaved in the manner required by her professional obligations, though she, and the core group of colleagues who supported her, paid a heavy career and personal price for doing her/their duty. Dr Brenda Gallie, Dr Helen Chan, Dr Peter Dune, and Dr John Dick were all colleagues of Olivieri at the Hospital for Sick Children. All supported Olivieri in her struggles with Apotex, Sick Kids' Hospital, and the University of Toronto, despite serious risks of harm to their own careers.3 It is difficult not to empathize with Olivieri when she laments: "It should not be so hard to protect children at Sick Kids' Hospital." This sentence was quoted by a colleague and supporter of Olivieri, Dr Paul Ranalli, in a letter published by the Globe and Mail, headed "Courage under Fire."4 It is true, of course, that Apotex had a legal contract with Dr Olivieri, which was signed in 1995. That contract contained a confidentiality provision—one that prohibited her from disclosure "to any third party" of data from her Apotex-sponsored clinical trial of the drug deferiprone, without the express permission of the company, for a period of three years after the termination of the trial. The non-disclosure clause of the LA—Ol contract between Olivieri and Apotex reads as follows: All information whether or not obtained or generated by the investigators during the term of this agreement and for a period of one year thereafter, shall be and remain secret and confidential and shall not be disclosed in any manner to any third party, except to an appropriate regulatory agency for the purposes of obtaining regulatory approval to manufacture, use or sell Ll unless the information has been previously disclosed to the public with the consent of Apotex. The investigator shall not submit any information for publication without the prior written approval of Apotex.

Schafer: Biomedical Conflicts of Interest

This clause, it should be noted, does not specifically list "patients," but they would clearly appear to be covered under the phrase "any third party." Olivieri claims, supported by tapes of telephone conversations with Apotex, that Apotex threatened repeatedly to sue her if she breached the confidentiality clause of the contract, and that they warned her not to disclose her concerns to patients and others. At the time Olivieri signed the contract such nondisclosure provisions were common. Olivieri readily admits that she failed to appreciate the potential significance of that contract, and concedes that she should never have signed it. The University of Toronto admits that it was guilty of an institutional oversight by permitting its researchers to agree to such terms, and the university subsequently took steps to preclude repetition by any of its faculty. Because Apotex refused, repeatedly, to give permission for disclosure, Olivieri might have been found legally liable for significant damages arising out of her disclosure of risks to her patients and colleagues. Since the protection of human life is, other things being equal, a higher value than respect for the sanctity of contracts, it is possible that the legal system would have "thrown out of court" any lawsuit for breach of contract brought by Apotex against Olivieri, as being against public policy and, hence, unenforceable. For our purposes, it matters little whether the nondisclosure provision of the contract Olivieri signed with Apotex would have been found by the courts to be nugatory. Nor, for our purposes, does it matter much whether the information Olivieri disclosed to patients and colleagues was information actually covered by the terms of the confidentiality agreement she signed—an issue also in dispute. Even if Nancy Olivieri were legally bound to keep confidential all information about the risks of deferiprone, she was morally obliged to disclose that information to her patients and to her colleagues worldwide. It could be argued, of course, that one has a moral obligation to keep the contracts one signs. This moral obligation is prima facie, however, rather than absolute, and should surely be overridden where the lives and health of patients are at stake. Apotex did not agree with Olivieri's interpretation of her data and they refused her request to disclose these risks to her patients. They also threatened to take legal action against her if she were to violate the nondisclosure clause of the contract. Olivieri proceeded anyway, in the face of these threats, to disclose her findings, and some time after these events, the company did take legal action against her. On 24 May 1996, for example, Apotex wrote to Olivieri that it was terminating both of the clinical trials she was conducting for

them, and warned her not to disclose information "in any manner to any third party except with the prior written consent of Apotex," and warned further that it would "vigorously pursue all legal remedies in the event that there is any breach of these obligations."5 Just prior to their suing her, she sued them for defamation. According to one standard account of heroism, the hero is a person who acts far beyond the call of duty. By this test, Olivieri's actions would not count as heroic. She only did that which it was her duty to do. But there is another account of heroism according to which the hero is a person who does her duty, at great risk to her own self-interest, when most others would resist from fear. Olivieri relates a story about her personal fears. Sitting in a restaurant with her scientific collaborator, Dr Garry Brittenham, she raised the concern with him that if they were to break their contract with Apotex by disclosing to patients the risks they had newly discovered, Apotex might act on its threats to sue, and the enormous costs of fighting such a lawsuit could mean that they would each lose their homes. Brittenham replied: "Red wine or white?" At that moment, Olivieri reports, it became obvious to her that there was no decision to make concerning whether or not to disclose. They had to disclose the risks. Better, therefore, to concentrate on those matters that still required a decision, such as the colour of the wine they were to consume with their dinner (The Current, CBC radio interview, 2 Mar 02). Apotex is currently suing Olivieri for damages, claiming that she defamed both the company and their drug (deferiprone). Olivieri is suing Apotex for defamation. For the benefit of those who have had the good fortune never to be involved in a legal action of this sort, it is perhaps worth noting that the costs of defending such an action (at least in North America and England) tend to be ruinously expensive; hence, utterly beyond the means of any except the wealthiest individuals. When Olivieri turned to her hospital and university for financial and other help in the face of intimidating threats of legal action against her, they provided little effective assistance.° Instead, both the University of Toronto and the Hospital for Sick Children "took actions that were harmful to Dr Olivieri's interests and professional reputation, and disrupted her work."7 In their public pronouncements about the case, none of the senior administrators of the university, the medical faculty, or the hospital gave any sign that they recognized that the case was one involving a serious issue of academic freedom. They justified their official "tread lightly" policy in part by characterizing the conflict as a "scientific dispute," to be resolved primarily between the parties themselves. Some University of Toronto officials did make efforts, behind the scenes, to promote

MI III

II

AI

II

j11

111111

9 Research with Humans

a settlement between Olivieri and Apotex but, as the Thompson report found, the support which they offered was "not effective."8 It was discovered during this period of conflict and controversy that the University of Toronto was negotiating for a twenty million dollar donation from Apotex (with additional millions promised for its affiliated hospitals). Some were led to speculate that the university's failure to recognize and support Olivieri's academic freedom might not have been unconnected to its eagerness to secure financial support from Apotex for the university's proposed molecular medicine building project. Indeed, it was subsequently revealed that the university's then-president had gone so far as to lobby the Government of Canada on behalf of Apotex. In a private letter to the Prime Minister of Canada, President Robert Prichard stated that the government's proposed changes to drug patent regulations would adversely affect Apotex revenues and could thereby jeopardize the building of the university's new medical research centre. President Prichard was unsuccessful in persuading the federal government to change its drug patent laws, but his action demonstrated the lengths to which the university was prepared to go in appeasing the company or promoting its interests. When Prichard's conduct became public knowledge, he apologized to the executive committee of the university for acting inappropriately in this matter.9 This embarrassing episode illustrates the dangers that can ensue from university reliance upon industry "philanthropy." When career success for university/ hospital presidents and deans is measured in significant part by their ability to raise vast sums of money from corporate donors, such fundraising can easily become a dominating priority. In North America, top university and hospital officials are now required to ride two horses their fundraiser's horse and, simultaneously, their academic horse (as guardians of core university values). Unfortunately, those who attempt to ride two horses can come to grief when, as sometimes happens, the horses pull in opposite directions. Perhaps it is time for a radical rethinking of the competing role responsibilities of top university and hospital officials. The word "philanthropy" is placed above in warning quotes, not to suggest that big pharma never behaves in a genuinely philanthropic manner but, rather, to flag the point that when corporate donors make substantial donations they often expect to gain substantial influence. Indeed, it is the legally mandated duty of corporate executives and board members to act in the "best interests" of the corporation, which is commonly interpreted to mean that they have a legal duty to maximize overall profitability. Corporate donations to universities

are typically viewed, at least in part, as an investment. This, in turn, raises the questions (to which an answer is supplied later): What exactly is being bought by such investments? What exactly is being sold? David Healy and the Centre for Addiction and Mental Health

In December of 2000, while the Olivieri affair was still capturing attention, both within and without the University of Toronto, a second major scandal, also raising basic issues of academic freedom and patient safety, was brewing at the same university. Some months previously, the Centre for Addiction and Mental Health had hired Dr David Healy to become the new director of its Mood and Anxiety Disorders Clinic. After accepting their offer and the offer of a joint appointment in the university's Department of Psychiatry, Healy notified his employer in Wales of his intention to resign, and prepared to move his family to Canada to take up this new appointment. Then, Healy's career plans came dramatically unstuck. On 30 November 2002, some months before his new appointment was officially scheduled to begin, Healy made a conference presentation at CAMH. (The symposium, called "Looking Back, Looking Ahead," was held to mark the 75th anniversary of the university Department of Psychiatry, as well as the 150th anniversary of the Queen Street Mental Health Service.) In this lecture, which he subsequently delivered at Cornell, and in Paris, Minneapolis, and Cambridge, Healy raised the question of whether the drug Prozac, manufactured by Eli Lilly, might be responsible for increasing the risk of suicide among certain kinds of patients. This issue was by no means the principal theme of Healy's talk, but the potential link of Prozac to patient suicides, and the call for further research on this matter, was almost certainly regarded by CAMH officials as the most controversial part of Healy's presentation. Healy reports that his talk was well received in all the places where he presented it, and it is noteworthy that the audience at the CAMH conference honoured his lecture with the highest rating for content.1° Despite this fact, senior administrators of CAMH were not well pleased. Within 24 hours of the talk they were trying to contact him. Within a week he received an email unilaterally rescinding their offer of employment. Why was David Healy's employment terminated so precipitately by both the Centre for Addiction and Mental Health and the university Department of Psychiatry? No one disputes that Healy is an internationally distinguished psychiatrist and researcher. The university and CAMH recruited him with enthusiasm and

Schafer: Biomedical Conflicts of Interest

persistence. Since he was unhired almost immediately after he gave his conference lecture at CAMH, the inference is inescapable that his contract for employment was cancelled because of the contents of his lecture that day. In this lecture, Healy expressed the view, referred to above, that the antidepressant drug Prozac might cause some patients to commit suicide. Although Healy did not condemn Prozac outright, he did advocate caution on the part of doctors who prescribe this drug, and he called for further research into possible adverse side effects. He was also critical of the practice whereby drug companies are engaged in ghostwriting some of the therapeutic literature. Some time prior to Healy's conference presentation, Eli Lilly had donated 1.5 million dollars to CAMH, and a new wing of the hospital, built with their financial assistance, was scheduled to have its official opening soon after. There is no evidence that Eli Lilly attempted to have Dr Healy fired from his new appointment at CAMH. The incident raises legitimate questions, however, about whether those involved with rescinding his contract offer were affected, consciously or unconsciously by the relationship between CAMH and Eli Lilly. In this connection, it is worth noting that six months before Healy delivered his fateful presentation at CAMH, he had published an article on Prozac in the biomedical ethics journal, the Hastings Center Report.11 In this article, Healy developed several of the themes which later became controversial at the University of Toronto, namely suicide and Prozac, and ghostwriting of scientific articles by drug companies. Eli Lilly, which had hitherto been the largest annual private donor to the Hastings Center, publisher of the Hastings Center Report, subsequently withdrew its financial support for the centre. The administrations of both the University of Toronto and CAMH claim that the unhiring of Dr Healy had nothing to do with academic freedom. Instead, they contend, his lecture gave rise to "clinical concerns" and revealed that he would be a "bad fit" with his new colleagues. It may be worth quoting a key paragraph from the email which the University of Toronto sent to Healy by way of explaining their decision to rescind his contract: Essentially, we believe that it is not a good fit between you and the role as leader of an academic program in mood and anxiety disorders at the centre. While you are held in high regard as a scholar of the history of modern psychiatry, we do not feel your approach is compatible with the goals for development of the academic and clinical resource that we have. This view was solidified by your recent appearance at the centre in the context of an academic lecture.0

University of Toronto officials later denied that Healy was unhired because of fears on their part that if Dr Healy were allowed to take up his position drug companies might be reluctant in future to donate money to or fund research at the centre. Notwithstanding their strenuous denials, however, many people understood the above quoted words to mean". . . the university was worried about the risk to the financial inflows to the department from pharmaceutical company sources."13 In September 2001, an international group of physicians published an open letter to the president of the University of Toronto, in which they protested against what they termed the "maltreatment" of Dr Healy. In their open letter they concluded: "To have sullied Dr Healy's reputation by withdrawing the job offer is an affront to the standards of free speech and academic freedom." The signatories, who included two Nobel Prize winners, chose not to focus on the possible involvement of a drug company in university affairs, but they nevertheless insisted that the central issue in the case was the failure of the University of Toronto and CAMH to uphold "the standards of open discussion and frank exchange in university life."14 That is, the issue was essentially one of academic freedom. The university's official response to the concerns expressed by this international group of scholars was dismissive: they (the protesting scholars) were illinformed outsiders, unaware of all the pertinent information. University of Toronto spokespeople went even further in their defence of the unhiring of Dr Healy by suggesting that his publicly expressed concerns were dangerously irresponsible. On the University of Toronto's website, Healy's warnings about the potential hazards of Prozac were compared to the "fool" who cries "fire" in a crowded theatre.'5 To this accusation, Healy responds: "But what if there is a fire in the theatre?"16 It is worth bearing in mind that Prozac or other drugs of its class, known as SSR1s, are often prescribed to healthy patients with problems in living. If, as Healy believes the evidence indicates, some of these healthy patients become suicidal because of their ingestion of SSRI-type drugs, then a failure to warn them and their physicians of this potential side effect would be grossly irresponsible. The argument underlying such an analogy is, presumably that Healy's warnings (of possible adverse side effects from taking Prozac) might deter some depressive patients from using Prozac or other SSRI drugs and this, in turn, might result in their committing suicide. In other words, the university's position seems to be that when the values of clinical care clash with the values of science, the former should trump the latter. The problem with this argument, however, is that if valid it

41

II del

IIILL1

9 Research with Humans

proves too much. It proves that researchers ought never to warn patients of potentially harmful side effects lest some patients thereupon forgo an effective medication. This manifestation of an approach often labelled "physician paternalism" would be morally objectionable because it would usurp the patient's right to give informed consent to treatment. How can patients weigh and balance the benefits and harms of treatment options (including the option of not taking any antidepressant medication) if evidence about potential harms is deliberately withheld from them? There is by now a vast literature, both legal and ethical, in which the near universal consensus of philosophers and jurists is that competent adult patients have a fundamental right to give informed consent to treatment. In practice, this means that research scientists must make the results of their research public, so that physicians can adequately inform their patients about potential risks. The duty to warn would seem, then, to be a fundamental obligation of every research scientist. Both Healy and Olivieri were alerting patients and the scientific community to the need for further researc a into potentially serious adverse consequences of the drugs they were investigating. For either to have remained silent about their preliminary adverse data would surely have been a violation of their legal, as well as their moral, duty. Interestingly, on 10 June 2003, the Medicines and Healthcare Products Regulatory Agency [MHRA] of the UK issued a caution to physicians that Seroxat (Paxil) was "contra-indicated" in children under 18 for the treatment of major depressive disorder. Potential side effects include dramatically increased risk of "potentially suicidal behaviour" (G. Duff, personal communication, 2003). Thus, it seems that recent evidence further confirms the wisdom of Healy's warnings about drugs of the SSRI category. Sadly, his scrupulous caution appears to have cost him his job at the University of Toronto.

A short time after the university's dismissive rejection of the open letter, described above, Dr Healy initiated what might have been the first legal action in the English-speaking world based, in part, on the alleged tort of violating academic freedom. ("Alleged" because until there is a legal precedent in which the courts find that such a tort exists, one cannot be sure of its validity) A settlement was subsequently negotiated, which included the appointment of Healy as visiting professor in the Department of the History of Medicine (with unrestricted academic freedom to speak out publicly about any of the issues). The appointment as visiting professor of the history of medicine was for one week a year during each of the following three years. He was not permitted, however, to assume the position for which he had originally been hired, as director of the Mood Disorders and Anxiety Clinic of CAMH. David Healy, himself, feels little doubt about the most important lesson to be learned from his experience at CAMH, and he insists that it is the same lesson that should be learned from the experience of Nancy Olivieri at Sick Kids' Hospital: "What is involved is a contrast between the values of science and the values of business."" Although the Thompson report dealt only with the Olivieri case, the Thompson authors, like Healy, conclude that the problem is system-wide: "The safety of research subjects in clinical trials and the integrity of the research project are more important than corporate interests.' Nathan and Weatherall, in their NEJM commentary on the Olivieri case, reach a similar conclusion,19 as does Somerville writing in Nature,2° and as do the authors of the CMAJ article "Dancing with porcupines."21 Together, the Healy and Olivieri cases have forced both the university community and the wider public to confront the ways in which university-industry partnerships can imperil the fundamental values of academic freedom, research integrity, and patient safety . . .

Notes 1. 2.

3.

O'Hara, J. 1998. "Whistleblowing," in Maclean's Magazine: 66. Nathan, D.G., and Weatherall, D.J. 2002. "Academic Freedom in Clinical Research," in New England Journal of Medicine 347: 1368-70. Naimark, A., Knoppers, B.M., and Lowry, EH. 1998.

Clinical Trials of Li (Deferiprone) at The Hospital for Sick Children: A Review of the Facts and Circumstances. Hospital 4.

for Sick Children: Toronto. Ranalli, P 1998. "Courage Ur. der Fire" [letter], in Globe

and Mail. 5.

Thompson, J., Baircl, P., and Downie, J. 2001. Report of the Committee of Inquiry on the Case Involving Dr Nancy Olivieri, the

Hospital for Sick Children, the Un iversity of Toronto, and Apotex Inc. Canadian Association of Uriversity Teachers: Toronto, 143.

6. Ibid., 29. 7. Ibid., 32. 8. Ibid. 9. Ibid., 13. 10. Healy, D. 2000. "Good Science or Good Business?," in Hastings Center Report 30. 11. Ibid., 19-22. 12. Ibid., 6. 13. Ibid., 6. 14. Ibid. 15. Clarke, C. 1998. 'Top Scientists Allege U of T Academic Chill," in The Globe and Mail. 16. Healy, "Good Science or Good Business?," 11. 17. Ibid., 11.

Schafer: Biomedical Conflicts of Interest

18. Thompson, Baird, and Downie, Report of the Committee of Inquiry, 17. 19. Nathan and Weatherall, "Academic Freedom in Clinical Research." 20. Somerville, M.A. 2002. "Post-Modern Tale: The Ethics of Research Relationships," in Nature 1: 316-20.

21. Lewis, S., Baird, P, and Evans, R.G., et al. 2001. "Dancing with the Porcupine: Rules for Governing the University/ Industry Relationship," in CMAJ 165: 783-5.

9.3 Ethical Issues in the Design and Conduct of Clinical Research

An Ethical Framework for Biomedical Research Ezekiel J. Emanuel David Wendler, and Christine Grady Over the last 60 years or so, there has been myriad guidance on the ethical conduct of research with humans1-13 Despite the profusion, the extant guidance seems flawed in several respects. First, most guidance was "born in scandal."14 That is, the guidelines or reports were a response to a specific controversy, and therefore tend to focus on what was perceived to be the transgression of that scandal. The Nuremberg Code directly addressed the atrocities of the Nazi physicians; the Belmont Report was a response to the Tuskegee Syphilis Study and other scandals;4 and the Advisory Committee on Human Radiation Experiments responded to covert radiation experiments during the Cold War and therefore emphasized deception.15 Second, regulatory guidance tends not to examine the overall ethics of research but to have a specific practical purpose. For instance, the International Conference on Harmonisation has the purpose of creating common rules across developed countries for the "registration of pharmaceuticals for human use."8 The aim is more to enhance the efficiency of drug approval than to protect research participants, for which it defers to the Declaration of Helsinki.3 In general, these regulatory guidelines emphasize the procedural safeguards of informed consent and independent review by an institutional review board or research ethics committee because these leave "paper trails" that can subsequently be audited. Both of these deficiencies contribute to a third: existing guidance is neither comprehensive nor systematic. The guidelines tend to be lists of claims or principles. For instance, the Nuremberg Code with its 10 statements and the Declaration of Helsinki, originally with

22 principles subsequently expanded to 32, contain no elaboration." Such sparse, oracular statements lack an overarching framework to ensure that all relevant ethical issues are addressed. They also lack justifications for their claims, implying that the ethical guidance is either self-evident or beyond debate. Consequently, when controversies arise about whether the principle itself is valid or how a principle should be applied to a case, there is nothing to appeal to other than the authority of these documents. Agreement can frequently be secured on the broad principles, but this often hides deep disagreements about how they should be interpreted and applied to specific situations.1° Finally, and maybe most important, the existing guidance seems mistaken on some important issues. For instance, the Nuremberg Code's strong statement that "the voluntary consent of the human subject is absolutely essential" seems to prohibit all pediatric research.2 Yet this seems wrong. Similarly, the 1993 Council for International Organizations of Medical Sciences (CIOMS) guidelines recommended that Phase I or II studies of drugs and vaccines should be conducted first in sponsoring countries before being done in developing countries.17 Because of strong objections, especially by developing countries, a decade later this was deleted from the revision.° The most recent version of the Declaration of Helsinki addresses conflicts of interest through disclosure, requiring that potential research participants be adequately informed about "any possible conflict of interest" and that these "should be declared in the publication."3 The value and importance of disclosing conflicts of interest to research participants is controversia1.18 More important, exclusive reliance on disclosure in the absence of prohibitions on certain conflicts of interest seems inadequate.19.20 Because of the deficiencies of existing research ethics guidance, there is a need for a broader, systematic, and comprehensive framework that includes an

Source: From The Oxford Textbook of Clinical Research Ethics (2011), Chapter 11:123-135. Reprinted with permission.

II .111

496

1111.11

9 Research with Humans

ethical justification and specification for how each principle is to be fulfilled in practice.21'22 Among other goals, this framework should incorporate those concerns that overlap in the existing guidance and organize them into a coherent whole. Fundamental Ethical Purpose

Informing this overarching framework is the understanding that the fundamental ethical challenge of all research with humans is to avoid exploitation.21.22 Research aims at obtaining generalizable knowledge that can be used to improve health and health care. Participants in research are a necessary means to obtaining this knowledge. Consequently, participants are used in the research process for the benefit of others and are at risk of being exploited. The fundamental purpose of research guidelines is to minimize the possibility of exploitation in clinical research. There are two distinct conceptions of exploitation. Both are important in protecing research participants. One is the traditional, Kantian notion of exploitation as using an individual merely as a means and not simultaneously as an end in itself.23,24 This Kantian conception of exploitation is grounded in the use of individuals for an end they do not agree with or to which they have not consented. Using individuals without their consent violates their autonomy.25 The remedy for the Kantian type of exploitation is obtaining informed consent and sometimes ensuring collaborative partnership with a larger community that agrees to the research. A second conception of exploitation elaborated by Alan Wertheimer rests on the unfair distribution of the benefits and burdens of an interaction.26,27 This is distinct from the Kantian conception because it concerns the distribution of benefits—who benefits and how much they benefit—rather than autonomy. Importantly, this type of exploitation can occur even when the interacting parties provide valid consent.26 Minimizing this type of exploitation is moire complex, requiring the fulfillment of multiple principles.27 Principles and Benchmarks of Ethical Clinical Research

The following eight ethical principles provide a comprehensive and systematic framework to guide the ethical conduct of clinical research and thereby minimize the possibility of exploitation21.22 (see Table 1). These principles are general and identify considerations necessary to justify research as ethical. They are conceptually included in most of the previously mentioned guidance, although existing guidelines do

not necessarily include all of them. In addition, they are presented sequentially, going from the development of research proposals to the conduct of research to monitoring during research. Each principle is specified by benchmarks that offer a specific elaboration and understanding of each principle.22 The benchmarks are practical interpretations of what is required to fulfill each principle.22,28.29 In this sense, the benchmarks should clarify and focus the kinds of values and considerations at stake in fulfilling each principle. No matter how specific and detailed, the benchmarks cannot eliminate all controversy over the principles.16.22 However, by specifying and clarifying the eight principles, these benchmarks should help to narrow any disagreement related to specific cases, making it easier to focus on the substance of the disagreement, assess the importance of the problems and concerns, and even identify potential solutions.22 Collaborative Partnership

Clinical research is meant to serve a social good, to enhance the health and health care of people. It is part of the way people collectively improve their well-being. Clinical research is not meant to be done to people but done with people." The principle of collaborative partnership recognizes that the community in which research is conducted should collaborate in the research endeavor.22,27 Seeking the community's agreement and input helps ensure that the particular community will not be exploited.27 In addition, collaboration helps ensure—although it does not guarantee—that the community will receive fair benefits from the conduct of the research.27." Collaborative partnership helps ensure that the community determines for itself whether the research is acceptable and responsive to its health problems. Finally, collaborative partnership is practically important. Without the engagement of researchers and community members, research is unlikely to have any lasting impact. Without the investment of health policy makers, the research results are unlikely to influence policy making and the allocation of scarce health-care resources.22 Collaborative partnership can be fulfilled through myriad formal and informal mechanisms. For instance, establishment of community advisory boards, consultations with advocacy groups, public meetings with community members, and advocacy for funding of research are approaches to developing collaborative partnerships.3o.32 Which method is preferred depends upon the nature of the particular research study. Because many of these mechanisms exist in the background without the need to launch explicit initiatives or are just part of "doing business," collaborative

Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research

partnership has infrequently been included as an explicit ethical requirement of clinical research?' One example of research that fails on collaborative partnership grounds includes "helicopter research" in which researchers arrive in a community, take samples, and leave, never to return. Several benchmarks are essential to fulfilling the principle of collaborative partnership.22 First, collaborative partnership obviously requires partners. This means identifying representatives of the target community to be involved in the research. Second, it requires collaboration. This entails sharing responsibility for assessing the importance of the health problem and the value of the research to the community, as well as for planning and conducting the study, disseminating the results, and ensuring that the results are used for health improvements. Third, a collaborative partnership requires mutual respect. This entails recognition of and respect for a community's distinctive values, circumstances, culture, and social practices.3° Importantly, respect does not mean uncritical acceptance of practices that might be oppressive or coercive. Indeed, some of these practices may be challenged in research. A true collaborative partnership based on respect also aspires toward equality between the partners. In this sense, collaborative partnership aspires to minimize the deprived circumstances of the involved community. Research aims to ameliorate deprivations usually of disease and sometimes of social circumstances. This could occur through a number of interventions directly related to the goals of the research project or ancillary mechanisms such as developing the general infrastructure necessary to actually conducting ethical research. Fourth, the community in which the research is being conducted should receive fair benefits from the conduct and/or results of the research.27,3' What level of benefits is fair depends upon the burdens the community bears for the conduct of the research.26 Such benefits might include direct benefits to the research participants as well as more indirect benefits such as employment and training for community members to augment health care services for the entire community.27•3' Finally, collaborative partnership requires a fair distribution of the tangible and intangible rewards of research among the partners. Very little can generate more resentment, mistrust, and sense of exploitation than an unfair distribution of the benefits of collaboration. This may require agreements regarding sharing intellectual property rights, royalties, and other sources of financial profit as well as appropriate authorship and other credit for contributions to the research.27,3'

Social Value

Clinical research is not an end in itself. It has instrumental value because it generates knowledge that leads to improvement in health or health care.33,34 It is such improvements in health that ultimately constitute the social value of research. Unfortunately, the emphasis on protection of research participants has displaced the importance of assessing research's social value. Without social value, research exposes participants to risks for no good reason and wastes resources.21•22•33-35 However, the process of translating research results into health improvements is complex, incremental, and haphazard.36 Typically, early studies are valuable because the data they generate informs additional research that ultimately could improve health. Priorities may change while a study is being conducted, and the cooperation of diverse groups is often needed to make changes based on research results. This makes the process of going from research to health improvement uncertain and arduous. Assessment of the value of research is made prospectively before any data are collected. Consequently, determinations of social value are uncertain and probabilistic, entailing judgments about the usefulness of a sequence of research and chances of implementing the results.35,36 Even in wealthy countries with well-established research studies and health system infrastructures, research results are imperfectly incorporated into clinical practice. Certain kinds of research clearly lack social value: for example, research that is nongeneralizable, that addresses a problem of little relevance to anyone, that will not enroll sufficient numbers of patients, that assesses proven or empirically well-established results, and research that could never be practically implemented to improve health or health care even if effective in the research setting.37.38 Consideration of four benchmarks helps to ensure fulfillment of the principle of social value.2',22 First, to whom will the research be valuable? It is important to delineate both the short-term and long-term prospective beneficiaries of the research study, specifying whether they include a specific group, similarly situated groups, a larger community from which research participants will be recruited, the country hosting the research, or people outside the host country.22 Second, what is the potential value of the research for each of the prospective beneficiaries? Potential beneficiaries may rank the health problem's importance differently and may receive different benefits from the research results. Factors to be considered might include how widespread the disease or condition is, the impact of the disease on individuals and communities, and the extent to which the research is likely to offer an

498

9 Research with Humans

Table 1 Principles and Benchmarks for Ethical Clinical Research Principles

Benchmarks

Collaborative partnership

• Which community representatives will be partners, involved in helping to plan and conduct the research, disseminate the results and use the results to improve health? • How will responsibility be shared with these partnersfor planning and conducting the research, disseminating the results and using the results to improve health? • How will respect for the community's values, circumstances, culture, social practices, and so forth, be demonstrated? • How will fair benefits for the community from the conduct and results of the research be assured? • How will the tangible benefits of the research, such as authorship credit and intellectual property rights, be distributed to ensure fairness?

Social value

• Who will benefit from the conduct and results of research? • What is the potential value of the research for each of the prospective beneficiaries? • How will the social value of the research be enhanced? • How can adverse impacts, if any, of conducting the research be minimized?

Scientific validity

• Do the scientific and statistical design and methods satisfy generally accepted standards and achieve the objectives of the study? If not, is there clear justification for the deviations? • Will the research results be interpretable and useful in the context of the health problem? • Does the study design ensure participants health-care services they are entitled to? If not, are there methodologically compelling reasons and are participants protected from serious harm? • Is the research design practically feasible given the social, political, economic, and cultural environment?

Fair participant selection

• Is the research population selected to ensure that the research complies with scientific norms and will generate valid and reliable data? • Is the research population selected to minimize risks to the participants? • Are the individual research participants selected to maximize social value and enhance the possibility of benefits to the participants? • Are the participants vulnerable based on age, clinical status, social marginalization, economic deprivation, and so forth? If so, what safeguards are included to protect the participants?

Favorable risk-benefit ratio

• Are the potential physical, psychological, social, and economic risks of the research for the individual participants delineated and their probability and magnitude quantified to the extent possible given the available data? • Are the potential physical, psychological, social, and economic benefits of the research for the individual participants delineated and their probability and magnitude quantified to the extent possible given the available data? • When compared, do the potential benefits to the individual participants outweigh the risks? If not, does the knowledge gained from the study for society justify the net risks to the individual participants?

Independent review

• Are the procedures for independent review established by law and regulation being properly followed? • Is the review body both independent and competent? • Is the review process transparent, and are reasons given for the review committee's decisions? • Are multiple reviews minimized and reconciled if they conflict?

Informed consent

• Are recruitment procedures and incentives consistent with cultural, political and social practices of the potential participants and their community? • Are disclosure forms and verbal disclosure procedures sensitive to participants' culture, language, and context? • Is the information presented to participants complete, accurate, and not overwhelming? • Are there appropriate plans in place for obtaining permission from legally authorized representatives for individuals unable to consent for themselves? • Are supplementary consents or permissions, for example, from spouses or community leaders, obtained? If so, are there ways to ensure that the individual participant can still decide whether to participate independent of the spouse or community leader? • Are the mechanisms to symbolize consent consistent with participants' culture and context? • How will individual participants be made aware of their right to refuse to participate and are they actually be free to refuse?

Respect for participants

• How will the health and well-being of participants be monitored to minimize harms? Are the criteria for changing doses or procedures for stopping the study for the health of participants adequate? • How will the confidentiality procedures actually be implemented? • How will it be ensured that participants who want to withdraw can withdraw without penalty? • How will results of the research be disseminated? • What are the plans for care of the participants after the research is completed?

Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research intervention or information useful to the beneficiaries. For example, because malaria is a substantially greater health problem for certain developing countries than for developed countries, research on cerebral malaria may be of substantial value to people in developing countries. Conversely, research on prophylactic medications for malaria is likely to be more valuable for tourists, whereas research on a malaria vaccine may be perceived as valuable to everyone, but to a different degree. Similarly, research on new HIV/AIDS medications in a developing country, although needed in that country, could benefit those outside the host country more than the community in which the research is being conducted if the ultimate cost of the medication is high. Third, it is important to develop mechanisms to enhance the social value of research. Through collaborative partnerships, strategies should be devised to disseminate results in appropriate ways to key stakeholders including people with the disease, practicing clinicians, advocacy groups, health policy makers, and sometimes international health-care organizations.22,3° In addition to presentations at scientific conferences and journal publications, this may require novel forms of dissemination such as letters to patients, articles in advocacy publications, presentations at community gatherings, public service announcements in the media, or letters to clinicians. Social value can also be enhanced when research is integrated into a long-term collaborative strategy, so that one research project forms part of a more comprehensive research and health delivery strategy to address significant health problems." Finally, consideration should be given to the impact of the research on the existing health-care infrastructure. The conduct of the research should not undermine a community's existing health-care services or social structures and leave it worse off at the end of the research. Supplementing the existing system and contributing to sustainable improvements in health through the provision of additional resources, equipment, medications, or training appropriate to the research can enhance value.

Scientific Validity

Contrary to many claims, in research, science and ethics do not conflict.21,22,34 Valid science is a fundamental ethical requirement.21.22,35 Unless research generates reliable and valid data that can be interpreted and used by the specified beneficiaries of the research, it will have no social value and participants may be exposed to risks for no benefits.39,4° Research must be designed in a way that provides valid and reliable data. Four benchmarks are important in fulfilling the principle of scientific validity. First, the scientific and

statistical design and methods of the research must plausibly realize the objectives of the research and must also satisfy the generally accepted norms of research. Research must have clear, justifiable objectives, an adequate sample size, and unbiased and reliable outcome measures and statistical analyses. Deviations from such standards, such as innovative designs, must be plausibly justifiable to the research community. Second, a research study must be designed to generate results that will be interpretable and useful in the context of the health problem.15 Interventions should be selected to ensure that the design is useful in identifying ineffective or appropriate interventions; implementing socially, culturally, and economically appropriate changes in the health-care system; or providing a reliable foundation for conducting subsequent research. Interventions should be selected to ensure that the design will realize social value and that the data are generalizable.21,2241 Third, the study design must realize the research objectives while neither denying health-care services that participants are otherwise entitled to nor requiring services that are not feasible to deliver in the context.37,38,42 However, studies can be ethically designed yet not provide a service or intervention individuals are entitled to under certain, restrictive conditions.41,43- 45 Specifically, it is ethical to use placebo or less than the diagnostic tests or treatments to which individuals are entitled when two conditions are fulfilled: (1) there is a methodologically compelling reason to do so, and (2) there is only minimal chance of serious harm—such as suffering irreversible morbidity or disability, or reversible but serious injury.41,43-45 Determining entitlement to medical services in studies is challenging because entitlements differ among countries, and may differ among groups within a country.46,47 Even in wealthy countries, participants are not entitled to every available or effective medical service, because justice necessitates establishing priorities for the distribution of scarce resources.460 For instance, some developed countries may not guarantee expensive drugs when inexpensive but more inconvenient yet effective drugs are available. Similarly, it is widely accepted that cardiac research conducted in developing countries need not be designed to require a coronary care unit because participants would not necessarily be entitled to this service under a just distribution of scarce resources in those countries.37,38,42,46,49 Conversely, in a study evaluating interventions to reduce mortality from cerebral malaria conducted in rural settings in which travel to hospitals is impracticable, provision of bed nets may be part of a valid design even if participants may not otherwise have them.5° However,

II

111111

9 Research with Humans

even if the study's objective is deemed socially valuable, especially to the enrolled participants' community, it is not ethically necessary to provide more comprehensive interventions beyond those to which participants are entitled, especially interventions that may not be feasible and sustainable. Doing so may even be unethical if it undermines the scientific objectives or makes the results irrelevant to the enrolled participants' community. Finally, the study must be designed in a way that is practically feasible given the social, political, and cultural environment in which it is being conducted.' Ensuring feasibility might require extensive community education and outreach as well as sustainable improvements to the health-care infrastructure, such as training of personnel, construction of additional facilities, or provision of an affordable drug. Feasibility also requires that it be possible tc achieve the sample size in a reasonable time frame. Fair Participant Selection

Historically, populations that were poor, uneducated, or powerless to defend their own interests were targeted for high-risk research, whereas promising research was offered to more privileged individuals.15,34,52 Fair selection of participants requires that the research objectives be the primary basis for determining eligibility.4,15,21,22,34 Once a target group is identified based on scientific objectives, considerations of minimizing risk, enhancing benefits, minimizing vulnerability, feasibility, as well as facilitating collaborative partnership, become determinative.22 Factors extraneous to the objectives, risks, benefits, and feasibility of conducting the research should not be the basis for selecting target communities or excluding individuals or communities.4,15,22,34 Four benchmarks are necessary to fulfill the principle of fair participant selection. First, the study population should be selected to ensure valid science .21,22,34,53 Scientific reasons for choosing a particular group of individuals or a community might be high prevalence or incidence of a disease, the magnitude of harms caused by the disease, high transmission rates of an infection, special drug resistance patterns, deprived social circumstances that increase susceptibility to a disease, or particular combinations of diseases. Social status that is irrelevant to the research objectives should not influence selection. Scientific considerations alone, however, will usually underdetermine which community or individuals are selected. Second, selecting participants in a way that minimizes risk is essentia1.54 For instance, in selecting a target population for an Hiv vaccine study, the extent to which a community protects Hiv-infected persons against discrimination and provides treatment for opportunistic

infections are important considerations to minimize risk Similarly, individuals with high creatinine clearance may be appropriately excluded from a trial of a potentially renal toxic drug in order to reduce risk. Third, individuals should be selected in order to enhance both the social value of the research and the possibility of benefits to participants.22'55-57 For example, assuring an adequate number of women in a study of a disease largely affecting women enhances benefits to women. Selecting individuals who are able to comply with the study requirements will enhance the chances that they will benefit from the intervention and that the study will yield valid data. Communities should be selected in which a collaborative partnership can be developed and in which social value can be realized. Consequently, it is preferable to select communities that have, or can establish, a system for identifying legitimate representatives and that will share responsibility for planning and conducting the study and ensuring that results are implemented through health system improvements or additional research. Finally, factors such as cognitive ability, age, clinical status, familial relationships, social marginalization, political powerlessness, and economic deprivation should be considered in order to determine the vulnerability of individuals or groups.58 For instance, if health policy makers suggest a particular group for research participation, the researchers should determine whether the group has been selected for good reasons, such as a high incidence of disease, or because of social subjugation. If scientifically appropriate individuals or groups are identified as vulnerable, specific safeguards to protect the population should be implemented, such as consent monitoring or independent capacity assessment, independent clinical monitoring, ensuring confidentiality, and ensuring that potential research participants are free to decline joining the study. Favorable Risk-Benefit Ratio

Like life itself, all research entails some risks. However, clinical research typically should offer individual participants a favorable net risk-benefit ratio 21,22,34 In cases in which potential risks outweigh benefits to individual participants, the social value of the study must be sufficient to justify these net risks.4'59 Because clinical research involves drugs, devices, and procedures about which there is limited knowledge, uncertainty about the degree of risks and benefits is inherent. And the uncertainty is greater in early phase research. The principle of a favorable net risk-benefit ratio requires fulfilling three benchmarks. First, the risks of the research should be delineated and minimized. Researchers should identify the type, probability, and

Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research 501

magnitude of the risks of the research. The risks are not limited to physical risks, but should also encompass potential psychological, social, and economic risks. To the extent possible, the assessment of risks should be based on available empirical data, not intuition or speculation. Within the context of good clinical practice, these risks should be minimized "by using procedures which are consistent with sound research design and which do not unnecessarily expose subjects to risk, and whenever appropriate, by using procedures already being performed on the subjects for diagnostic or treatment purposes."5 In addition, research procedures should be performed by trained and competent individuals who adhere to the standards of clinical practice.3 Second, the type, probability, and magnitude of the benefits of the research should be identified. The benefits to individual participants, such as health improvements, are relevant. The specification of potential benefits to individual participants should consider only health-related potential benefits derived from the research intervention itself.21,22,34 The benefits to society through the generation of knowledge are assumed if the research is deemed to be of social value and scientifically valid. Secondary benefits, such as payment, or adjunct medical services, such as the possibility of receiving a hepatitis vaccine not related to the research, should not be considered in the risk-benefit evaluation; otherwise simply increasing payment or adding more unrelated services could allow the benefits to justify even the riskiest research.22,6° Furthermore, although participants in clinical research often receive some health services and benefits, the purpose of clinical research is not the provision of health services. Services directly related to clinical research are necessary to ensure scientific validity and to protect the well-being of the individual participants. As a matter of general beneficence, consideration should be given to enhancing benefits to participants and their community, especially when such benefits can be provided easily and will not compromise the scientific validity of the study. However, such enhancements of benefits are not to be considered in the assessment of the risk-benefit ratio—or even of the social value—of the research study itself. Third, the risks and potential benefits of the clinical research interventions to individual participants should be compared. In general, the more likely and/ or more severe the potential risks, the greater in likelihood and/or magnitude the prospective benefits must be; conversely, research entailing potential risks that are less likely and/or of lower severity can have more uncertain and/or circumscribed potential benefits. Importantly, this comparison of risks and benefits should take

into account the context in which the participants live and the risks they actually face. The underlying risks of a particular disease can vary because of differences in incidence, drug resistance, genetic susceptibility, or social or environmental factors. When participants confront a higher risk of disease, riskier research may be justifiable.61 Similarly, the net risk-benefit ratio for a particular study may be favorable in communities in which the social value of the research is high, yet may be unfavorable in communities in which the potential value is lower. When potential benefits to participants from the research are proportional to the risks they face, then the additional social value of the research, assured by the fulfillment of the value and validity requirements, implies that the cumulative benefits of the research outweigh its net risks. The notions of "proportionality" and potential benefits "outweighing" risks are metaphorical!' Yet the absence of a mathematical formula to determine when the balance of risks and potential benefits is proportionate does not connote that such judgments are inherently haphazard or subjective. Instead, assessments of risks and potential benefits to the same individuals can appeal to explicit standards, informed by existing data on the potential types of harms and benefits, their likelihood of occurring, and their longterm consequences.4 Evaluations of the quality of books are not quantifiable either, but neither are they merely matters of subjective taste; comparing the quality of Shakespeare or Dostoevsky with Danielle Steel entails judgments based on shared standards that can be justified to others. Similarly people routinely make discursively justifiable intrapersonal comparisons of risks and benefits for themselves, and even for others, such as children, friends, and employees without the aid of mathematical formulae.62 Finally a more complex evaluation is necessary when clinical research presents no or few potential benefits to individual participants, such as in Phase I safety and pharmacokinetic studies, and even in some epidemiology research, or when the risks outweigh the potential benefits to individual participants. In this case, a more complex evaluation, what Charles Weijer calls a "'risk-knowledge' calculus," is necessary.' This calculus assesses whether the societal benefits in terms of knowledge gained justify the "excess" risks to individual participants.63 Determining when potential social benefits outweigh net risks to individual participants requires interpersonal comparisons that are conceptually and practically more difficult than intrapersonal comparisons.62 However, policy makers are often required to make these kinds of comparisons,

II :ill

111111

9 Research with Humans

for example, when considering whether pollution and its attendant harms to some people are worth the potential benefits of higher employment and tax revenues to others. There is no settled framework for how potential social benefits shculd be "balanced" against individual risks. Indeed, the appeal to a utilitarian approach of maximization, as in cost-benefit analysis, is quite controversial both morally and because many risks and benefits of research are not readily quantifiable on commensurable scalzs.64-66 Nevertheless, these comparisons are made,67 and regulations mandate that investigators and research review committees make them with respect to clinical research.4'5 When research risks exceed the combination of potential medical benefits to individuals and the benefit of useful knowledge to society, clinical research is not justifiable.

Independent Review Independent ethical review of all clinical research protocols is necessary for two reasons: (1) to minimize concerns regarding researchers' conflicts of interest and (2) to ensure public accountability.21'22 Investigators inherently have multiple, legitimate interests—interests to conduct high quality research, to complete the research expeditiously, to protect research participants, to obtain funding and advance their careers, and so forth.'" Even for well-intentioned investigators, these diverse interests can generate conflicts that may unwittingly distort or undermine their judgments regarding the design, conduct, and analysis of research, as well as adherence to ethical requirements.190-7° Wanting to complete a study quickly may lead to the use of questionable scientific methods or to the use of readily available participants rather than fairer participant selection criteria; enthusiasm for and commitment to the research project may lead to overemphasis of potential benefits and under emphasis of potential harms to participants. Independent review by individuals unaffiliated with the clinical research study helps to minimize the potential impact of such conflicts of interest.21,22,34,71 In this way, in dependent reviewers can assure potential research participants that the study they are considering is ethical—that is, it will generate socially valuable information and the risk-benefit ratio is favorable. Independent review of clinical research is also important for a second, less emphasized, reason: social accountability.21 Clinical research imposes risks on participants for the benefit of society. An independent review of a study's compliance with ethical requirements assures members of society that people who enroll in trials will be treated ethically. Based on this review,

members of society can have confidence that they will not benefit from the exploitation of other humans. Four benchmarks help in fulfilling this principle. First, procedures established by law and regulation should be followed. Research has not revealed the best mechanism to conduct independent review.72 Consequently, the actual review mechanisms are usually determined by laws and regulations that vary both internationally and locally. For instance, some countries and institutions separate scientific and ethical review, whereas others integrate scientific and ethical assessments into a single review. Similarly, some countries have ethics review committees composed only of laypersons, whereas others have committees dominated by medical scientists and physicians. Nevertheless, prevailing laws and regulations establish the standards that should be followed for independent review They should be amended as better processes are identified. Second, whatever the process, the review must be independent and competent. Members of the review committees must be free of any conflicts with the researchers or the research study. The reviewers should not be collaborators on the research or with the researchers, and should not have any financial interests in the outcomes of the study. Similarly, reviewers should be excluded from the review if they have other conflicting interests, such as responsibility for the financial interests of the institution in which the research is conducted, that might preclude them from evaluating the protocols according to ethical principles and without bias. Similarly, the reviewers should have sufficient expertise—or be able to access advice—in the scientific, clinical, and statistical areas necessary to assess the research protocol. Training in research ethics for the reviewers may be necessary. Third, the review should be transparent. This is especially important in multinational research in which differences in culture, practices, and understandings may yield different judgments. One fundamental aspect of transparency is that the reasons for decisions of the independent review committee are explained. This allows observers to assess whether the reasons are appropriate and relevant considerations have been addressed. Finally given the increasing complexity of research, multiple independent reviews frequently occur 73-75 Multiple independent reviews may seem to be required by law or regulation for multisite studies or studies conducted by investigators from multiple institutions. Importantly, however, the ethical principle of independent review does not require multiple reviews.76 The only requirement is that the reviewers competently and independently assess relevant scientific and ethical considerations. Indeed, multiple reviews may have no added

EmanueVWendler/Grady: An Ethical Framework for Biomedical Research

value or may even be counterproductive, by taking time and requiring adjudication without added protections.72 Such situations are unethical—resources are expended that produce no value or even waste value.40 If there is disagreement among such reviews, it is important to clarify its nature. Disagreement may reflect different ways of balancing various principles and benchmarks, or the appropriateness of different ways of fulfilling them. That is, disagreement might reflect how the ethical principles are met, rather than whether they are met .77 Conflicts may also arise because of different guidelines or regulatory requirements, which themselves may not have good ethical justification or may be insensitive to particular cultural or social circumstances.78 Only rarely are there fundamental disagreements about whether ethical principles and benchmarks are fulfilled. Unfortunately, there is no widely accepted procedure for adjudicating such conflicts. In practice, the requirements specified by the sponsor's review board are often determinative. This contravenes the principle of collaborative partnership and the notion that the community that assumes the risks of the research should make the assessment about the research protocol." Informed Consent

No requirement has received as much explication as informed consent. The purpose of informed consent is to show respect for the autonomy of individuals.4,6,15,25,34,80-85 To enroll individuals in clinical research without their authorization is to treat them merely as a means to purposes and ends they may not endorse or even know about, denying them the opportunity to choose what projects they will pursue and subjecting them to Kantian-type exploitation.23-25 By allowing individuals to decide if—and how—they contribute to research, informed consent respects persons and their autonomy.4,25 Valid informed consent requires that the consenting person has the capacity to understand and make decisions, receives relevant information about the research study, understands that information, and consents voluntarily and without coercion.4,15,25,34,80-84 Each of these elements is necessary to ensure that individuals make rational and free determinations of whether the research trial is consonant with their interests.86 Seven benchmarks are necessary to fulfill the principle of informed consent. First, recruitment procedures and incentives for participants should be consistent with cultural, political and social practices of the potential participants. In some communities, compensation for participation in research may be expected, whereas in others, it may be considered offensive. The appropriate form and

level of compensation depends upon the local economic and social context.°7 Although concerns about undue inducement are frequently raised,45,84 high potential social value and a favorable risk-benefit ratio—implying minimal net risks to the participants—dispel these concems.88-91 Indeed, worry about undue inducement could reduce compensation and some other benefits for participants and host communities. Paradoxically, balancing fair compensation and undue inducement may result in less compensation for members of impoverished communities and raise the specter of exploitation.26.88 Second, both written and verbal disclosure of information should be sensitive to participants' culture and context. Disclosures should use the language, culturally appropriate idioms, and analogies of the prospective participants at a level they can understand. This entails a need for collaborative partnership. After disclosure, investigators should feel confident that participants understand the information and are consenting without any pressure or major misconceptions. In some cases, a formal assessment of understanding, monitoring of the consent process, or independent assessment of participants' capacity to consent may be warranted.92 Third, the disclosure of information relevant to the research study must be complete and accurate, but not overwhelming. Providing less than complete and accurate information raises concerns about potential deception of participants. However, complete information does not imply lengthy or exhaustive disclosure forms detailing every aspect of the research study, which may be overwhelming to the participants. Indeed, shorter, more focused forms, without repetition and boilerplate disclosures, may be more effective.93 Disclosure forms must balance completeness with not being overwhelming. Fourth, some research entails enrollment of individuals unable to consent because of their age, permanent mental incapacity, an acute loss of mental functions, or other reasons. In these cases, researchers must have a strategy for obtaining permission from legally authorized representatives of the potential participants.15,83,84,94-99 In some cases, "spheres of consent" ranging from spouses to heads of households to school principals to village elders or community leaders may be required before researchers can invite individual participation.30.100,'0i With a few exceptions, such as emergency research, it is unacceptable to supplant individual consent of competent adults by family or community consent.1°2 The family or community gives permission only to approach individuals. When family or community permission to approach individuals is reasonable, special care should be given to assure that the individual can still refuse participation—that is, that there is no coercion.

II

JII

11111.1

9 Research with Humans

Sixth, researchers should utilize consent procedures that are acceptable within the local context, while ensuring that an independent observer could verify voluntary participation by the individuals. For instance, US regulations require a written signature.5 In many cases, this is an acceptable and efficient way to document consent authorization. However, in some cases, because of limited literacy or cultural differences, such requirements may be inappropriate and unethical." Alternative methods to express consent, such as handshakes, embracing, or sharing a meal, are known." Appropriate alternative procedures for documenting informed consent might include tape recordings or witnessed written documentation of these methods of consent. Finally, special attention must be given to ensure that individuals are aware of their right to, and are actually free to, refuse to participate or to withdraw from research. A key element of informed consent is the ability to refuse or withdraw participation without penalty.m3 Prorating offered compensation and other research-related benefits may help to obviate possible familial or community coercion or retribution. Respect for Participant;

The ethical conduct of clinical research does not end when informed consent is obtained.21,22,104 Researchers have ongoing obligations to treat individuals with respect from the time they are approached—even if they refuse enrollment—throughout their participation and even after their participation ends. Respecting potential and enrolled participants entails multiple activities. First, and arguably most important, this principle requires monitoring the health and well-being of participants, and intervening to prevent or treat harms that might result from the adverse reactions, untoward events, or changes in clinical status associated with the research.10" In some cases, research studies need to include procedures to adjust drug doses and even withdraw study participants because of adverse events. Furthermore, specific stopping rules may be necessary if excessive adverse events or benefits are identi:ied. Second, pledges of confidentiality should be honored and procedures to protect confidentiality implemented. Such procedures ir.clude securing databases, locking file cabinets containing data, coding specimens and data forms, as well as interviewing participants in private spaces where they cannot be overheard. In addition, it is important to alert participants that despite researchers' best efforts, absolute confidentiality cannot be guaranteed.

Third, respect includes permitting participants to change their minds, to decide that the research does not comport with their interests or preferences, and to withdraw without penalty. Fourth, as new information about the impact of the intervention or about the participant's clinical condition is gained during the course of the research, respect requires providing this new information to the participants. Researchers should also develop explicit strategies to inform participants and host communities of the results of the research. Having participated in research and assumed risks, the participants and host communities have a right to know what was found and its implications for public health and health-care policies. Finally, plans should be made regarding the care of participants when the trial is over. In some cases, this may simply involve referral to a primary care provider. In other cases, this may require researchers to find creative strategies for providing access to treatments benefiting the participants, even when these interventions are unlicensed. Characteristics of the Principles

The eight general principles and the benchmarks delineate a systematic and comprehensive way of assessing the ethics of particular clinical research.21,22 They provide a coherent and organized way for researchers, ethics reviewers, participants, and others to evaluate a research protocol and to determine whether it fulfills ethical standards. They should not be seen as adding ethical requirements, but rather distilling and coherently articulating the ethical norms underlying much of the prevailing guidance. These principles and benchmarks offer a more organized and systematic delineation of what many researchers, ethics reviewers, and others already do. Importantly, these principles are not independent of all other ethical principles. They operate within and presume compliance with more general moral norms, such as honesty and promise keeping.22 Similarly, these principles focus on what is required to evaluate research studies, not on the enforcement or proper conduct of the research itself. Having ethical researchers is important for implementation of the framework but not a requirement for evaluating the research protocol. Determining what is ethical and what needs to be enforced must be done prior to and should not be confused with how to implement an ethical protocol or to enforce the requirements.21,22 These eight principles are necessary. The presumption is that they must all be fulfilled for a research protocol to be ethical. There is no picking and choosing.

EmanueVWendler/Grady: An Ethical Framework for Biomedical Research

However, in specific cases, such as emergency research, informed consent may be legitimately waived. These principles are justified by ethical values that are widely recognized and accepted, that reasonable people would want to be treated in accordance with—avoidance of exploitation, the just distribution of benefits and burdens, beneficence, respect for persons, and so forth.105,106 These requirements are precisely the types of considerations that would be invoked to justify clinical research if it were challenged. The benchmarks provide more practical considerations for discerning satisfaction of the general principles. The principles are sufficient. Fulfilling these eight principles means the research is ethical. Failing on any one principle—except for waiving informed consent in specific cases, in which waiving consent must be justified—makes the research unethical. The proposed benchmarks, however, may not be sufficient, and may need revision with experience and time. They certainly provide a useful first estimation of the kind of specific elements that need to be fulfilled. These eight principles are universal; they apply in all countries and contexts, regardless of sponsorship. The principles are general statements of value; they must be elaborated by traditions of interpretation and require practical interpretation and specification. The benchmarks offer a first level of specification, indicating how to fulfill these principles. However, the details of this specification will inherently be context and culture dependent. This does not make them relativistic or less universal. It simply recognizes that applying ethical principles in the world requires taking facts into account, and these facts depend upon the context. Moral arguments take place in context, and they therefore depend at least implicitly on matters of fact, estimates of risk, suppositions about feasibility, and beliefs about human nature and social processes. . . . Even those who rely on what they regard as universal moral principles do not presume that their practical conclusions

References

S. International Guidelines on Bioethics. Geneva, Switzerland: European Forum on Good Clinical Practim/cioms, 1998. 2. The Nuremberg Code. In Trials of War Criminals Before 1.

are independent of reliable facts and plausible assumptions about particular societies. The arguments begin from where we are, and appeal to those with whom we now live. This is why moral relativism is seldom as important an issue in practical as it is in theoretical ethics.107 Importantly, that there are eight principles suggests that the ethics of research is complex. Adherence to a single ethical principle rarely provides a complete solution; most situations implicate multiple principles.48,62,64.105,107-110 Consequently, the various principles and benchmarks will sometimes conflict. What is fair participant selection could at times increase risk; what is required for informed consent may sometimes compromise scientific validity. Unfortunately, there is no simple algorithm for determining how to balance or weigh these principles when they conflict. Different researchers and communities will balance the principles in different ways, some emphasizing informed consent, others the importance of minimizing risks or enhancing social value. Ignoring or rejecting basic principles in designing or conducting a research study could render it unethical. Conversely, accepting the principles and benchmarks, yet disagreeing about how to balance them in a particular case, highlights the intricacies of ethical judgments entailing multiple considerations. Disagreement on the balancing of the various benchmarks does not necessarily make one assessment ethical and the other unethical. Rather, it reflects different, but perhaps legitimate, ways of resolving competing ethical claims.m In fact, this framework can help narrow disagreements and elucidate the different underlying views. When conflicts between principles and benchmarks occur, or when different groups weigh the principles differently the important point is to be clear about the reasons for the evaluation and the differences. Ultimately, a thoughtful process of balancing ethical considerations can be as important as any particular judgment in the effort to ensure that research is conducted ethically.

3.

ical Principles for Medical Research Involving Human Subjects. Tokyo, Japan: WMA; October 2004. [Online] 2004.

Fluss,

the Nuremberg Military Tribunals Under Control Council Law No. 10. Volume 2. Washington, DC: US Government

Printing Office; 1949:181-2. [Online] Available: httpl/ ohsr. od.nih.gov/guidelines/nuremberg.html.

World Medical Association. Declaration of Helsinki: Eth-

4.

Available: http://wwwwma.net/e/policy/b3.htm. The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. Washington, DC: Department of Health,

Education and Welfare;

DHEW

Publication OS

505

11

11

111111

9 Research with Humans 78-0012 1978. [Online: April 18, 1979. Available: 17. http i/www. hhs gov/oht p/hum ansu bj e c ts/g uid a nce/ belmont.htm. 5. Department of Health and Human Services, National Institutes of Health, and Office for Human Research Protections. The Common Rule, Title 45 (Public Welfare), 18. Code of Federal Regulations, Part 46 (Protection of Human Subjects). [Online) June 23, 2005. Available: http://www 19. .hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm. 6. Council for International Organizations of Medical Sciences, in collaboration with the World Health Organiza- 20. tion. International Ethical Guidelines for Biomedical Research Involving Human Subjects. Geneva, Switzerland: CIOMS and WHO; 2002. [Online) November 2002. Available: http://wwwcioms.ch/framc_guidelines_nov_2002.htm. 21. 7. Council of Europe, Directorate of Legal Affairs Convention for the Protection of Human Rights and Dignity of the Human Being 22.

With Regard to the Applicatior of Biology and Medicine: Convention on Human Rights and Biomedicine. Strasbourg, France: Council of Europe; 1997. [Online) April 4, 1997. Available: 23. httpl/conventions.coe.int/tnaty/en/treaties/html/164.htm. 8. InternationalConf erence on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human 24. Use. The ICH Harmonised Tripartite Guideline-Guideline for Good Clinical Practice. Geneva: ICH; 1996. [Online) Avail- 25. able: http//wwwichorg/LOB/media/MEDIA482.pdf. 9. UK Medical Research Council. Guidelinesfor Good Clinical 26. Practice in Clinical Trials. London, England: MRC; 1998. 27. [Online) Available: http://www.mrc.ac.uk/pdf-ctg.pdf. 10. Uganda National Council of Science and Technology (uNCST).

Guidelines for The Conduct of .qealth Research Involving Human Subjects in Uganda. Kampala, Uganda UNCST, 1998. 11. Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council, 28. and the Social Sciences and Humanities Research Council. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. [Online) 1998 (with 2000, 2002 and 2005 amendments). Available: 29. http://www.pre.ethics.gc.ca/english/policystatement/ policystatement.cfm. 12. National Health and Medical Research Council, Com- 30. monwealth of Australia. National Statement on Ethical Conduct in Research Involving Humans. Canberra: NHMRC, 31. 1999. [Online) Available: http://www.nhmrc.gov.au/ publications/ files/e35.pdf 13. Department of Health and Human Services, Office for 32. Human Research Protecticns. International Compilation of Human Subject Research Protections. 2nd edn [Online) October 1, 2005. Available: http://www.hhs.gov/ohrp/ 33. international/HSPCompilation.pdf. 14. Levine, C. Has AIDS changel the ethics of human subjects research? Law Medicine and Health Care 1988, 16:167-73. 34. 15. Advisory Committee on Human Radiation Experiments Final Report of the Advisory Committee on Human Radiation 35. Experiments. New York, N.Y.: Oxford University Press 1996. 16. Macklin, R. After Helsinki: Unresolved issues in international research. Kennedy Institute of Ethics Journal 36. 2001, 11:17-36.

Council for International Organizations of Medical Sciences, in collaboration with the World Health Organization. International Ethical Guidelines for Biomedical Research Involving Human Subjects. Geneva, Switzerland: CIOMS and WHO, 1993. Forster, H.P., Emanuel, E., Grady, C. The 2000 revision of the Declaration of Helsinki: A step forward or more confusion. Lancet 2001, 358: 1449-53. Thompson D.F. Understanding financial conflicts of interest. New England Journal of Medicine 1993, 329: 573-6. American Society of Clinical Oncology. Revised Conflict of Interest Policy (adopted on November 7, 2002, by the American Society of Clinical Oncology). Journal of Clinical Oncology 2003, 21: 2394-6. Emanuel EJ., Wendler D., Grady C. What makes clinical research ethical? JAMA 2000, 283: 2701-11. Emanuel E.J., Wendler D., Killen J., Grady C. What makes clinical research in developing countries ethical? The benchmarks of ethical research. Journal of Infectious Diseases 2004, 189: 930-7. Wood A.W. Exploitation. Social Philosophy and Policy 1995, 12: 135-58. Buchanan A. Ethics, Efficiency and the Market. Lanham, Md.: Rowman & Littlefield, 1985. Beauchamp T.L., Childress J. The Principles of Biomedical Ethics. 5th edn. New York: Oxford University Press, 1996. Wertheimer A. Exploitation. Princeton, NJ: Princeton University Press, 1996. Participants in the 2001 Conference on Ethical Aspects of Research in Developing Countries. Moral standards for research in developing countries: From "reasonable availability" to "fair benefits." Hastings Center Report 2004, 34(3): 17-27. Daniels, N., Bryant, J., Castano, R.A., et al. Benchmarks of fairness for health care reform a policy tool for developing countries. Bulletin of the World Health Organization 2000, 78: 40-50. Daniels, N., Light D.W., Caplan R.L. Benchmarks of Fairness for Health Care Reform. New York: Oxford University Press, 1996. Weijer, C., Emanuel, E.J. Protecting communities in biomedical research. Science 2000, 289: 1142-4. Participants in the 2001 Conference on Ethical Aspects of Research in Developing Countries Fair benefits for research in developing countries. Science 2002, 298: 2133-4. Diallo, D.A., Doumbo, O.K., Plowe, C.V., et al. Community permission for research in developing countries. Clinical Infectious Diseases 2005, 41: 255-9. Vanderpool, H.Y., ed. The Ethics of Research Involving Human Subjects: Facing the 21st Century. Frederick, Md.: University Publishing Group, 1996. Levine, R.J. Ethics and Regulation of Clinical Research, 2nd edn. New Haven, CT: Yale University Press, 1988. Freedman, B. Scientific value and validity as ethical requirements for research. IRB: A Review of Human Subjects Research 1987, 9(6): 7-10. Black, N. Evidence based policy: Proceed with care. British Medical Journal 2001, 323: 275-8.

Emanuel/Wendler/Grady: An Ethical Framework for Biomedical Research

37. Grady, C Science in the service of healing. Hastings 57. Center Report 1998, 28(6): 34-8. 38. Crouch, R.A., Arras, J.D. AZT trials and tribulations. Hastings Center Report 1998, 28(6): 26-34. 58. 39. President's Commission f or the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Summing Up. Washington, DC: US Government Printing Office, 1983. [Online] Available: httpi/www.bioethics .gov/reports/past_commissions/summing_up.pdf. 40. Rutstein, D.D. The ethical design of human experiments. 59. In Freund, PA., ed. Experimentation with Human Subjects. New York: George Braziller, 1970: 383- 401. 60. 41. Emanuel, E.J., Miller, F.G. The ethics of placebocontrolled trials-A middle ground. New England Journal of Medicine 2001, 345: 915-9. 61. 42. Freedman, B. Placebo-controlled trials and the logic of clinical purpose. IRB: A Review of Human Subjects Research 62. 1990, 12(6): 1-6. 43. Temple, R., Ellenberg, S.S. Placebo controlled trials and 63. active control trials in the evaluation of new treatments. I. Ethical and scientific issues. Annals of Internal Medicine 2000, 133: 455-63. 64. 44. Wendler, D., Emanuel, E.J., Lie, R.K. The standard of care debate: Can research in developing countries be both ethical and responsive to those countries' health 65. needs? American Journal of Public Health 2004, 94: 923-8. 45. Nuffield Council on Bioethics. The Ethics of Research Related to Healthcare in Developing Countries. London, England: Nuffield Council on Bioethics, 2002. [Online] Available: 66. httpi/nuffieldbioethics.org/fileLibrary/pdf/errhdc_f ull report001.pdf. 46. Emanuel, E.J. The Ends of Human Life. Cambridge, MA: 67. Harvard University Press, 1991: chapters 4 and 5. 47. Daniels, N., Sabin, J.E. Setting Limits Fairly. New York Oxford University Press; 2002: chapter 2. 68. 48. Walzer, M. Spheres of Justice. New York: Basic Books, 1983. 49. Faden, R.R., Kass, N. HIV research, ethics and the developing 69. world. American Journal of Public Health 1998, 8a 548-50. 50. Kidane, G., Morrow, R.H. Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: A ran- 70. domized trial Lancet 2000, 356: 550-5. 51. Bloom, B.R. The highest attainable standard: Ethical issues in AIDS vaccines. Science 1998, 279: 186-8. 71. 52. Lederer, S.E. Subject to Science. Baltimore, Md.: Johns Hopkins University Press, 1995. 53. National Institutes of Health. NIH policy and guidelines on the inclusion of children as participants in research involving 72. human subjects. [Online] March 6, 1998. Available: http:// grants2.nihgov/grants/guide/notim-filestriot98-024.htmL 54. Weijer, C., Fuks, A. The duty to exclude: Excluding people at undue risk from research. Clinical and Investi- 73. gative Medicine 1994, 17: 115-22. 55. DeBruin, D. Justice and the inclusion of women in clinical studies. Kennedy Institute of Ethics Journal 1994, 4: 117-46. 74. 56. Mastroianni, A.C., Faden, R.R., Federman, D.D. Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies. Washington, DC: National 75. Academies Press 1994.

Weijer, C. Evolving issues in the selection of subjects for clinical research. Cambridge Quarterly of Healthcare Ethics 1996, 5. 334-5. Kipnis, K. Vulnerability in research subjects: A bioethical taxonomy. In National Bioethics Advisory Commission. Ethical and Policy Issues in Research Involving Human Participants, Vol. II. Rockville, Md.:NBAC, 2001. [Online] Available: http://www.georgetown.eduhrsearch/nrcbl/nbac/human/ overv012.pdf. Weijer, C. The ethical analysis of risk. Journal of Law, Medicine and Ethics 2000, 28: 344-61. Freedman, B., Fuks, A., Weijer, C. Demarcating research and treatment: A systematic approach fbr the analysis of the ethics of clinical research. Clinical Research 1992, 40: 653-60. Weijer, C. The future of research into rotavirus vaccine. British MedicalJournal 2000, 321: 525-6. Anderson, E Value in Ethics and Economics. Cambridge, MA. Harvard University Press, 1993: chapter 9. Weijer, C. Thinking clearly about research risks: Implications of the work of Benjamin Freedman. IRB: A Review of Human Subjects Research 1999, 21(6): 1-5. Sen, A., Williams, B. Introduction. In Sen, A., Williams, B., eds., Utilitarianism and Beyond. New York: Cambridge University Press, 1982: 1-22. Leonard, H., Zeckhauser, R. Cost-benefit analysis and the management of risk: Philosophy and legitimacy. In MacLean, D., ed. Values at Risk. Totowa, NJ: Rowman & Littlefield Publishers, 1986: 31- 48. Gold, M.R., Siegel, J.E., Russell, L.B., Weinstein, M.C., eds. Cost- Effectiveness in Health and Medicine. New York: Oxford University Press, 1996. Sen, A. Interpersonal comparisons of welfare. In Sen, A., Choice, Welfare, and Measurement. Cambridge, MA: Harvard University Press, 1982: 264-84. Relman, A.S. Economic incentives in clinical investigations. New EnglandJournal of Medicine 1989, 320: 933- 4. Porter, R.J., Malone, T.E. Biomedical Research: Collaboration and Conflicts of Interest. Baltimore, MD: Johns Hopkins University Press, 1992. Spece, R.G., Shimm, D.S., Buchanan, A.E., eds. Conflicts of Interest in Clinical Practice and Research. New York: Oxford University Press, 1996. National Commission for the Protection of Human Subjects of Bio- medical and Behavioral Research. Institutional Review Boards: Report and Recommendations. Washington, DC: US Government Printing Office, 1978. Emanuel, E.J., Wood, A., Fleischman, A., et al. Oversight of human participants research: Identifying problems to evaluate reform proposals. Annals of Internal Medicine 2004,141: 282-91. McWilliams, R., Hoover-Fong, J., Hamosh, A., et al. Problematic variation in local institutional review of a multicenter genetic epidemiology study. JAMA 2003, 290: 360-6. Roberts, L.M., Bowyer, L, Homer, C.S., Brown, M.A. Multicentre research Negotiating the ethics approval obstacle course [letter]. MedicalJournal of Australia 2004, 180: 139. Stair, T.O., Reed, C.R., Radeos, M.S., et al., for the MARC Investigators. Variation in Institutional Review Board

al

II

;iii

IIIIJJ

9 Research with Humans

responses to a standard protocol for a multicenter clinical trial. Academic Emergency Medicine 2001, 8: 636- 41. 76. Wood, A., Grady, C., Emanuel, E.J. Regional ethics organizations for protection of human research participants. Nature Medicine 2004, 10: 1283-8. 77. Mulholland, K., Smith, P.G., Broome, C.V., et al. A randomized trial of a Haemophilus influenzae type b conjugate vaccine in a developing country for the prevention of pneumonia-Ethical considerations. International Journal of Tuberculosis and Lung Disease 1999, 3: 749-55. 78. Wendler, D., Rackoff, J.E. :nformed consent and respecting individual autonomy: What's a signature got to do with it? IRB: Ethics and Human Research 2001, 23(3): 1- 4. 79. White, M.T. Guidelines for IRB review of international collaborative medical research: A proposal. Journal of Law, Medicine and Ethics 1999, 27: 87-94. 80. Freedman, B. A moral theory of informed consent. Hastings Center Report 1975, 5(4): 32-9. 81. President's Commission for the Study of Ethical Problems in Medicine and Biomedical Research. Making Health Care Decisions: Ethical and Legal Implications of Informed Consent in the Physician-Practitioner Relationship. Washington, DC: US Government Printing Office, 1982. [Online] Available http: //www.bioethics.gov/ reports/past_commissions/making_health_care_ decisions.pdf. 82. Donagan, A. Informed cons= in therapy and experimentation. Journal of Medicine ar. d Philosophy 1977, 2: 318-29. 83. Faden, R.R., Beauchamp, TL., with King, N.M.P. A History and Theory of Informed Consent. New York: Oxford University Press, 1986: chapters 5-9. 84. Berg, J.W., Applebaum, P.S., Lidz, C.W., Parker, L.S. Informed Consent: Legal Them), and Clinical Practice. 2nd edn. New York: Oxford University Press, 2001: chapters 2, 3, 11, and 12. 85. Dworkin, G. The Them), and Practice of Autonomy. New York. Cambridge University Press; 1988: chapters 1, 6, and 7. 86. Sreenivasan, G. Does informed consent to research require comprehension? Lancet 2003, 362:,2016-8. 87. Dickert, N., Grady, C What's the price of a research subject? Approaches to payment for research participation. New England Journal of Medicine 1999, 341: 198-203. 88. Emanuel, E.J., Currie, X.E., Herman, A„ on behalf of Project Phidisa. Undue inducement in clinical research in developing countries: Is it a worry? Lancet 2005, 366: 336- 40. 89. Emanuel, E.J. Ending concerns about undue inducement. Journal of Law, Medicine and Ethics 2004, 32: 100-5. 90. Harris, J. Wonderwoman and Superman: The Ethics of Human Biotechnology. New York: Oxford University Press, 1992: chapter 6. 91. Wilkinson, M., Moore, A. Inducement in research. Bioethics 1997, 11: 373-89. 92. Grisso, T., Applebaum, P.S. Assessing Competence to Consent to Treatment. New York: Oxford University Press, 1998.

93. Flory, J.H., Emanuel, EJ. Interventions to improve research participants' understanding in informed consent for research: A systematic review. JAMA 2004, 292: 1593-1601. 94. National Bioethics Advisory Commission. Research Involving Persons with Mental Disorders That May Affect Decisionmaking Capacity. Rockville, MD: NBAC, 1998. [Online] Available: http://www.georgetown.edu/research/nrcb1/ nbac/capacityfrOC.htm. 95. Buchanan, A.E., Brock, D.W. Deciding for Others: The Ethics of Surrogate Decision Making. New York: Cambridge University Press, 1990: chapter 2. 96. Dresser, R. Mentally disabled research subjects: The enduring policy issues. JAMA 1996, 276: 67-72. 97. Michels, R. Are research ethics bad for our mental health? New EnglandJournal of Medicine 1999, 340: 959-61. 98. Capron, A.M. Ethical and human-rights issues in research on mental disorders that may affect decisionmaking capacity. New England Journal of Medicine 1999, 340: 1430- 4. 99. Wendler, D., Prasad, K. Core safeguards for clinical research with adults who are unable to consent. Annals of Internal Medicine 2001, 135: 514-23. 100. Weijer, C., Goldsand, G., Emanuel, E.J. Protecting communities in research: Current guidelines and limits of extrapolation. Nature Genetics 1999, 23: 275-80. 101. Macaulay, A.C., et al. Participatory research with native community of Kahnawake creates innovative code of research ethics. Canadian Journal of Public Health 1998, 89: 105-8. 102. IJsselmuiden, C.B., Faden, R.R. Research and informed consent in Africa: Another look. New England Journal of Medicine 1992, 326: 830- 4. 103. Karim, Q.A., Karim, S.S.A., Coovadia, H.M., Susser, M. Informed consent for HIV testing in a South African hospital: Is it truly informed and truly voluntary? American Journal of Public Health 1998, 88: 637- 40. 104. Weijer, C., Shapiro, S., Fuks, A., Glass, K.C., Skrutkowska, M Monitoring clinical research: An obligation unfulfilled. Canadian Medical Association Journal 1995, 152: 1973-80. 105. Scanlon, T.M. What We Owe to Each Other. Cambridge, MA: Harvard University Press; 1999: chapters 1 and 8. 106. Kymlicka, W. Liberalism, Community and Culture. New York: Oxford University Press, 1989. 107. Gutmann, A., Thompson, D. Democracy and Disagreement. Cambridge, MA: Harvard University Press, 1996: chapter 1. 108. Nagel, T. The fragmentation of value. In Nagel, T., Mortal Questions. New York: Cambridge University Press, 1979: 128- 41. 109. Temkin, L. Inequality. New York: Oxford University Press, 1993: chapter 2. 110. Richardson, H.S. Specifying norms as a way to resolve concrete ethical problems. Philosophy and Public Affairs 1990, 19: 279-310.

Miller/Brody: A Critique of Clinical Equipoise

A Critique of Clinical Equipoise: Therapeutic Misconception in the Ethics of Clinical Trials

could two such radically divergent viewpoints exist, without research ethics being widely regarded as in a state of crisis?

Franklin G. Miller and Howard Brody

Therapeutic Misconceptions

The Hypericum Depression Trial Study Group published in 2002 the results of a randomized trial comparing hypericum (St. John's Wort), sertraline (Zoloft), and placebo in the treatment of major depression.' In the study, funded by the National Institutes of Health, 340 subjects from twelve participating centers were randomized to three trial arms for an eight-week period, with careful monitoring to assure that patients who worsened significantly or who became suicidal were removed from the study and received adequate treatment. Neither hypericum nor sertraline was found to be superior to placebo on the primary outcome measures. The authors noted, "From a methodological point of view, this study can be considered an example of the importance of including inactive and active comparators in trials testing the possible antidepressant effects of medications. In fact, without a placebo, hypericum could easily have been considered as effective as sertraline."2 What can we conclude about the ethics of this trial? One dominant viewpoint in research ethics would have prohibited the study. On this viewpoint, a randomized trial is ethical only in circumstances of "clinical equipoise"—a genuine uncertainty within the medical community as to whether (in this case) any of the three treatment arms are superior to the other two. No such uncertainty exists. Approximately twenty-five clinically available antidepressants, including sertraline, have been shown to be superior to placebo.3 Moreover, the majority opinion within psychiatry probably holds that sertraline is definitely superior to hypericum for major depression, even if hypericum has potential for the treatment of mild to moderate depression. But another widespread viewpoint would hold that the trial was ethically sound. Depressed individuals widely use hypericum, a "natural" agent, despite the lack of proven efficacy. Accordingly, a rigorous evaluation offered scientific, clinical, and social value. According to the report of trial results, the study was approved by institutional review boards (IRBs) at twelve sites, and subjects provided written informed consent. But if clinical equipoise is a basic requirement for ethical research, how could all these review boards be blind to the unethical nature of this trial? And how

The prevailing ethical perspective on clinical trials holds that physician-investigators can discharge their "therapeutic obligation" to patients in the context of randomized clinical trials (RCTs) as long as treatments being tested scientifically satisfy clinical equipoise. We contend that this ethical perspective is fundamentally flawed. An ethical framework that provides normative guidance about a practice should accurately characterize the practice. The prevailing ethical perspective fails this test: All sound ethical thinking about clinical research, and the regulatory framework for review of protocols for clinical investigation, depends on a basic distinction between research and therapy. But the claims in the prevailing ethical perspective on clinical trials conflate research and therapy. These claims are that the ethics of the physician—patient relationship must govern RCTs, that physicians who conduct these trials have a "therapeutic obligation" to patients enrolled in them, and that RCTs must be compatible with some form of equipoise. Certainly, investigators and ethicists recognize that clinical trials are scientific experiments, which differ from standard medical care. They also recognize that they are subject to regulatory requirements which do not apply to routine medical practice. However, the prevailing ethical framework views clinical trials through a therapeutic lens. The mainstream ethical approach to clinical trials attempts to have it both ways: to view the clinical trial as a scientific experiment, aimed at producing knowledge that can help improve the care of future patients, and as treatment conducted by physicians who retain fidelity to the principles of therapeutic beneficence and therapeutic non-maleficence that govern the ethics of clinical medicine. The doctrine of clinical equipoise has emerged as the bridge between medical care and scientific experimentation, allegedly making it possible to conduct RCTs without sacrificing the therapeutic obligation of physicians to provide treatment according to a scientifically validated standard of care. This constitutes a "therapeutic misconception" concerning the ethics of clinical trials, analogous to the tendency of patient volunteers to confuse treatment in the context of RCTs with routine medical care.4 As Paul Appelbaum has recently observed, "In fact, this confusion between the ethics of research

Source: From The Hastings Center Report 33 (2003) no. 3, 19-28. Reprinted with permission of the Hastings Center.

II

311

111111

510 9 Research with Humans

and of ordinary clinical care appears rampant in the world of clinical trials. "5 The therapeutic misconception in the ethics of clinical trials is reflected in the language commonly used within the clinical research enterprise. Clinical trials are often described as "therapeutic research," and investigators are regarC ed as having a "therapeutic intent." Research participants who are being studied because they have a medical condition under investigation are referred to as "patients" and investigators as "physicians" or "doctors," without qualification. To demonstrate our contention about the mainstream approach to the ethics of clinical trials, we will offer an intellectual reconstruction of some of the history of research ethics since the 1970s. This history is characterized by incoherence resulting from commitment to two incompatible positions, each approaching research ethics in a fundamentally different way. The therapeutic misconception about the ethics of clinical trials has emerged from the "similarity position," which argues that ultimately, the ethics of clinical trials rest on the same moral consithrations that underlie the ethics of therapeutic medicine. The "difference position" argues that the ethics of clinical trials must start with the realization that medical research and medical treatment are two distinct forms of activity, governed by different ethical principles. The reigning ethical paradigm for clinical trials has coexisted with clinical trials practice that departs from its guidance. Clinical equipcise, the cornerstone of the similarity position, rules out placebo-controlled trials whenever there is a proven effective treatment for the disorder under investigation.° However, TRBs have routinely approved such placebo-controlled trials. These two anomalies—unappreciated theoretical incoherence and conflict between the theoretical paradigm and the practice of ethical review of clinical trials—call for critical examination of the similarity position and the doctrine of clinical equipoise. The Distinction between Research and Therapy

In 1979, Robert Levine summarized "the most important achievements of the National Commission" for the Protection of Human Subjects of Biomedical and Behavioral Research in "correcting the conceptual and semantic errors that had undermined virtually all previous attempts to develop rational public policy on research involving human subjects."' Two portions of Levine's summary capture the essential ingredients of the difference position: recognizing the distinction between research and therapy and, accordingly,

abandoning the distinction between therapeutic and nontherapeutic research. Clinical research shares with medical care the fact that both are performed by physicians in clinical settings and both often use similar diagnostic and treatment interventions. When the commission began its work, physicians commonly regarded clinical research and medical therapy as inextricably connected. One authority quoted by Levine claimed that "Every time a physician administers a drug to a patient, he is in a sense performing an experiment." But the commission recognized the importance of determining the boundaries between routine medical practice and research. For Levine, the commission's conceptual breakthrough came with the realization that the physicians of the day were thinking about clinical research in the wrong way and that the boundary between research and therapy was clear rather than fuzzy. The commission came to hold that clinical research is fundamentally different from medical practice.° Clinical medicine aims at providing optimal medical care for individual patients. Ethically, it is governed by the principles of therapeutic beneficence and therapeutic nonmaleficence. Therapeutic beneficence directs physicians to practice medicine with primary fidelity to promoting the health of particular patients. According to therapeutic nonmaleficence, the risks of medical care to which a patient is exposed are to be justified by the prospect of compensating medical benefits for that patient. The physician uses scientific knowledge to care for the patient and engages in therapeutic experimentation with the aim only of finding optimal treatment. It is not part of the role of the physician in providing medical care to develop scientific knowledge that can help future patients. Clinical research, in contrast, is not a therapeutic activity devoted to the personal care of patients. It is designed for answering a scientific question, with the aim of producing "generalizable knowledge." The investigator seeks to learn about disease and its treatment in groups of patients, with the ultimate aim of improving medical care. Scientific interest in any particular patient concerns what can be learned that is applicable to other patients. In view of the nature and purpose of clinical research, the principles of beneficence and nonmaleficence applicable to clinical research lack the therapeutic meaning that guides their application to medical care. Clinical research is dedicated primarily to promoting the medical good of future patients by means of scientific knowledge derived from experimentation with current research participants—a frankly utilitarian purpose. A major reason for distinguishing research from therapy is to underscore that clinical research has an

Miller/Brody: A Critique of Clinical Equipoise

inherent potential for exploiting research participants.9 Exploitation also may occur in clinical medicine—venal physicians sometimes perform medically unnecessary procedures for the sake of profit, for example. Yet when physicians of integrity practice medicine, physicians' and patients' interests converge. The patient desires to regain or maintain health or to relieve suffering the physician is dedicated to providing the medical help that the patient needs. In clinical research, by contrast, the interests of investigators and patient volunteers are likely to diverge, even when the investigator acts with complete integrity. Patient volunteers, especially in clinical trials, typically seek therapeutic benefit, though they also may be motivated by altruism.1° Investigators are interested primarily in developing scientific knowledge about groups of patients. Regardless of investigators' motivations, patient volunteers are at risk of having their well-being compromised in the course of scientific investigation. Clinical research involves an inherent tension between pursuing rigorous science and protecting research participants from harm.il Historically, the ethical distinction between research and therapy emerged out of concern about exploitive abuses of patients in clinical research. Reflection on this dark history gave rise to a major development in the ethics of clinical research: the requirement for independent, prospective review and approval of research protocols.0 Prior independent review was considered necessary for clinical research because of the divergence between the interests of the investigator and the research participant. Self-regulation by physician-investigators could not be trusted in the research context to the same extent that self-regulation by physicians was appropriate in the therapeutic context. The basic rationale for prospective, independent research review depends on the distinction between research and therapy. The point of distinguishing research and therapy is not to make an invidious comparison, implying that clinical trials are more risky or ethically problematic than routine clinical practice. Indeed, there is some evidence that patients receive more favorable medical outcomes in many clinical trials,13 and clinical medicine is certainly rife with ethical problems. Further, since research is more carefully regulated than medical practice, it is quite likely that fewer ethical violations occur in research. To say that two activities are ethically different is not to say that either is inherently better than the other.

Abandoning the Distinction The distinction between research and therapy is most likely to be obfuscated in the context of clinical trials,

which test the safety or efficacy of investigational and standard treatments. Since patients may derive medical benefit from trial participation, especially in phase III RCTs (the final stage of testing, which many investigational drugs never even reach), clinical trials are often characterized as "therapeutic research." Nonetheless, the process of treatment in RCTs differs radically from routine clinical practice.14 Consider the contrast between the hypericum—sertraline trial and routine medical care for depression. If a physician treated 340 patients for major depression, she would not decide which drug to administer by flipping a coin. If the physician elected to use sertraline, she would judge each case individually to determine dose, when to change the dose, and whether to prescribe a second antidepressant or recommend other treatment. We would expect to find considerable variation in the treatment administered to those 340 patients after eight weeks or so. From the vantage point of therapy, this is what it means to provide care to patients. From the vantage point of research, such variation would wreak havoc on experimental design and the validity and generalizability of findings. So when patients are randomized to one or another experimental drug and are treated according to relatively inflexible protocols, the activity is very different from therapeutic medicine. In many other ways, too, routine aspects of research deviate from what would be required by the duties of therapeutic beneficence and nonmaleficence. Volunteer patients and physician investigators are often ignorant of assignment to the experimental or control treatment, which may be a placebo. Trials often include interventions such as blood draws, lumbar punctures, radiation imaging, or biopsies that measure trial outcomes but in no way benefit participants. RCTs often contain a drug "washout" phase before randomization to avoid confounding the evaluation of the investigational treatment with the effects of medication that patients were receiving prior to the trial. These various features of research design promote scientific validity, they carry risks to participants without the prospect of compensating therapeutic benefit. For these reasons, Levine argued that the second major contribution of the commission was to abandon the "illogical" distinction between therapeutic and nontherapeutic research, which previous policymakers thought was essential to the proper regulation of research and the protection of human subjects.15 Because research and therapy are distinct activities, and the ethics of therapeutic medicine therefore cannot be automatically extended to guide research, it is mistaken to label research as "therapeutic" or "nontherapeutic,"

Al

9 Research with Humans

as if that made any funda::nental ethical difference. Many research trials consist of a complex mix of therapeutic and nontherapeutic elements—the placebocontrolled trial being only one obvious example—such that labeling the trial as a whole as "therapeutic" or "nontherapeutic" is misleading. In addition, the therapeutic—nontherapeutic distinction diverts attention from key ethical issues. Consider a nontherapeutic trial in which one interviews subjects and takes saliva samples and a therapeutic trial in which one is testing a new cancer drug that has some promise for creating remission but also has potentially life-threatening toxicity. Is the latter trial less in need of stringent regulatory oversight because it is "therapeutic"? Or does the therapeutic—nontherapeutic distinction distract the observer from those aspects of the trials that assume far greater moral weight, such as the level of risks and the potential vulnerability of subjects? Once one understands the distinction between research and therapy, one realizes that "therapeutic" research is still research and that the ethical rules appropriate to it are those appropriate for clinical research generally. Even though the patient may derive benefit from treatment being evaluated, the basic goal of the activity is not personal therapy, but rather the acquisition of generally applicable scientific knowledge. The basic goal and nature of the activity determines the ethical standards that ought to apply. Writing in 1993, Jay Katz affirmed the vital importance of the distinction between research and therapy and deplored its blurring in practice: "The astronomical increase in clinical research has, in practice, not led to a clear demarcation between therapy and research, bioethical theories notwithstanding. This vital distinction remains blurred when physician-investigators view subjects as patients, and then believe that patients' interests and not science's are being served by participation in randomized clinical trials that are so commonly conducted in today's world."15 One of the reasons investigators (and bioethicists) 1-tave failed to appreciate the distinction between research and therapy is that the similarity position has conceived the ethics of clinical trials within the context of the physician—patient relationship. Charles Fried and the Similarity Position

In 1974, Fried published Medical Experimentation: Personal Integrity and Social Po icy, which launched the similarity position within bioethics." Fried assumed that answers to ethical dilemmas in research would have to be found within the ethics of therapeutic medicine. He defended fidelity to the interests of the

individual patient against a model in which "medicine is to be viewed as caring for populations."18What made the RCT ethically suspect was that it seemed to him a prime example of population-focused—rather than individualized—and utilitarian medicine. Fried devoted most of his book to defending patients—rights in personal care."19 Returning to medical research, he took issue with trials in which patients were randomized to receive either the experimental intervention or standard care. Fried coined the term "equipoise" to describe the ethically necessary condition for conducting an RCT: physician-investigators must be indifferent to the therapeutic value of the experimental and control treatments evaluated in the trial. The basic idea of equipoise had previously been articulated by Bradford Hill, a pioneer in the development of RCTs.2° But what Fried objected to primarily in RCTs was not randomization per se but the fact that no informed consent had been obtained. Fried saw the threat of "care for groups" (instead of "care for individuals") as residing primarily in the idea that it was legitimate to enroll subjects in an RCT without explicit, informed consent because the results of the trial would provide new medical knowledge that would improve the lot of future patients.21 Because Fried was concerned chiefly about informed consent, an essential ingredient of both medical research and therapeutic medicine, he saw no problem in applying the ethics of medical therapy to medical research. In the 1970s, the "respect for patient autonomy" movement was gaining steam as a replacement for the old Hippocratic ethic of paternalistic beneficence. Since both Fried and the National Commission seemed on the surface to be championing patient autonomy, it was easy to miss the point that they were proposing two fundamentally different strategies for approaching the ethics of clinical trials. Put another way, so long as the bioethics debate of the moment has to do with whether research ethics requires all competent subjects to give fully informed consent, any fundamental divergence between the similarity and the difference positions is likely to be obscured. The Emergence of Clinical Equipoise

During the 1980s, philosophers interested in research ethics recognized a tension between the obligation of physicians to offer optimal care to their patients ("the therapeutic obligation") and the provision of medical treatment in the context of clinical trials. Don Marquis addressed this problem in a 1983 essay, "Leaving Therapy to Chance."22 The title is significant, suggesting that the RCT is a form of therapy rather than an ethically

Miller/Brody: A Critique of Clinical Equipoise

distinct activity. Marquis began his essay "Consider this dilemma: according to an argument that is hard to refute, the procedure for conducting randomized clinical trials of anticancer drugs is incompatible with the ethics of the physician—patient relationship. If this problem is to be resolved, then either a key procedure for achieving scientific knowledge in medicine must be given up or unethical behavior by physicians must be tolerated."23 In framing this "RCT dilemma," Marquis assumed that the appropriate ethic for clinical trials was that of the (therapeutic) physician—patient relationship. Fred Gifford, following the lead of Marquis, examined the RCT dilemma in greater depth "The central dilemma concerning randomized clinical trials (RcTs) arises out of some simple facts about causal methodology (Ras are the best way to generate the reliable causal knowledge necessary for optimally-informed action) and a prima facie plausible principle concerning how physicians should treat their patients (always do what it is most reasonable to believe will be best for the patient)."24 Neither Marquis nor Gifford found what they regarded as a satisfactory solution, and neither considered the possibility that the difference position could dismiss the "RCT dilemma" as misguided to begin with. In a landmark 1987 article, Benjamin Freedman offered a solution to the RCT dilemma that gained widespread acceptance within bioethics. He argued that the tension between ethically legitimate scientific experimentation and the therapeutic obligation of physicians could be overcome by the principle of "clinical equipoise."25 Freedman agreed with Fried and Marquis that ethical clinical trials had to be compatible with therapeutic beneficence and nonmaleficence. But he argued that Fried's formulation of equipoise was too constraining. Freedman called Fried's original concept "theoretical equipoise" (sometimes called "individual equipoise") and contrasted it with his favored concept of "clinical equipoise" (sometimes called "collective equipoise"). In the latter sense of equipoise, any individual investigator or physician might have reasons to believe that one arm of the RCT offers a therapeutic benefit over the other arm, but the medical profession as a whole remains divided. According to Freedman, an RCT is ethical so long as the professional community has not yet reached a consensus, which recognizes that "medicine is social rather than individual in nature."26 When, and only when, clinical equipoise is satisfied will patients enrolled in a clinical trial be assured that they will not be randomized to treatment known to be inferior. Freedman thus asserted in a later article that clinical equipoise is "grounded in the normative nature of clinical practice, the view that a patient is

ethically entitled to expect treatment from his or her physician—an entitlement that cannot be sacrificed to scientific curiosity."27 The bioethics community perceived Freedman's concept of clinical equipoise as both a theoretical and a practical advance. Theoretically, it appeared to offer a more intellectually compelling argument than Fried's initial formulation. Practically, it would permit useful RCTs that would otherwise be ethically proscribed to go forward. Since it appeared to solve the RCT dilemma by accommodating the conduct of clinical trials with the therapeutic obligation of physicians to offer optimal medical care, clinical equipoise gained wide currency as a fundamental concept of the ethics of clinical trials.28 The persuasive way in which Freedman fortified the similarity position diverted attention from the fact that clinical equipoise collapsed the distinction between research and therapy. The similarity position and clinical equipoise have been popular not only among bioethicists but also among investigators. We speculate that this ethical perspective helps to address investigators' psychological needs. Physician-investigators, after all, went to medical school, not investigator school. To think of research with patients outside the ethical framework of the physician—patient relationship, as the difference position requires, may be difficult and threatening to them. Clinical equipoise offers a formula that seems to allow them to mix both physician and investigator roles—even if the psychological comfort is purchased at the price of ethical obfuscation. The anomaly therefore exists that much of today's bioethical thinking accepts clinical equipoise as an outgrowth of the similarity position, while the Federal regulations grew out of the work of the National Commission, which largely endorsed the difference position. One would imagine that sooner or later proponents of clinical equipoise would realize the need to defend this doctrine from the charge that it conflates the ethics of clinical trials with the ethics of medical care. But this is precisely what has not yet happened. The Case of Placebo-Controlled Trials

Although the similarity position, bolstered by clinical equipoise, became the reigning paradigm in the ethics of clinical trials, its dominion over practice was limited. This divorce between theory and practice has been particularly pronounced in the case of placebocontrolled trials. Freedman and his colleagues argued that the use of placebo controls is unethical whenever proven effective treatment exists for the medical condition under investigation in a clinical trial because

■ II I

II 411

IIILLI

514 9 Research with Humans

those randomized to placebo would receive treatment known to be inferior.29 Despite the clear implications of clinical equipoise for the ethics of placebo-controlled trials, numerous trials, such as the hypericum—sertraline trial, continued to use placebo controls despite proven effective treatment. Placebo controls have typically been used in trials of new treatments for a wide range of chronic conditions—including mood and anxiety disorders, asthma, stable angina, hypertension, and migraine headaches—all of which can be treated with medication of proven efficacy. There are two explanations for this incoherence between theory and practice. First, the FDA has encouraged the use of placebo controls in trials concerning these and other chronic conditions.3° Active-controlled trials designed to test the equivalence of the experimental treatment with a standard treatment suffer from serious methodological limitations. Whenever active-controlled trials show no statistically significant difference between the invest gational treatment and an active comparator, two conclusions are possible. Either both were effective in the trial sample of patients, or neither was effective. Withou: the use of a placebo control, such trials lack internal validity. Accordingly, the FDA has insisted that pharmaceutical companies use placebo controls in trials of new treatments for conditions characterized by fluctuating symptoms and high rates of placebo response.31 Second, the U.S. federal regulations governing human subjects research do not provide any explicit guidance on the use of placebo controls.32 utBs have been free to approve such placebocontrolled trials, provided that they meet regulatory requirements for a favorable risk-benefit ratio, including the potential value of knowledge to be gained and informed consent. For the most part, this lack of fit between theory and practice received little critical attention until the publication in 1994 of an article in the New England Journal of Medicine entitled "The Continuing Unethical Use of Placebo Controls."33 Kenneth Rothman and Karin Michels castigated the practice of placebo-controlled trials in the face of proven ef 'ective treatment and the role of the FDA in encouraging these trials. They cited the Declaration of Helsinki, which relies heavily on the similarity position, as prohibiting this widespread "unethical" practice. Their article stimulated a lively debate over the ethics of placebo-controlled trials. Freedman and his colleagues attacked "the placebo orthodoxy" in a two-part article that challenged the scientific value of placebo-controlled trials and reiterated that they are unethical when proven effective treatments exist because

they contravene clinical equipoise.34 Other commentators, writing in leading medical journals, defended more or less extensive use of placebo-controlled trials on methodological and ethical grounds.35 Without directly challenging the doctrine of clinical equipoise, they implied that clinical equipoise provides erroneous ethical guidance for placebo-controlled trials. Accordingly, the debate over placebo-controlled trials jeopardizes the reigning ethical paradigm of the similarity position and clinical equipoise. Critique of the Similarity Position and Clinical Equipoise Our reconstruction of the recent history of the ethics of clinical trials has traced the emergence and dominance of the similarity position. This history also reveals cracks in the foundation of this ethical paradigm. Simultaneous endorsement of the difference position, reflected in the federal regulatory system and the Belmont Report, and the similarity position, which invokes the doctrine of clinical equipoise, has left the ethics of clinical trials in a state of incoherence. Although this incoherence has not received critical attention, it becomes apparent once the assumptions underlying the similarity position and clinical equipoise are challenged. In addition, the divorce between research ethics theory and clinical trials practice in the case of placebo-controlled trials suggests that a critique of the similarity position and clinical equipoise is overdue. We contend that clinical equipoise is fundamentally mistaken because "the RCT dilemma," for which it was proposed as a solution, is false. Clinical equipoise and all other forms of equipoise make sense as a normative requirement for clinical trials only on the assumption that investigators have a therapeutic obligation to the research participants. The "therapeutic obligation" of investigators, forming one horn of the RCT dilemma, constitutes a therapeutic misconception about the ethics of clinical trials. The presumption that RCTs must be compatible with the ethics of the physician—patient relationship assumes erroneously that the RCT is a form of therapy, thus inappropriately applying the principles of therapeutic beneficence and nonmaleficence that govern clinical medicine to the fundamentally different practice of clinical research. It is impossible to maintain fidelity to doing what is best medically for patients in the context of RCTs because these are not designed for, and may conflict with, personalized care. Although ethically appealing, the project of bridging the gap between therapy and research via the doctrine of clinical equipoise is doomed to fail.

Miller/Brody: A Critique of Clinical Equipoise

The insight that the RCT contravenes the ethics of the physician—patient relationship led Samuel Hellman and Debra Hellman to argue that the RCT is unethical and that other methods of evaluating treatments should be employed.36 This stance, however, would deprive patients and society of the benefits that flow from rigorous scientific evaluation of experimental and standard treatments. The more reasonable conclusion is that Rcrs should be governed by ethical norms appropriate to clinical research, which are distinct from therapeutic beneficence and therapeutic nonmaleficence. Clinical equipoise is neither necessary nor sufficient for ethically justifiable Rcrs. The use of placebo controls when proven effective treatment exists violates clinical equipoise; however, when methodologically indicated, their use is no different in principle from any research intervention that poses risks to subjects without the prospect of benefiting them.37 In many cases, the risks of withholding effective treatment are excessive, and the use of placebo controls would thus be unethicaL Nevertheless, it is the unacceptable level of risk, not the violation of investigators' alleged "therapeutic obligation," that makes these trials unethical. In other cases, including the hypericum—sertraline trial, use of placebo controls when proven effective treatment exists is ethically justifiable. By conflating the ethics of clinical trials with the ethics of therapeutic medicine, proponents of the similarity position may also contribute to the lack of adequate informed consent. If investigators view the ethics of clinical trials through a therapeutic lens, they may explicitly or implicitly foster the therapeutic misconception among research participants—that is, the tendency of participants in trials to confuse clinical trials with medical care. Research participants need to know that the overall activity is aimed not at their own ultimate benefit but at discovering new knowledge to help future patients. If they think that clinical trial participation is a form of therapy, then they cannot give informed consent. Moreover, unlike the therapeutic context, the patient-subject cannot delegate the decision to the physician-researcher. In the therapeutic setting, a patient can decide to trust the physician to choose the best treatment because the physician has the patient's best interests at heart. The investigator has the interests of future patients at heart and so cannot decide for the subject whether or not to participate in the research. To be trustworthy, investigators must themselves understand clearly the ways in which clinical research differs from clinical practice and convey this forthrightly to potential research subjects.

It is worth pondering, however, the practical consequences that might ensue if physicians, investigators, patients, and ethicists understood clinical trials without distortion by therapeutic misconceptions. Would recruitment of participants for valuable clinical trials become substantially more difficult, slowing progress in medical care? The fact that clinical trials are no longer seen as a mode of therapy leaves unchanged the real prospect of therapeutic benefits offered to patients from trial participation, including the opportunity to receive promising investigational agents, ancillary medical care, expert diagnostic evaluations, and education about their disorder. Nonetheless, some patients might be less inclined to participate in clinical trials when they appreciate the differences between these scientific experiments and medical care. To attract enough subjects, researchers might have to pay people for their participation, as researchers in industry-sponsored clinical trials already do with increasing frequency Payments would add to the cost of conducting clinical trials, but it might help prevent the therapeutic misconception among trial participants.38 To be paid signifies that the trial participant is not merely a patient seeking therapy. If additional expenditure is necessary to motivate clinical trial participation, then this is a price worth paying for enhanced professional integrity and informed consent. An Alternative Ethical Framework

In view of the theoretical and practical problems associated with the similarity position and its logical offspring, clinical equipoise, an alternative framework for the ethics of clinical trials is needed. The most promising recent treatment of research ethics has been developed by Ezekiel Emanuel, David Wendler, and Christine Grady.39 They proposes seven ethical requirements for all clinical research: (1) scientific or social value; (2) scientific validity; (3) fair subject selection; (4) favorable risk-benefit ratio; (5) independent review; (6) informed consent; and (7) respect for enrolled research participants. This framework is built on the difference between research and therapy and on the core value of protecting research participants from exploitation. Yet even this formulation of an ethical framework appropriate to clinical research testifies to the hold of the similarity position. The authors endorse clinical equipoise, claiming it is implied by the requirements of value, validity, and risk-benefit ratio. We contend, by contrast, that the endorsement o f clinical equipoise renders incoherent any account that arises from the difference position. The most important next step for

II

3111

1111:61

9 Research with Humans

research ethics is to develop this "non-exploitation" framework systematically in a way that avoids any conflation of clinical research wit:1 medical care. Those who agree that physician-investigators who conduct clinical trials are not governed by therapeutic beneficence still might argue that clinical equipoise provides important methodological guidance for justifying clinical trials. Freedman and his colleagues have argued that clinical equipoise is both an ethical and a scientific principle: "That principle can be put into normative or scientific language. As a normative matter, it defines ethical trial design as prohibiting any compromise of a patient's right to medical treatment by enrolling in a study. The same concern is often stated scientifically when we assert that a study must start with an honest null hypothesis, genuine mechcal uncertainty concerning the relative merits of the various treatment arms included in the trial's design."i° Nevertheless, whatever is valid methodologically in clinical equipoise—the honest null hypothesis—can be stated more clearly and without confusion with the therapeutic obligation, by appeal to the requirement of scientific value: no research participants should be exposed to the risks of valueless research. Clinical trials must be designed to answer valuable scientific questions. If the answer is already known or the question is trivial, then there is no honest null hypothesis, arid a clinical trial should not be conducted. But this is logically independent of whether all the patients enrolled in the trial would receive medical treatment that is believed by the expert medical community to be at least as good as the standard of care. This alternative framework provides accurate ethical guidance concerning clinical research without presuming that the ethics of therapeutic medicine should govern clinical trials. We illustrate this by applying the seven ethical requirements to the example of the hypericum—sertraline trial. Scientific or Social Value and Scientific Validity

The study has social value owing to the widespread use of herbal remedies. Since the efficacy of hypericum in treating depression (especially major depression) was uncertain, there was an honest null hypothesis that hypericum would be no better than placebo. It would have been unreasonable to design the trial as an activecontrolled superiority trial, since it is highly unlikely that hypericum could be shown to be more effective than sertraline. An active-con:rolled equivalence trial would lack "assay sensitivity" because the finding that the reduction in symptoms of depression experienced by those trial participants receiving hypericum was

not significantly different for those receiving sertraline would not validly support the inference that hypericum was effective.'" It would remain possible that neither treatment was effective in the study sample—as was in fact shown. The study, therefore, was properly designed as a three-arm placebo-controlled trial. Fair Subject Selection

There is no evidence to suggest that particularly vulnerable patients were recruited inappropriately for this study, which included a sample representative of depressed patients. Favorable Risk-Benefit Ratio

Risk-benefit assessment of research protocols ultimately comes down to a matter of judgment. With respect to the use of the placebo control—the aspect of the trial that violated clinical equipoise—the risks to participants from an eight-week trial, with careful exclusionary criteria and monitoring, were not excessive and were justifiable by the anticipated value of the knowledge to be gained from the research. Hence, the placebo component of the study had a favorable riskbenefit ratio. Eliminating the placebo would have made the risk-benefit ratio unfavorable by virtue of undermining the scientific validity of the research. Independent Review, Informed Consent, and Respect for Enrolled Research Participants

The report of the study asserted that IRB approval was obtained at all sites and that all subjects gave informed consent. In addition, the described procedures for monitoring subjects for possible risk of harm indicated an acceptable level of respect. In sum, this study was ethically justifiable despite violating clinical equipoise; moreover, had it been designed in accordance with clinical equipoise, it would have been methodologically deficient and therefore ethically questionable. Charles Weijer, a leading advocate of clinical equipoise and the similarity position, has recently claimed that "Placebo-controlled trials in the context of serious illnesses such as depression or schizophrenia are ethically egregious precisely because no competent physician would fail to offer therapy to a patient with the condition."'" Although we agree that depression is a serious illness, the hypericum—sertraline trial demonstrates that there is nothing "ethically egregious" about the use of placebo controls in trials of treatment for depression, as long as the ethical requirements for clinical research are satisfied. Whether or not one agrees that, all things considered, the placebo control was

Miller/Brody: A Critique of Clinical Equipoise

ethical in this trial, the ethical justification of placebo controls has nothing to do with the therapeutic practice of competent physicians. In any case, the alternative ethical framework with its seven requirements provides

adequate guidance for clinical trials without appeal to the incoherent doctrine of clinical equipoise and without conflating the ethics of research with the ethics of therapy.

References

1. Hypericum Depression Trial Study Group, Effect of 19. Hypericum Perforatum (St John's Wort) in Major Depres- 20. sive Disorder: A Randomized Controlled Trial," JAMA 287 (2002):1807-1814. 21. 2. Ibid., 1813. 3. S.M. Stahl, Essential Psychopharmacology of Depression 22. and Bipolar Disorder (New York: Cambridge University Press, 2000). 23. 4. P.S. Appelbaum, L.H. Roth, C.W. Lidz, P. Benson, and 24. W. Winslade, "False Hopes and Best Data: Consent to Research and the Therapeutic Misconception," Hastings Center Report 17, no. 2 (1987):20-24. 25. 5. P.S. Appelbaum, "Clarifying the Ethics of Clinical Research: A Path toward Avoiding the Therapeutic Miscon- 26. ception," American Journal of Bioethics 2, no. 2 (2002):22. 27. 6. B. Freedman, "Placebo-Controlled Trials and the Logic of Clinical Purpose," IRB 12, no. 6 (1990):1-6. 28. 7. R.J. Levine, "Clarifying the Concepts of Research Ethics," Hastings Center Report 9, no. 3 (1979):21-26. 8. National Commission for the Protection of Human 29. Subjects of Biomedical and Behavioral Research, The Belmont Report (Washington, D.C.: U.S. Government Printing Office, 1979) p. 3. 9. E.J. Emanuel, D. Wendler, and C. Grady, "What Makes 30. Clinical Research Ethical?" JAMA 283 (2000):2701-2711. 10. J. Sugarman, N.E. Kass, S.N. Goodman, P. Perentesis, P. Fernandes, and R.R. Faden, "What Patients Say about Medical Research," IRB 20, no. 4 (1998):1-7. 31. 11. E.G. Miller, D.L. Rosenstein, and E.G. DeRenzo, "Professional Integrity in Clinical Research," JAMA 280 (1998):1449-1454. 32. 12. R.R. Faden and T.L. Beauchamp, A History and Theory of Informed Consent (New York: Oxford University Press, 33. 1986):200-232. 13. D.A. Braunholtz, S.J.L. Edwards, and RJ. Lilford, "Are Randomized Clinical Trials Good for Us (in the Short 34. Term)? Evidence for a 'Trial Effect,"' Journal of Clinical Epidemiology 54 (2001):217-224. 14. J.W. Berg, P.S. Appelbaum, C.W. Lidz, and L.S. Parker, Informed Consent: Legal Theory and Clinical Practice, 2nd ed. (New York: Oxford University Press, 2001):280-283. 15. R.J. Levine, Ethics and Regulation of Clinical Research, 2nd ed. (New Haven: Yale University Press,1986):8-10. 35. 16. J. Katz, "'Ethics and Clinical Research' Revisited: A Tribute to Henry K. Beecher," Hastings Center Report 23, no. 5 (1993):36. 17. C. Fried, Medical Experimentation: Personal Integrity and Social Policy (New York: American Elsevier, 1974). 18. Ibid., 5

Ibid., 94. A.B. Hill, "Medical Ethics and Controlled Trials," British Medical Journal 1 (1963):1043-1049. C. Fried, Medical Experimentation: Personal Integrity and Social Policy (New York: American Elsevier, 1974):8. D. Marquis, "Leaving Therapy to Chance," Hastings Center Report 13, no. 4 (1983):40-47. Ibid., 40. F. Gifford, "The Conflict between Randomized Clinical Trials and the Therapeutic Obligation," Journal of Medicine and Philosophy (1986):347-366. B. Freedman, "Equipoise and the Ethics of Clinical Research," NEJM 317 (1987):141-145. Ibid., 144 B. Freedman, "Placebo-Controlled Trials and the Logic of Scientific Purpose," IRB 12, no. 6 (1990):5. T.L. Beauchamp and J.E. Childress, Principles of Biomedical Ethics, 5th ed. (New York: Oxford University Press, 2001):323-327. B. Freedman, K.C. Glass, and C. Weijer, "Placebo Orthodoxy in Clinical Research. II: Ethical, Legal and Regulatory Myths," Journal of Law, Medicine & Ethics 24 (1996):252-259. R. Temple and S.E. Ellenberg, "Placebo-Controlled Trials and Active-Control Trials in the Evaluation of New Treatments: Part 1: Ethical and Scientific Issues," Annals of Internal Medicine 133 (2000):455-463. T.P. Laughren, "The Scientific and Ethical Basis for PlaceboControlled Trials in Depression and Schizophrenia: An FDA Perspective," European Psychiatry 16 (2001):418-423. Department of Health and Human Services. "Protection of Human Subjects. Code of Federal Regulations." 45 CFR 46,1991. K. J. Rothman and K.B. Michels, "The Continuing Unethical Use of Placebo Controls, NEJM 331 (1994):394-398. See B. Freedman, K.C. Glass, and C. Weijer, "Placebo Orthodoxy in Clinical Research. I: Empirical and Methodological Myths," Journal of Law, Medicine & Ethics 24 (1996):243-251; and B. Freedman, K.C. Glass, and C. Weijer, "Placebo Orthodoxy in Clinical Research. II: Ethical, Legal and Regulatory Myths," Journal of Law, Medicine & Ethics 24 (1996):252-259. R. Temple and S.E. Ellenberg, "Placebo-Controlled Trials and Active-Control Trials in the Evaluation of New Treatments: Part 1: Ethical and Scientific Issues," Annals of Internal Medicine 133 (2000):455-463; E.J. Emanuel and E.G. Miller, "The Ethics of PlaceboControlled Trials-A Middle Ground," NEJM 345 (2001):915-919.

II J11

1111.1j

9 Research with Humans

36. S. Hellman and D.S. Hellman, "Of Mice but Not Men: Problems of the Randomized Controlled Trial," NE JM 324 (1991):1585-1589. 37. E.G. Miller and H. Brody, "What Makes PlaceboControlled Trials Unethical?" American Journal of Bioethics 2, no. 2 (2002):3-9. 38. N. Dickert and C Grady, "What's the Price of a Research Subject? Approaches to Payment for Research Participation," New EnglandJournal of Medicine 341(1999):198-203. 39. See E.J. Emanuel, D. Wendler, and C. Grady, 'What Makes Clinical Research Ethical?" JAMA 283 (2000):2701-2711.

40. B. Freedman, K.C. Glass, and C. Weijer, "Placebo Orthodoxy in Clinical Research. II: Ethical, Legal and Regulatory Myths," Journal of Law, Medicine & Ethics 24 (1996):253. 41. R. Temple and S.E. Ellenberg, "Placebo-Controlled Trials and Active-Control Trials in the Evaluation of New Treatments: Part 1: Ethical and Scientific Issues," Annals of Internal Medicine 133 (2000): 455-463. 42. C. Weijer, "When Argument Fails," American Journal of Bioethics 2, no. 2 (2002):10.

Protecting Communiities in Biomedical Research

new Canadian research guidelines (the Tri-Council policy statement) applied guidelines for the protection of aboriginal communities in biomedical research to a wide variety of other communities, including Ashkenazi Jews and families." This effort was curtailed as it soon became apparent that many of the specific protections could not be applied to these other communities. Rather than simply applying existing guidelines, a rational strategy for the development of protections for communities in biomedical research must entail a series of steps. In discussing this strategy, we will describe characteristics and types of communities; these are not meant to be exhaustive lists but to be the ones relevant to medical research.

Charles Weijer and Ezekiel J. Emanuel

Geneticists and other scientists have increasingly targeted communities for biomedical research into the etiology, especially the genetic determinants, of common diseases and have met with some wellknown successes. For example, particular mutations predisposing to breast, ovarian, and colon cancer have been identified through studies of Ashkenazi Jews.'"Z Although these discoveries will undoubtedly have important implications for cancer prevention and treatment, the community has expressed concern that they may become the target of discrimination, and there is growing public concern that added protections for communities in biomedical research are required.3 Protections for communities in biomedical research have been developed in limited circumstances. The U.S. Food and Drug Administration has issued regulations allowing for a waiver of informed consent in certain emergency room research, provided there is public disclosure of the research plans and consultation with community representatives.4 Guidelines on consulting communities involved in research on HIV/AIDS have been proposed and implemented.5 The most developed protections for communities in biomedical research are found in guidelines for research involving aboriginal communities, exemplified by those of the Australia National Health and Medical Research Council.° Attempts to generalize these approaches by extending them from one community to another have been problematic.' For example, a 1996 draft of the

Community Characteristics

The term community delineates a wide variety of human associations, from tribes to municipalities to religious adherents. A single set of regulations to fit all types of communities is doomed to failure. What is needed are morally relevant criteria that distinguish communities. Characteristics of particular importance or relevance to communities in biomedical research can be identified and used to delineate seven types of communities [Table 1; for definitions see Web table 119 Communities may be arrayed along a spectrum of cohesiveness, from those that have all the characteristics to those that have only a few. At one end of the spectrum, a cohesive aboriginal community often (or nearly always) has all of the characteristics listed. Conversely, a less-cohesive occupational community embodies only two of the characteristics: common culture and a communications network.

Source: From Science 18 (2000) 289,1142-4. Reprinted with permission.

Weijer/Emanuel: Protecting Communities in Biomedical Research

Table 1 Characteristics of Types of Communities in Biomedical Research. Examples are aboriginal, Kahnawake; geographic/political, Jackson, MI, and Iceland; religious, Amish; disease, HIV; ethnic/racial, Ashkenazim; occupational, nurses; and virtual, e-mail discussion group Type of community Community characteristic

Aboriginal

Geographic/ Political

Religious

Disease

++

+/-

Aboriginal Geographic/Political Religious Disease Ethnic/Racial Occupational Virtual

++

Comprehensiveness of culture

++

Health-related common culture

++

Legitimate political Authority

++

++

+/-

Representative group/individuals

++

++

++

+/-

Common economy/ shared resources

++

Communication network

++

Self-identification as community

++

Virtual

+/++

+/+/+/-

++ +/-

+/-

Geographic localization

Occupational ++

++ ++

Mechanism for priority setting in health care

Ethnic/ Racial

++

+/-

+/-

+/-

++

+/-

+/-

+/-

+

+/-

+/-

++

++

+/-

+/-

+/-

++ +/-

++The community nearly always or always possesses the characteristic. +The community often possesses the characteristic. +/-The community occasionally or rarely possesses the characteristic. -The community very rarely or never possesses the characteristic.

Potential Protections Potential community protections extend from the genesis of the research project to the publication of the results.' Consultation in protocol development. The researcher must show respect for the community culture, seek community input on protocol development, ensure research is useful to the community, and respect the community's knowledge and experience. Information disclosure and informed consent. Disclosure to the community should be nontechnical, and appropriate face-to-face meetings are encouraged. The community ought to have adequate time for review, the researcher must obtain the consent of the community, and community consent is required for protocol changes. Involvement in research conduct. The researcher should ensure that skills and research expertise are transferred to members of the community, employment is offered to members of the community the community is reimbursed for research costs, and the community is informed about research progress. Access to data and samples. The researcher must seek community consent for further use of samples, and storage of data should be negotiated. Dissemination and publication of results. The researcher should involve the community in manuscript preparation, transmit a draft report to the community

for comment, acknowledge community contributions, seek consent to identify the community, provide a final report to the community, and obtain community consent for media interviews. The researcher must report compliance with guidelines to the Institutional Review Board (IRB) and the publication. Connecting Guideline Requirements to Community Characteristics It is possible to identify particular characteristics of communities that are necessary for the implementation of specific protections. In this way, each of the identified community characteristics is linked to one or more of the protections (Table 2). For example, if the community is to have input on the protocol, the community must have representatives who can provide this input on behalf of the community. Similarly, if community consent is to be sought before individuals are approached for study participation, the community must have a legitimate political authority that is empowered to speak authoritatively for, and make binding decisions on behalf of, the community, more than mere representation is required. In addition, if the community is to be reimbursed for research costs, the community must have a common economy or shared resources. Furthermore, community consent for further use of samples requires not only that the

II

an

II Lid

9 Research with Humans

community have a legitimate political authority but also that they have a health-related common culture. Unless there are shared ideas about health, the disposition of samples is likely to be of little or no relevance to the community as a whole, although particular

individuals within it may have strong feelings. Finally, if a draft report for comment is to be provided meaningfully to the community, a communication network must be in place so that the report can be distributed to community members.

Table 2 Appropriate Protections for Communities Depend on Their Characteristics. Italics indicate community protections that require consent; nonitalics, protections that require only community consultation COMMUNITY CHARACTERISTICS REQUIRED FOR PARTICULAR PROTECTIONS Community characteristics Proposed protections

H

LPA

Rep.

PS

GL

CE/SR

CN

SI

Consultation in protocol development Respect for culture Input on protocol Research useful Respect for knowledge and experience Process of providing information and obtaining informed consent Nontechnical and appropriate disclosure

V

Face-to-face meetings Adequate time for review Consent Consent required for protocol change.; May withdraw consent Involvement in research conduct Transfer of skills and expertise Employment Reimbursement for research costs

V

Informed about research progress Access to data and samples Consent for further use of samples

V

Storage of data negotiated Dissemination and publication Involvement in manuscript preparation Draft report for comment Acknowledgment

V

Consent to identify Final report

V

Consent for researcher media interview

Synthesizing Appropriate Protections

Three general regimes of protection can be delineated, based on the specific protections appropriate to the distinct types of communities: (i) community consent and

consultation, (ii) community consultation alone, and (iii) no added protections. [See Web table 2.9 Community consent is only possible if the community has a legitimate political authority, which could be a legislative assembly, mayor, or tribal council that

Weijer/Emanuel: Protecting Communities in Biomedical Research

has the authority to make binding decisions on behalf of its members. For instance, the Ashkenazim have no legitimate political authority, and hence, suggesting that community consent be sought from them is neither morally nor pragmatically justifiable. However, this does not undermine the importance of respect for communities or the possibility that community consent is appropriate for some. Not surprisingly, communities that have legitimate political authorities are among the most cohesive communities and have all or most of the characteristics relevant to the implementation of guidelines requirements (Table 1). Thus, protections for aboriginal and geographic or political communities in research indude the full list of guideline requirements; in other words, community consent and consultation are required (Table 2). Even though community consent is not possible for the Ashkenazi Jewish community, other protections that may be characterized as "community consultation" are appropriate. Community consultation encompasses the involvement of community representatives to a limited degree in study planning, informing the community as a whole of the study at its start and as progress is made, consulting with community representatives regarding the disposition of data, and providing them with a draft report on which to comment. Communities that share a religion, a disease, ethnicity, or race may be relatively cohesive and share many although not all, of the characteristics required for the implementation of guideline requirements. Reflecting the intermediate degree of cohesiveness, the list of potential protections listed is roughly half that for aboriginal and geographical or political communities (Table 2). Occupational and virtual communities are the least cohesive of the communities in the typology (Table 1). Generally, they have few of the morally relevant community characteristics, and accordingly no added protections are required for research involving them. However, there may be exceptions to these guidelines. Actual communities are diverse and can deviate from the ideal types. For instance, for historical and social reasons, farm workers or coal miners with strong union representation, geographic localization, unionbased health insurance, and other social security programs may be more cohesive than typical occupational communities. Their cohesiveness may be so extensive that they have all or most of the characteristics legitimizing the additional protections of community consent and consultation. Furthermore, it is important to recognize that human associations are not static but dynamic; bonds within a group may strengthen over time, and novel social structures may emerge as a new community

develops, necessitating reconsideration of the level of protections. For example, it is not meaningful to speak of many disease groups, such as asthmatics, as communities. However, at some time in the future they may come together for support and to advocate for more research funding and a voice in setting research agenda. In this evolution, such a group accrues these same morally relevant characteristics that confer the ability and obligation to enact protections when the community is targeted for research. Similarly a relatively cohesive community may experience disagreements, rupture, and disintegration over time, losing characteristics and thereby losing attendant protections.

Possible Questions How do community protections relate to individual informed consent? Ultimately, no person can be enrolled in research without his or her individual consent. Properly understood, community consent is an additional protection; a study may not proceed without informed consent from both the community and the individual research subject. However, protections for communities are asymmetric If the community consents to research participation, individuals may still refuse to participate; if the community does not consent, then individuals who are identified because they are members of the community should not be approached for study enrollment. Conflicts may arise when individual and community interests conflict. For example, what if the community withdraws consent for study participation when individual research subjects seem to be deriving medical benefit and wish to continue participating? Inevitably, the answer to such conflicts will depend on the circumstances. Is it more appropriate to conceive of a community as a vulnerable group protected by current regulations? Research ethics and protections have largely been shaped in reaction to instances of unethical and exploitative research on prisoners, children, the elderly, the poor, and racial or ethnic minorities." Vulnerable groups are socially, economically, and otherwise disadvantaged and, therefore, are more susceptible to exploitation or harm. Regulations protecting such groups include added consent requirements and limits on the nontherapeutic research risk to which they may be exposed.0 Conversely, the driving issue for protections for communities is not vulnerability but rather that communities have interests that are entitled to respect and protection.° Respecting and protecting the interests of a community call for a partnership between community and researcher. This is a fundamentally different relation than with vulnerable groups, for which there is more of a protective guardianship.'4 Consequently, the protections typically afforded

IIA

9 Research with Humans

vulnerable groups, such as limits upon risk, would be inappropriate for communities in research. Might a community use added protections for research to legitimize the oppression of groups within the community? For example, a community with a male-dominated leadership may silence the voices of women within the community. Such a community may be reluctant to permit research into the prevalence and causes of spousal abuse, because it may reflect poorly on the community. Careful reflection is needed to ensure that the desire to protect the community in research does not perpetuate oppression. All the relevant values, including respect for persons, beneficence, justice, and respect for communities, must be used in the assessment of potential oppression. Viewed in isolation, any of the ethical principles articulated in the "Belmont report"15 may be used to justify unethical ends. One formulation of justice requires that the burdens and benefits of research participation be distributed equitably; another aspect of justice calls for the elimination of domination.16,17 Thus, a community that seeks to perpetuate oppression might be legitimately criticized on grounds of justice; community safeguards used to perpetuate such oppression have no moral force. Who counts as the community leader? In some communities, a multiplicity of legitimate leaders may make it difficult to discern with whom researchers ought to interact; an aboriginal community may have both a tribal council and an elected mayor. The decision will depend on the values and traditions of particular communities and whose authority encompasses the questions raised. For instance, political leaders may be appropriate for

interaction regarding reimbursement for use of resources, whereas religious leaders have a stake in cases involving the disposition of tissue samples. What if the community wants to suppress adverse or undesirable research findings? This problem is not restricted to research with communities but also exists in relation to research funded by pharmaceutical companies, managed care and other health care institutions, or other research organizations.18 Experience in research with the aboriginal community provides a useful guide to the negotiation of disparate interests. Researchers and the Kahnawake community have negotiated a mechanism in which consensus between the researcher and the community on data interpretation is sought.19 If consensus cannot be attained within a reasonable amount of time, the competing interpretations of the study will both be published. Further experience may generate other examples of creative and equitable solutions to these problems that are less threatening to academic freedom than many existing agreements between researchers and for-profit companies. Putting Principle into Practice

Undoubtedly, difficult questions do remain. By providing precision in distinguishing different types of communities in research, their characteristics, and protections appropriate for each, this analysis should make discussion of community consent and consultation more focused. By distinguishing between community consent and community consultation, skepticism as to the feasibility and appropriateness of additional protections for communities in research ought to be allayed.

References and Notes 1. 2. 3. 4. 5. 6.

7. 8.

9.

J. P Streuwing e t al., Nature Genet. 11, 198 (1995). S. J. Laken et al., Nature Genet. 17, 79 (1997). S. Lehrman, Nature 389, 322 (1997). 61 Code of Federal Regulations 51497. C. Levine, N. N. Dubler, R. J. Levine, IRB: Rev. Human Subj. Res. 13, 1 (January-April 1991). Australia National Health and Medical Research Council, "Guidelines on ethical matters in aboriginal and Torres Strait islander research" (NHMRC, 1991). C. Weijer, G. Goldsand, E. J. Emanuel, Nature Genet. 23, 275 (1999). Canada Tri-Council Working Group on Ethics, "Code of conduct for research involving humans (draft)" (Minister of Supply and Services, Ottawa, 1996). Medical Research Council of Canada, Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, "Tri-council policy statement: Ethical conduct for research involving

10.

11. 12. 13. 14. 15.

16. 17. 18. 19.

humans" (Public Works and Government Services Canada, Ottawa, 1998). Supplemental material is available to Science Online subscribers at www.sciencemag.org/feature/data/ 1050363.shl. R. J. Levine, Ethics and Regulation of Clinical Research (2nd ed.) (Yale Univ. Press, New Haven, CT, 1988). C Weijer, Accountability Res. 7, 21 (1999). C. Weijer, Cambridge Q. Healthcare Ethics 8, 501 (1999). B. Freedman, A. Fuks, C. Weijer, Hastings Center Rep. 23, 13 (March-April 1993). National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, "Belmont report" (Department of Health, Education, and Welfare, 1979). J. Rawls, A Theory of Justice (Harvard Univ. Press, Cambridge, MA, 1971). M. Walzer, Spheres of fustice (Basic Books, New York, 1983). D. G. Nathan and D. J. Weatherall, Lancet 353, 771 (1999). A. C. Macaulay et al., Can. J. Public Health 89, 105 (1998).

Titus/Wells/Rhoades: Repairing Research Integrity

Repairing Research Integrity Sandra L. Titus, James A. Wells, and Lawrence J. Rhoades Misconduct jeopardizes the good name of any institution. Inevitably, the way in which research misconduct is policed and corrected reflects the integrity of the whole enterprise of science. The US National Academy of Sciences has asserted that scientists share an "obligation to act" when suspected research misconduct is observed.' But it has been unclear how well scientists are meeting that obligation. In the United States, the Office of Research Integrity (ORI) evaluates all the investigation records submitted by institutions and plays an oversight role in determining whether there has been misconduct at institutions that receive support from the Department of Health and Human Services (DDHS). The reported number of investigations submitted to ORI has remained low: on average 24 institutional investigation reports per year.2 ORI focuses resources, not only on evaluating institutional reports of research misconduct but also on preventing misconduct and promoting research integrity through deterrence and education. To evaluate these initiatives, we investigated whether the low number of misconduct cases reported to ORI is an accurate reflection of misconduct incidence, or the tip of a much larger iceberg. The latter seems to be the case. The 2,212 researchers we surveyed observed 201 instances of likely misconduct over a three-year period. That's 3 incidents per 100 researchers per year. A conservative extrapolation from our findings to all DHHSfunded researchers predicts that more than 2,300 observations of potential misconduct are made every year. Not all are being reported to universities and few of these are being reported to the ORI. No regulatory office can hope to catch all research misconduct and we think that the primary deterrent must be at the institutional level. Institutions must establish the culture that promotes safeguards for whistleblowers and establishes zero tolerance both for those who commit misconduct and for those who turn a blind eye to it. Defining Misconduct

A first step in developing that culture is taking stock of misconduct's frequency. Several investigators

have addressed research misconduct incidence with limited results because of methodological problems, such as applying inconsistent definitions of misconduct or not accounting for duplicate reports of the same incident.3.4,5 So, we used the US federal definition of research misconduct6—fabrication, falsification, or plagiarism in proposing, performing or reviewing research, or in reporting research results—and verified whether reports accurately fitted that definition. The possibility of duplicate reports was virtually eliminated by selecting only one National Institutes of Health (NIH)—funded researcher in a given department to respond. We asked about events only from the past three academic years to avoid inclusion of distant events and to have a consistent time parameter. We used frequent and varied reminders to secure a high response rate to the survey. Previous research has treated survey reports of misconduct as if the observer could make the determination that they had observed misconduct. Instead, we consider the observations to be "possible research misconduct" and not all such observations will result in a finding of misconduct. In all we asked 4,298 scientists holding NIH extramural research funds at 605 institutions to respond to the survey so that our findings would be representative of a broad spectrum of research fields as well as varied sizes of institutions. What Scientists Saw

In 2006, we asked participants to indicate the number of times they had observed suspected research misconduct in their own department in the past three academic years (2002-05). A total of 2,212 scientists provided complete responses to questions concerning research misconduct (51 percent response rate). Of these, 192 scientists (8.7 percent) indicated that they had observed or had direct evidence of researchers in their own department committing one or more incidents of suspected research misconduct over the past three academic years. The 192 scientists described a total of 265 incidents. Scientists were asked to indicate how they became aware of the possible misconduct and were told to report observations and not hearsay. Suspected misconduct was observed at all scientific ranks including postdocs, students, and tenured faculty members. The following are examples of how scientists described such incidents. We used these descriptions to validate

Source: From Nature 453 (19 June 2008), 980-2, doi:10.1038/453980a.

II JE

9 Research with Humans

whether the observation met the federal definition of research misconduct. "A post doc changed the numbers in assays in order to 'improve' the data." "A colleague duplicated results between three different papers but differently labelled data in each paper." "A co-investigator on a large, interdisciplinary grant application reported that a postdoctoral fellow in his laboratory falsified data submitted as preliminary data in the grant. As principal investigator of the grant, I submitted supplementary data to correct the application " "A colleague used Photoshop to eliminate background bands on a western blot to make the data look more specific than they were." Two people independently coded and evaluated the 265 descriptions to determine whether each met the federal definition of research misconduct. In all, 64 reports (24 percent of the total) did not meet the threshold of the federal definition—which left 201 observations of potential misconduct made by 164 scientists (7.4 percent). These 201 misconduct observations included fabrication or falsification (60 percent) and plagiarism only (36 percent). According to our respondents, 58 percent of the observed incidents had been reported to officials at their institutions. In 24 perce:at of incidents it was the survey respondent who reported it and in 33 percent of the incidents it was someone other than the respondent. Responses indicated that 37 percent of incidents were not reported by anyone and for 5 percent of the cases respondents did not know Study Limitations

Extrapolating the survey results—even conservatively— projects an alarming picture of under-reporting. NIH extramural research grants in 2007 supported an estimated 155,000 people, which includes principal investigators and other research personnel.' In our survey, 201 cases were observed over three years by 2,212 respondents, essentially 3 cases per 100 people per year. Most conservatively, we assumed that non-responders (roughly half of our sample) did not witness any misconduct. Thus, applying 1.5 cases in 100 scientists

to 155,000 researchers suggests that there could be, minimally, 2,325 possible research misconduct observations in a year. If 58 percent of these cases were reported to institutional officials as in our survey, approximately 1,350 would have been reported whereas almost 1,000 could be assumed to go unreported to any official. These numbers indicate a sizeable disconnect between what universities are seeing and the 24 investigations evaluated by the ORI annually. Could all the predicted cases be found to lack evidence? Could all the cases be concluded at the inquiry stage? Could the cases be primarily occurring in research that is not funded by the Public Health Service and hence not reportable to the ORI? Can duplicate observations of misconduct account for this disparity? We doubt that affirmative answers to these questions could sufficiently explain the discrepancy. We recognize that this estimate is not perfect. First we are applying our findings from a defined context to a much larger context and one that also includes the staff of the investigator. Another weakness of the prediction is the fact that scientists in our study would have been narrowly reporting observations restricted to their own experience. A single observer in a department cannot be expected to have been exposed to all instances of misconduct. Thus, our estimate may be off by an order of magnitude in either direction. On an individual level, many reasons for underreporting are easy to understand because they involve motivations we might all have experienced. For example, one does not want to accuse falsely. One may also fear that reporting would take time away from research, or have concerns and fears about possible retaliation. One may assume someone else will or should report it. Or one may have sympathy toward a researcher, and might think "it's not too bad," it can be sorted out without a careerdamaging investigation. Reporting also necessitates confidence that the issue will be examined carefully and thoroughly. Keeping It Quiet

The leaders of institutions may also have concerns about handling research misconduct. Because public image is important to institutions, some may try to minimize reporting and keep unfavourable information from reaching the ORI and the press. An institution may choose to ignore conducting an investigation and instead they may simply dismiss an

Titus/Wells/Rhoades: Repairing Research Integrity

accused person or even a whistleblower in the hope that the problem will go away without needing further examination. Additionally, institutional leaders may wish to ignore or minimize allegations of possible research misconduct to protect the revenue that the researcher generates; some may avoid investigations because they are costly in terms of time and money. Administrators may not recognize the significance of evaluating research misconduct and of course they may be poorly equipped to conduct an investigation in an appropriate manner. Fundamentally all explanations seem to share a common denominator—the failure to foster a culture of integrity. An analysis commissioned by the ORI found in 2000 that only 29 percent of institutional misconduct polices explicitly obligate members to report scientific misconduct.8 Individuals and institutions, not the federal government, are the guardians of research integrity. Therefore, we urge action and recommend six strategies to champion integrity.

Train the Mentors

If we want to build a stronger culture of integrity, then the current generation of researchers has to be educated to pay more attention to how they work with their junior team members. Social science has a long history of describing how group standards affect individual behaviour. Mentors specifically need to become more aware of their roles in establishing and maintaining research rules and minimizing opportunities to commit research misconduct.10 Only 34 percent of scientists in a study with 2,206 laboratory directors strongly agreed that their mentor had prepared them to be a good mentor to others.11 An institutional investment in building better mentors is an important vehicle to promoting research integrity. Use Alternative Mechanisms

To create a zero-tolerance culture, we think that it is essential that an institution specifies and implements the requirements that all suspected misconduct must be reported, and all allegations must be thoroughly and fairly investigated. Social responsibility to the academic community and to the public who fund the research will be strengthened when it is apparent that an institution has a real commitment to integrity.

Institutions must start to use other means to protect the integrity of their studies. The Institute of Medicine recommends that "Universities should not rely upon formal complaints of scientific misconduct as the sole source of monitoring the integrity and quality of the research conducted under their auspices. They need continuing mechanisms to review and evaluate the research and training environment of their institution."12 Auditing research records would be one such means. Mechanisms of review are needed to reduce deficient record keeping, improper protection of human or animal subjects or the utilization of questionable research behaviour.13

Protect Whistleblowers

Model Ethical Behaviour

Careful attention must be paid to the creation and dissemination of measures to protect whistleblowers. Responders to our survey said that reporting would be most likely to improve if institutions and the federal government increased the whistleblower protection. Indeed, more than two-thirds of whistleblowers, in a Research Triangle Institute study, experienced at least one negative outcome as a direct result of their actions.9 Plus, 43 percent reported that institutions encouraged them to drop the allegation.

People imitate the behaviour of powerful role models. Institutions successfully stop cheating, for example, when they have leaders who communicate what is acceptable behaviour, encourage faculty members and staff to follow the policies, develop fair and appropriate procedures for handling misconduct cases, focus on ways to develop and promote ethical behaviour, and provide clear deterrents that are communicated.14 Nearly one generation after the effort to reduce misconduct in science began, the responses by NIH scientists suggests that falsified and fabricated research records, publications, dissertations, and grant applications are much more prevalent than has been suspected to date. Our study calls into question the effectiveness of self-regulation. We hope it will lead individuals and institutions to evaluate their commitment to research integrity.

Adopt Zero Tolerance

Clarify How to Report

Researchers in our study also emphasized what would promote reporting: establishing a reporting system that clearly identifies the individuals to whom allegations should be brought, and establishing clear policies, procedures and guidelines related to misconduct and responsible conduct.

II

VIII

IIILLI

9 Research with Humans

References 1. http://www.nap.edu/html/obas/. 2. httpi/ori.dhhs.gov/research/intra/documents/Investig ations1994-2003-2.pdf. 3. Swazey, J., Anderson, M. & Lewis, K. Am. Sci. 81, 542-553 (1993). 4. St James-Roberts, I. New Sci. 72, 466-469 (1976). 5. Rankin, M. & Esteves, M D. Nurs. Res. 46, 270-276 (1997). 6. Department of Health and Human Services Public Health Service Policies on Research Misconduct 42 CFR 93 (2005). 7. Lederhendler, I. National Institutes of Health, personal communication.

8. http://ori.dhhs.gov/documents/institutional_policies. pdf. 9. http://ori.hhs.gov/documents/consequences.pdf. 10. Adams, D. & Pimple, K D. Account. Res. 12, 225-240 (2005). 11. http://ori.dhhs.gov/documents/research/intergity_measures_final_report_11_07_03.pdf. 12. Institute of Medicine, The Responsible Conduct of Research in Health Sciences (National Academies Press, Washington DC, 1989). 13. Martinson, B., Anderson, M. & DeVries, R. Nature 435, 737-738 (2005). 14. McCabe, D. L., Trevino, L. K. & Butterfield, K. D. Ethics Behay. 11, 219-232 (2001).

9.4 Use of Vulnerable Populations in Research

Vulnerability in Research and Health Care; Describing the Elephant in the Room? Sarnia A.Hurst

Introduction Broadly, we agree that the vulnerable should be afforded some kind of special attention, or protection. Defining vulnerable persons or populations, however, has proved more difficult than we would like. This is both a theoretical and a practical problem. On a theoretical level, uncertainty as to what we mean by vulnerability is unsatisfactory because although we agree that this notion has a strong pull., we cannot account for this pull, justify it, or define its limits. On a practical level, we cannot know who should be afforded the protection due to vulnerable persons, or what form this protection should take. Contradictory definitions can lead to confusion for those who are supposed to protect the vulnerable, and wrong definitions may be acted upon. Attempts to define vulnerability have differed in their scope.' At the broad end, we find a European "principle of vulnerability," which should be considered as a universal expression of the human condition2 and

requires us to respect "the right balance between this logic of the struggle for immortality and the finitude of the earthly presence of human suffering.'3 This and other such broad definitions4 encompass humanity in its entirety. At the restrictive end of the spectrum, "vulnerability" in research on human subjects is often applied to individuals who are unable to give informed consent or who are more likely to be exploited.5 These restrictive definitions have been critiqued, even in this context, as both "too broad and too narrow" by authors who proposed to replace special scrutiny for vulnerable populations with ongoing protection of individuals according to existing regulations, and focused attention on "characteristics of the research protocol and environment that present ethical challenges." 6 The definition of vulnerability for the purposes of health care and research with human participants is thus an unanswered question. That such an important question is open should be regretted. Giving up the concept could represent abandonment of a needed moral safeguard.' A definition that includes humanity itself does not provide such an account, as it cannot provide reason for special protection. A restrictive definition will thus be required. It cannot, however, be restrictive to the point of forcing us either to exclude persons that should validly be considered vulnerable, or to pretend they fulfill some

Source From Bioethics 22, 4 (2008), 191-202. Reprinted with permission

Hurst: Vulnerability in Research and Health Care

other criteria for vulnerability when in fact they do not. For example, if we consider that vulnerability in research ethics is centred on the inability to give voluntary informed consent, we may have a convincing case for excluding terminally ill patients from vulnerable groups. If we remained convinced that they are nevertheless vulnerable, our argument could only be that they are indeed incapable of giving valid informed consent. As it has been pointed out, this is unsatisfactory.8 Fortunately existing definitions seem mistaken only in part and may be mutually complementary. Descriptions sometimes resemble those of the proverbial elephant described by people with only partial views. In this paper, I will show how existing definitions of vulnerability in medical research and clinical care are insufficient, and attempt a definition of vulnerability as a claim to special protection that is both comprehensive and usable. I will also address some possible concerns with this proposal. What Is Vulnerability in Research and Health Care?

Restrictive definitions of vulnerability in research and health care can be roughly described as consent-based, harm-based, or comprehensive. Consent-based definitions include that proposed by the ICH tripartite guidelines,9 or by CIOMS, which defines the vulnerable as "Those who are relatively (or absolutely) incapable of protecting their own interests." 1° The Belmont report defines the vulnerable explicitly on grounds of "their dependent status and their frequently compromised capacity for free consent." 11 Perhaps in an attempt to complement their definitions and give a more comprehensive picture, guidelines for ethical conduct of human subjects research have also provided lists of vulnerable groups (Table 1). Although they do provide useful examples, these lists can be long, and lack an organizing principle. It is not clear that they are based on the definitions offered by the same guidelines, or even on a solid family resemblance between the listed groups. More basically, however, it is not clear that the sort of thing we mean by vulnerability should refer strictly to being at risk of giving faulty consent. This understanding of vulnerability is appealing in situations where we do, indeed, count on people's choices to protect them. If we believe that informed consent is the principal protection of human subjects of research, and crucial in clinical care as well, then

anchoring vulnerability to consent is tempting. We let people make their own choices and count on this to protect them. So we should only afford better protection to those less able to protect themselves in this way. Human subjects research, however, present us with examples of human activities where consent is a necessary but insufficient condition to ethical practice.12 The same is true to a degree in health care as well. As a patient, I will often lack crucial information and will thus need health care providers to have my interest at heart more than in other types of choices. In some clinical situations, decisions have the structure of a "prisoners' dilemma": what we consider to be our best option can only be chosen if we are sufficiently confident that others will choose the same option. If everyone chooses generic substitution for a benign ailment, the money available will benefit everyone including those in situations of greater need. If, however, I am the only one to sacrifice the small degree of comfort involved by, for example, taking a pill less comfortable to swallow, no significant benefit will accrue and I will lose out.13 In both of these activities, we do not, in fact, expect consent to do all the protective work. Whether this is because we believe that consent is often flawed to the degree of requiring other safeguards,14 or because we believe it to be intrinsically insufficient, does not change the impact this has on vulnerability. If additional safeguards other than consent are required for everyone, and if vulnerability can exist in relation to these other safeguards, then consent-based accounts of vulnerability are insufficient. A variant of this view expands consent-based vulnerability to include limits on the ability to avoid exploitation.15 But this sort of definition turns out to be either too broad or too narrow, depending on what we mean by exploitation. If we include harm, disrespect, and injustice° within the definition of exploitation, we are left with a poor understanding of the sort of wrong exploitation is. A stricter definition such as an "unfair distribution of burdens and benefits from an interaction," 17 or treating someone "in a way to which he could not possibly consent." 18 is more useful but means that the consent-based concept of vulnerability is not expanded very much by the inclusion of an added risk of exploitation. Moreover, if exploitation is a consent-based kind of wrong, then the problems outlined above will also apply here. For example, if there is indeed a right to access to health care, then being denied such access would constitute a wrong. It would, however, certainly not amount to exploitation. In research, not

II

,MI

nu.1.1

9 Research with Humans

Table 1 Examples of Vulnerability in International Guidelines for Research Ethics Source

Cited examples of vulnerability in human subjects research

Belmont report

• Racial minorities • The economically disadvantaged • The very sick • The institutionalized

45 CFR 46

• Children • Prisoners • Pregnant women and fetuses

Declaration of Helsinki

• Incompetent persons • Persons susceptible to coercion • Persons who will not derive direct benefits from participation • Persons for whom research is mixed with clinical care

CIOMS

• Those with limited capacity or freedom to consent or to decline to consent ... [including] children, and persons who because of mental or behavioural disorders are incapable of giving informed consent • Junior or subordinate members of a hierarchical group ... [such as] medical and nursing students, subordinate hospital and laboratory personnel, employees of pharmaceutical companies, and members of the armed forces or police • Elderly persons • Residents of nursing homes • People receiving welfare benefits or social assistance and other poor people • The unemployed • Patients in emergency rooms • Some ethnic and racial minority groups • Homeless persons • Nomads • Refugees or displaced persons • Prisoners • Patients with incurable disease • Individuals who are politically powerless • Members of communities unfamiliar with modern medical concepts

ICH tripartite guidelines

• Members of a group with a hierarchical structure such as medical, pharmacy, dental, and nursing students, subordinate hospital and laboratory personnel, employees in the pharmaceutical industry, members of the armed forces, and persons kept in detention • Patients with incurable diseases • Persons in nursing homes • Unemployed or impoverished persons • Patients in emergency situations • Ethnic minority groups • Homeless persons • Nomads • Refugees • Minors • Those incapable of giving consent

giving research subjects the result of the project they participated in can signify a lack of respect but it would be a stretch to call it exploitation. Whether we count on Institutional Review Boards ORB), researchers,19 or clinicians to provide additional safeguards for non consent-based wror gs, we do recognize that such wrongs exist. Alternatives to consent-based views of vulnerability include harm-based definitions. One such view accepts a broad definition of vulnerability as universal human fragility, and goes on to define the sort of concept useful to health care and research as susceptibility

to compound additional harms.2° This susceptibility is present when we become biologically weak or diseased, and this is what justifies additional protection. This, however, only recognizes added likelihood of additional harm. To give a counter-example, women and children may frequently be at greater risk of incurring harms but—a ccording to this view—this may not count unless they have already incurred a first harm. Furthermore, it is not clear that people who are at higher risk of being disrespected or of giving invalid consent would be included under this definition. What these persons risk is being wronged, rather than being harmed. Thus,

Hurst: Vulnerability in Research and Health Care 529

though it captures important aspects of the intuitions that drive our notion of who is vulnerable and resembles safeguards sometimes intended specifically to exclude persons at increased risk of harm in research,2' this definition is too narrow. In a way it is the mirror image of consent-based views of vulnerability and shares the same fault: insufficient comprehensiveness. Faced with the risk that a definition of vulnerability may be too narrow, some have proposed combined or otherwise more comprehensive definitions.22 Vulnerability may, for example, be considered to include "groups of people (i) whose capacity to safeguard their own interests as research participants, through the process of informed consent or refusal, is compromised; or (ii) who are more likely to take on the burdens of participation in research, in virtue of some feature they share, and this is not compensated for by other suitably related benefits (not money), or, (iii) who are less likely to gain the benefits of participation in research, in virtue of some feature they share, and this is not compensated for by other benefits to them (or to others similarly situated)."23 This bases vulnerability on both consent and fairness in subject selection. This view is appealing because it covers both equity and freedom. Inasmuch as there are wrongs other than transgressing the requirement for consent or fairness, it will however be insufficient as well. If we are more likely to have our confidential information disclosed, for example, this definition will not consider us to be vulnerable. A promising broad definition is offered by Agrawal, who views vulnerability as "increased potential that one's interests cannot be protected."24 As he correctly notes, labelling the inclusion of vulnerable subjects as automatically unethical is incorrect. It can put these people at a disadvantage, for example if their systematic exclusion from research leads to missed opportunities to gain knowledge useful to them.25 This definition recognizes that we have all sorts of interests that may require protection, and that we may be at risk as regards each of them. Ethical research with vulnerable populations requires more of investigators and IRBs. In some cases, this greater effort will be successful, and the interests of vulnerable persons will be protected in the end. Sometimes, however, this will not be feasible without, for example, excluding them from a protocol. In some instances, however, the reason why our interests cannot be protected is that some of them are limited by human finitude. In such cases, our interests will be truly impossible to protect in a way that places them clearly outside the responsibility of clinicians and researchers. Someone who is at a high risk of dying of a terminal disease is certainly at "increased potential" that her "interests cannot be protected." But this

could not ground any claim that ins give greater than usual protection in this case to the specific interest of living a longer life. Moreover, in some instances our vulnerability may lie in a greater likelihood of requiring protection in the first place. This will happen whenever our interests are discounted, however easy they may be to protect if the concern is present. Garcia proposes to base the requirement for special scrutiny on the concept of equal protection, which requires justifications for unequal treatment.26 Affording equal ethical protection, however, is not limited to avoiding unequal treatment per se and can include using different treatments to compensate higher risks. Mixed definitions have been offered for vulnerability in clinical care also, and tend to focus on the risk of neglect and lack of access to care.27 The Agency for Healthcare Research and Quality defines vulnerable populations as those less able to safeguard their own needs and interests adequately, a view based on consent or self-determination, but also as populations who may incur different health outcomes traceable to unwarranted disparities in their care, or stemming from special needs for care or barriers to care (Table 2).28 The prominence of barriers to care, or the risk of neglect in this wording, reflects the visibility of lack of access as a potential wrong incurred in clinical care. Lack of access to health care, or even to health, is however not the only source of vulnerability in clinical care. Individuals can be wronged without incurring loss of their opportunity to achieve maximum possible health, if they are deprived of the means for self-determination. People at greater risk of having confidential information disclosed are also left out of these definitions. The difficulties involved in navigating between insufficient comprehensiveness and excessive broadness, if all are to be considered vulnerable,29 have fuelled a critique of using the concept of vulnerability in research at al1.3° The authors of this critique argue that the concept of vulnerability has lost force through the inclusion of too many groups identified as vulnerable. This may be true, but only insofar as individuals who belong to such groups are systematically and sometimes inaccurately labelled as vulnerable. This is an avoidable practice.3' In addition, classifying groups as vulnerable can be stereotyping: for example classifying the poor or pregnant women as vulnerable is insulting if we mean that they are not capable of decision-making. If we recognize that their vulnerability is due to greater likelihood of being offered a bad risk-benefit ratio or of being disrespected, then this classification constitutes a justified attack on the perpetrators rather than a slur on the victims. The authors also note that vulnerability is often understood in relation to enrolment in research, whereas certain individuals

uu

I

II

.111

II I N.L1

9 Research with Humans

Table 2 Examples of Vulnerability in Clinical Care Source

Cited examples of vulnerability in health care

Agency for Healthcare Research and Quality

• Populations less able than others to safeguard their own needs and interests adequately • Populations who may incur different health outcomes traceable to unwarranted disparities in their care or stemming from special needs for care of barriers to care

Aday

• Social status • Age: infants, children, adolescents, elderly • Gender: females • Race and ethnicity: African Americans, Hispanics, native Americans, Asian Americans • Social capital • Family structure: living alone, female head • Marital status: single, separated, divorced, widowed • Voluntary organizations: non-member • Social networks: weak Human capital • School: less than high school • Jobs: unemployed, blue collar • Income: poor, low income • Housing: substandard

Danis and Patrick

Those at risk at any particular point in time for unequal opportunity to achieve maximum possible health and quality of rife because of differences in intrinsic and extrinsic resources that are associated with good health • Financial circumstances • Place of residence • Cultural background and ethnicity • Age • Health conditions (such as terminal illness or mental illness, impairments, including psychological and cognitive ones, and functional status or disability, such as inability to communicate effectively)

require ongoing protection. Although it does not tell us what vulnerability is, this is an important point. Vulnerability as a Claim to Special Protection I propose that vulnerability as a claim to special protection should be understood as an identifiably increased likelihood of incurring additional or greater wrong. Vulnerability in this sense is not restricted to the likelihood of faulty consent or even to the limited capacity to defend one's own interests. If we understood the worst, or perhaps the only, wrong to be lack of respect for self-determination, and that no harm is done to the willing, then this definition would be identical to previous ones. The examples of research and clinical care, however, show us situations where we do not expect even ordinary patients or subjects to protect all their interests themselves. When even ordinary, nonvulnerable people are not expected to protect their own interest, then additional elements become visible. Some interests are more likely than others to be placed at risk. They may be harder to achieve and thus to defend. This affects some individuals even when they would be completely capable of defending their own interests in other settings. Even for those able to consent, an increased likelihood of incun-ing additional or greater

wrong means that we expect self-determination to do more work in their case. If a claim exists that we should afford the same protection to all regarding a claim we consider valid, then additional requirements other than complementing faulty consent will be required. This definition is restricted to wrongs, including wrongful harms and the wrongs that we incur when something to which we have a valid claim is denied us. It cannot extend to any additional harm, or any interest more likely to be difficult to protect, because it is not the case that we have a duty to protect all interests from all harms. I could, for example, decide to enrol in research as a sales representative to be introduced to potential future customers. If this did not function as expected, one of my interests would certainly have been harmed, but we would hardly expect an IRB to protect me from this sort of frustration. This definition requires vulnerability as a claim to special protection, to be defined starting from the sorts of wrongs likely to occur. We thus agree with Levine and colleagues,32 who propose that attention must be focused on the characteristics of the research protocol and environment, rather than restricted to characteristics of potential subjects, and that it should regard the ongoing conduct of research as well as enrolment. In the definition proposed here, there is no single specific

Hurst: Vulnerability in Research and Health Care

transgression linked with vulnerability. The substantial contents both of vulnerability itself and of the transgressions associated with it will change with the nature of the wrongs involved. In one sense or another, many individuals will be vulnerable, but this will not mean that they are vulnerable in identical ways. This could partly explain the difficulties in defining vulnerability for the purposes of research with human subjects and clinical care. We seem to be caught between only two alternatives: defining everyone as vulnerable or sticking to notions based on a very limited number of wrongs. These in turn either fail to recognize some valid claims to special protection, or require conceptual contortions such as that which consists in describing someone as vulnerable on a consent-based view when their capacity to consent should really not be questioned. This problem is solved if we define vulnerability as proposed here. This definition is limited in the sense that it does not identify all forms of vulnerability, such as those associated in general with being human, fallible, mortal, and capable of suffering. It is, however, relevant in an important way, as it circumscribes the forms of vulnerability requiring additional attention as compared with the care we usually take to avoid perpetrating wrongs. It is also limited as a definition in that it does not provide a clear cut-off line between the vulnerable and the nonvulnerable. Inasmuch as some individuals and groups will indeed be identifiably more likely to suffer wrongs, however, it provides a framework both for recognizing these groups systematically and for designing ways to address their specific kinds of vulnerability. If an identifiably increased likelihood of incurring additional or greater wrong exists, including any wrongful harm, then there is an increased risk of moral transgression. Vulnerability in this sense is thus a twoway street and affects those who practise health care and research as well. This will be the case any time that an identifiable agent acts in way that predictably affects vulnerable persons.

Applying this definition of vulnerability could take the form of a four step approach: 1. Is there an identifiable potential wrong? 2. If yes, are some people identifiably more likely than others to incur this wrong, or likely to incur it to a greater degree? 3. Who shares in the duty to minimize, or avoid, this wrong, and does it include us in any way? 4. What should we do to minimize this increased likelihood or degree, or to compensate for it in ethically justifiable ways? The mere definition of vulnerability does not identify those with a share in the responsibility for protecting the vulnerable. Applying this definition thus requires the addition of step 3 (Figure 1). Depending on the sort of wrong identified, those responsible for preventing it will vary. Application to Research Ethics

If we accept the proposed definition, it is not surprising that vulnerability proves hard to define: it is as multiple as potential wrongs and as sources of greater likelihood of suffering them. It also involves some judgment, because it is a matter of degree. This, however, does not void the concept of its usefulness. We may disagree in a grey zone; but differences in the likelihood of suffering wrong will often be sufficiently marked to be uncontroversial. Using the concept of vulnerability in research ethics has proved difficult for the same sort of reason. General principles of research ethics exist to protect subjects from wrongs, including wrongful harms. Thus, the special scrutiny required by IRBs to deal with vulnerability is not a difference in kind but a difference in degree of care for ethical criteria that are, indeed, the same. Again, the concept of vulnerability is not rendered useless by this, as it serves to identify groups of individuals that do, in fact, need

......................... ..•••

•

.•". • Identifiably greater likelihood or likely degree of wrongs

Application of .4.„ vulnerability as a claim to pecial protection/

A share in the duty to avoid identifiable wrongs

............................... . Figure 1 Two Ingredients to Apply Vulnerability as a Claim to Special Protection

1

,111

IIIISI

9 Research with Humans

and merit this special care in the application of criteria for ethical research. One application to research ethics, then, is to aid IRBs in applying special scrutiny to protect vulnerable subjects but in a targeted way. The first of our four questions would thus be what IRBs examine as part of their regular work: 1. Are any potential research subjects at risk of being wronged in any way by this research project? For each potential wrong, the other three questions would then become: 2. Are some potential subjects identifiably more likely than other persons to incur this wrong, or likely to incur it to a greater degree? 3. Is our IRB among those who share i n the duty to minimize, or avoid, this wrong? 4. If yes, what should we do to avoid this wrong, or minimize this increased likelihood or degree, or ensure it is compensated in ethically justifiable ways? Based on recognized requirements for ethical research with human subjects,I5 examples of vulnerability are shown in Table 3. Actions required for special protection would need to be tailored to the sort of wrong to be avoided, and to the source of the specific vulnerability (Figure 2).

As outlined above, the relevant question for an IRB at the third step will be "does this include us"? rather than "are we solely responsible for this?." This is an important point, as understanding protection of the vulnerable in research in this way would expand the responsibilities of IRBs. Rather than checking a list of predefined vulnerable groups, they would have to identify who was vulnerable based on the wrongs likely to occur in the case of each protocol they reviewed. This is consistent with their role. Before being regulatory institutions, IRBs primarily have a moral function to protect human subjects of research. Moreover, this expansion of their role would remain limited to wrongs actually linked to research. They would not become responsible for protecting the vulnerable from any kind of wrong whatsoever. Rather than going through a list of pre-identified vulnerable populations, then, IRBs would go through a list of potential research-related wrongs (Table 3). Another application to research ethics could be in thinking through double standards. Critiques have put forward that lower ethical standards are used in developing countries.36 While this is often true, this accusation is sometimes leveled at protocols that actually use all the same standards used in rich countries. The Tenofovir trial was one such example.37 A possible interpretation using the definition of vulnerability proposed here could run like this. If subjects in developing countries are, on the

Table 3 Using This Definition in Research with Human Subjects Requirements

Examples of vulnerability

Social or scientific value

• Lack of access to either benefit or knowledge derived from research

Scientific validity

• Rare disease, leading to difficulties in reaching statistical power to demonstrate therapeutic effectiveness

Fair subjects selection

• All persons likely to be victims of discrimination

Favorable risk-benefit ratio

• Potentially higher risks: unstable patients, emergency research, fetuses, pregnant women • Potentially lower benefits: subjects in phase I studies, terminally ill patients • Subjects whose risk-benefit ratio might sometimes be the object of lesser concern to those responsible for protection:terminally ill patients, disenfranchised persons, poor subjects in developing countries, subjects without access to health care outside of research

Independent review

All persons likely to be victims of discrimination, if those responsible for review share discriminatory views

Informed consent

Difficulties in receiving or understanding the relevant information: not knowing the language used, or how to read Lack of decision-making capacity: some children, some patients with mental disorders, comatose patients Lack of freedom to make a voluntary choice • Through limited freedom: prisoners • Through social weakness: minorities, refugees, sometimes women • Through hierarchical weakness lab employees, students

Respect for potential and enrolled subjects

Health care providers, researchers and students close to the study team who are at increased risk of faulty confidentiality Groups and communities at risk of stigmatisation in the interpretation of study results

Hurst: Vulnerability in Research and Health Care

Are potential research subjects at risk of being wronged by this research project?

Are some potential subject identifiably more likely than other persons to incur this wrong, or likely to incur it to a greater degree?

Is our IRB among those who share in the duty to minimize, or avoid, this wrong?

If yes, what should we do to avoid this wrong, or minimize this increased likelihood or degree, or ensure it is compensated in ethically justifiable ways?

Example 1: breach of confidentiality

• Health care providers are at greater risk. • IRBs share in the duty of protection. • Minimization: could require specific anonymisation of data to limit colleagues' access to their personal information.

Example 2: unfavourable risk/benefit ratio

• If they stand to benefit less, terminally ill patients may be at greater risk. • IRBs share in the duty of protection. • Their risk/benefit ratio should be specifically examined by researchers and IRBs rather than assumed to be the same as for other potential subjects.

Example 3: being enrolled without valid consent

Example 4: being denied the benefit of research

• Subjects of emergency research lack time to think through the options. • IRBs share in the duty of protection. • This can be minimized if consent is asked at that time onlyfor those parts of the protocol that are truly urgent. • The remaining problems with consent at that time can be compensated by including a requirement that an independent clinician confirm that enrolment is not contrary to the potential subject's interest. • Patients in developing countries who lack access to care are excluded from an important part of the social benefits of research. • Although IRBs are not alone in bearing some responsibility for this, it is among the points they should examine in general, and thus also for the purposes of protecting the vulnerable. • Minimization: reasonable availability33 aims to minimize this problem. • Compensation:fair benefits34 aim to compensate it.

Figure 2 Examples of Application of This Approach by IRBs

whole, less likely than subjects in Western countries to see their claims met, additional efforts may be required for them. Not just equivalent efforts, but actually greater ones. This also helps to distinguish problematic double standards based on research-related wrongs from those that may be problematic based on other sorts of wrongs. Not providing additional protection against the risk that dangers to the poor could be discounted is clearly problematic and within the remit of IRBs. Not providing adequate roads may very well be problematic, but lack of roads is clearly not a research-related risk and thus outside the remit of IRBs. Not providing access to treatment will be harder to attribute precisely because it is not quite clear to what extent it can be considered a research-related wrong. Once we know this, however, applying the definition of vulnerability presented here does distinguish double standards that IRBs should concern themselves with from those outside their scope.

Application to the Ethics of Clinical Care

Basic requirements for the ethical conduct of research with human subjects are relatively uncontroversial. The nature of valid claims in clinical care is debated to a greater degree. Application of the concept of vulnerability proposed here to this field can be expected to reflect this. Importantly, however, this does not invalidate the use of this notion. We should expect greater controversy as to what constitutes a valid claim; but once we admit a claim as valid, we become able to identify vulnerable populations as those more likely to be denied fulfillment of this claim, and to identify measures likely to prevent this. Examples of requirements proposed for ethical clinical care are shown in Table 4. They include having access to health care, adequate financial coverage, not being harmed, self-determination, confidentiality, getting fair consideration in resource allocation, and having

I I ill

534

111111

9 Research with Humans

Table 4 Using This Definition in Clinical Care Examples of requirements

Examples of vulnerability

Having access to health care 1. Availability 2. Accessibility 3. Accommodation 4. Affordability 5. Acceptability

• Patients with rare diseases, need for interventions requiring expensive technology • Distance from health service, responsibility for dependent relatives • Long or inflexible working hours • Poverty, uninsurance, underinsurance, distance (high transportation costs) • Populations with reason to distrust the health care system

Adequate financial coverage

• Greater likelihood of being denied insurance, such as pre-existing conditions, or risk factors

Not being harmed

• Patients more likely to be treated in unusual ways, such as health care providers and their families, or patients from whom litigation is feared

Self-determination/Autonomy

• Difficulties in receiving or understanding the relevant information: not knowing the language used, or how to read • Difficulties in requesting a role in decision-making • Lack of decision-making capacity: some children, some patients with mental disorders, comatose patients • Lack of freedom to make a voluntary choice • through limited freedom: prisoners • through social weakness: minorities, refugees, sometimes women • through hierarchical weakness: hospital employees, students

Confidentiality

• Public figures, health care providers, families of health care providers

Getting fair consideration in resource allocation.

• Patients at risk of seeing their interest discounted, such as: • terminally ill patients • elderly patients • cognitively impaired or handicapped persons • disenfranchised or socially marginalized persons

Having a voice as a stakeholder in health care.

• Disenfranchised or socially marginalized persons

Avoiding illness?

• Persons with less access to health literacy • Persons at the lower end of the socio-economic spectrum

a voice as a stakeholder in heiIth care. Access can be further divided into the elements proposed by Penchansky and Thomas: availability, or the degree to which the provider has the necessary resources to meet the patient's needs, accessibility, or the ease with which the patient can physically reach the location of health services, accommodation, or the degree to which the health service is organized in ways that mee: the constraints and preferences of patients, affordability, or how the provider's charges fit with the patients' ability and willingness to pay, and acceptability, or the extent to which patients are comfortable with the characteristics of the health services and vice versa.38 Examples of vulnerability linked to these requirements are shown in Table 4. Whenever resources are allocated, vulnerability also applies to those more likely to see their claims transgressed or discounted in allocation processes. Importantly this may not mean that anyone less likely to have resources allocated :o them is vulnerable in this way: it would depend on their risk of having their claims discounted, rather than on the final result. For example, if decisions about admission to intensive care

for people with short life-expectancies were likely to be based on an evaluation that their needs should count for less, then this would constitute vulnerability. If the same decisions were systematically made with the same sort of considerations afforded anyone, such as likelihood of benefit and quality of life, then they might still be allocated intensive care less frequently, but their claim would not have been discounted. Clearly some populations will only be considered vulnerable in the sense proposed here if they do, indeed, have a valid claim to whatever it is they are more likely to be denied. On some counts, this may be more controversial than in others. For example, some populations are more at risk of becoming ill in the first place.' If we have a right limited to access to health care, then this would constitute a part of normal human fragility, not vulnerability as a claim to special protection. If, however, the claim to provide health care is based on a requirement to equalize health itself as a precondition of fair equality of opportunity,4° then a greater likelihood of becoming ill would constitute vulnerability, a claim to special protection as outlined here.

Hurst: Vulnerability in Research and Health Care

Some Possible Concerns

Importantly, this definition accepts that the claim to protection is based not on vulnerability itself, but on some other valid source. Attempts to ground obligations directly in vulnerability have been made both in continental philosophy`a and in bioethics.42 Concern could exist that if we do not have a requirement to respect a principle of vulnerability, then the vulnerable could lack protection. This concern, however, presupposes an existing claim to protect the vulnerable. Clearly, then, such a claim cannot originate in the principle itself, as it seems to ground the very need for it. The present proposal accepts that the vulnerable have a daim to protection; but this daim is grounded in other claims that we recognize anyway. If a daim for anything exists, then the higher likelihood that this claim will be transgressed generates a requirement for greater attention that this claim be fulfilled whatever this, pre-existing, daim may be. The obligation to avoid wronging is not derived directly from a principle of vulnerability but from another source; specifically, from a valid claim that some wrong should be avoided, including the wrong we incur when a good to which we have a valid claim is denied us. If vulnerability increases the likelihood of being wronged, it also increases the attention required to avoid any wrong that we should avoid for other reasons. This both clarifies and strengthens the claim for protection: we do not need to recognize a specific requirement based on vulnerability but only a situation where fulfilling existing requirements requires additional care. Some may say that defining vulnerability in this way makes the concept superfluous, since it means nothing further than a claim to fulfill duties that we have anyway. This does not, however, void the notion of its use. An increased risk is morally relevant. It can change the actions required to provide protection in degree (for example providing more security for confidentiality) or in nature (as when potential research subjects should be excluded from a protocol). Vulnerability thus truly means something different from the mere existence of a pre-existing claim. We may also wonder whether the strength of the pre-existing claim might not be affected by the added difficulty. There are, after all, instances where duties exist in part because the burden to the agent is not too great, as in the rule of rescue.43 As this suggests, however, this would depend on the sort of claim considered. In research with human subjects, for example, a claim that was too difficult to fulfill might have to lead to the exclusion of potential subjects from the protocol. This is because we do not actually have a claim to participate in research per se. In health care, this would play out very differently. If we have any claim to health care,

then added difficulty would not affect it directly. How much added effort is required when more is needed is indeed an open question, but this is due to the need to balance claims against those of others,44 rather than because the claim itself is diminished. This definition of vulnerability is silent as to whose duty it is to fulfill existing claims. The simple answer might be that whoever had this duty in the first place still has it in the case of vulnerable persons. This, however, should mean anyone who shares in the duty to avoid the identifiable wrong (Figure 1). Asking who is the agent responsible, as if there had to be a single one, is simplistic, as we consider that different people may have different sorts of duties to fulfill the same claim. A child's parents may have a duty to make sure she does not fall into a pond but this does not relieve me of a duty to rescue her if I happen to be there when she does. The problem of requirements for ancillary care, which has proved particularly thorny in research ethics, is an example. If governments have a duty to provide health care to their citizens but fail to do so, how much of this duty falls to researchers? Although we will not attempt to answer this question here, it is noteworthy that there are four counts on which researchers may have such a duty to some degree. Entrustment of their health by research subjects was proposed as grounds for this." That they are on the spot and able to help, as in the rule of rescue, could constitute another. If we should do our share of a collective duty to fulfil such claims," then researchers and sponsoring institutions could do their share by providing ancillary care. Finally, the advantages that researchers and sponsoring institutions sometimes reap, from the very fragilities that make their subjects vulnerable, could also ground such a duty. Increasing recognition that some claim to ancillary care exists47 could be understood as a growing realization of just this problem. Finally, the definition of vulnerability proposed here does not address differences in the way that people may have become vulnerable. As has been proposed, differences between vulnerabilities that originate in injustice, or misfortune, or that are the fault of the vulnerable person herself, could well be relevant.48 According to the view proposed here, however, this would affect the legitimacy of the claim being considered, rather than the definition of vulnerability itself. Acknowledgments

The author warmly thanks Bernard Baertschi PhD, Marion Danis MD, and Alex Mauron PhD, for very insightful and constructive comments, as well as her students. This work was funded by the Bioethics Institute at the University of Geneva, and by the Swiss National Science Foundation.

II ,111

1111j41

9 Research with Humans

Notes 1. M.C. Ruof. Vulnerability, vulnerable populations, and 20. policy. Kennedy Inst Ethics J 2004, 14: 411-25. 2. E. Levinas. 1961. Totalite et infini. Den Haag: 21. Phenomenologica. 3. J.D. Rendtorff. Basic ethical principles in European bioethics and biolaw: autonomy, dignity, integrity and 22. vulnerability-towards a foundation of bioethics and biolaw. Med Health Care Philos 2002, 5: 235-44. 4. D. Callahan. 2000. The Vulnerability of the Human Condition. In Bioethics and Biolaw, Volume II: Four Ethical 23. Principles. P Kemp, et al., ec.s. Copenhagen: Rhodos 24. International Science and AA Publishers; and Centre for 25. Ethics and Law in Nature and Society. A. Maclntyre. 1999. Dependent Rational Animals. Chicago and La Salle, Illinois: Open Court. M.H. Kottow. Vulnerability what kind of 26. principle is it? Med Health Care Philos 2004, 7: 281-7. 5. J.P. Lott. Module three: vulnerable/special participant populations. Developing World Bioeth 2005, 5: 30-54. 27. 6. C. Levine, et al. The limitatims of 'vulnerability' as a protection for human researzh participants. AmJ Bioeth 2004, 4: 44-9. 7. J.P. DeMarco. Vulnerability a needed moral safeguard. Am J Bioeth 2004, 4: 82-4; discussion W32. 8. M. Agrawal. Voluntariness in clinical research at the end of life. J Pain Symptom Manage 2003, 25: S25-S32. 9. ICH Steering Committee. 1956. ICH Harmonized Tripartite Guideline. In Guideline for Good Clinical Practice E6. 28. 10. CIOMS. 2002. International Ethical Guidelines for 29. Biomedical Research Involving Human Subjects. Geneva, Switzerland: CIOMS. 11. The National Commission for the Protection of Human 30. Subjects of Biomedical and Behavioral Research. 1979. 31. The Belmont Report: Ethical Principles and Guidelines for 32. the Protection of Human Subjects of Research. Washington 33. DC: Department of Health, Education, and Welfare. 12. E.J. Emanuel, et al. What makes clinical research ethical? Jama 2000, 283: 2701-11. 34. 13. S.A. Hurst, et al. Conserving scarce resources: willingness of health insurance enrollees to choose cheaper options. J Law Med Ethics 2004, 32: 496-9. 14. S.J. Edwards. Restricted treatments, inducements, and research participation. Bioethics 2006,20: 77-91. 35. 15. K. Kipnis. 2001. Vulnerability in Research Subjects: A 36. Bioethical Taxonomy. In Ethical and policy issues in research involving human research participants. National Bioethics Advisory Commission, ed. Bethesda, MD. 37. 16. D.B. Resnik. Exploitation in biomedical research. Theor Med Bioeth 2003, 24: 233-59. 38. 17. E.J. Emanuel, et al. Undue inducement in clinical research in developing countries: is it a worry? Lancet 2005, 366: 336-40. 18. C. Korsgaard. The Reasons We Can Share. Social philosophy and policy 1993, 10: 24-51. 19. G.B. Tangwa Moral agency, :moral worth and the question 39. of double standards in medical research in developing countries. Developing World Bioeth 2001, 1: 156-62.

Kottow 2004 op. cit. note 4, M.H. Kottow. The vulnerable and the susceptible. Bioethics 2003, 17: 460-71. D. Wendler. When should 'riskier' subjects be excluded from research participation? Kennedy Inst Ethics J 1998, 8: 307-27. P.J. Nickel. Vulnerable populations in research: the case of the seriously ill. Theor Med Bioeth 2006, 27: 245-64. R. Macklin. Bioethics, vulnerability, and protection. Bioethics 2003, 17: 472-86. Nickel op. cit. note 22. Agrawal op. cit. note 8. V. Merton. The exclusion of pregnant, pregnable, and once-pregnable people (a.k.a. women) from biomedical research. AmJ Law Med 1993, 19: 369-451. S.A. Garcia Equal protection clause enforcement as a model for protecting vulnerable human research subjects. AmJ Bioeth 2004, 4: 81-2; discussion W32. Agency for Healthcare Research and Quality. 1999. Request for applications on measures of quality of care for vulnerable populations. LA. Aday: 2001. At Risk in

Amedica; The Health and health Care Needs of Vulnerable Populations in the United States. San Fransisco, CA: JoseyBass, M. Danis, & D.L. Patrick. 2002. Health Policy, vulnerability, and vulnerable populations. In Ethical Dimensions of Health Policy. M. Danis, et al., eds. Oxford, New York Oxford University Press. Agency for Healthcare Research and Quality op. cit. note 27. F.J. Leavitt. Is any medical research population not vulnerable? Cambridge Quarterly of Healthcare Ethics 2006, 15: 81-8. Levine op. cit. note 6 p. 4. DeMarco op. cit. note 7. Levine op. cit. note 6. World Medical Association. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. Edinburgh; 2000. http://www.wma.net/e/policy/ b3.htm (accessed July 17th 2007). Participants in the 2001 Conference on Ethical Aspects of Research in Developing Countries. Ethics. Fair benefits for research in developing countries. Science 2002, 298(5601): 2133-4. Emanuel op. cit. note 12. P Farmer & N.G. Campos. New malaise: bioethics and human rights in the global era J Law Med Ethics 2004, 32: 243-51,190-241. K. Page-Shafer, et al. HIV prevention research in a resource-limited setting: the experience of planning a trial in Cambodia. Lancet 2005; 366: 1499-503. R. Penchansky & J.W. Thomas. The concept of access: definition and relationship to consumer satisfaction. Med Care 1981, 19: 127-40. C.G. McLaughlin & L Wyszewianski. Access to care: remembering old lessons. Health Sery Res 2002, 37: 1441-3. M. Marmot. 2004. The Status Syndrome How Social Standing Affects Our Health and Longevity. London: Bloomsbury Publishings.

Hurst: Vulnerability in Research and Health Care

40. N. Daniels. 1985. Just Health Care: Cambridge University Press. 41. Levinas op. cit. note 2. 42. D.C. Thomasma. The Vulnerability of the Sick. Bioethics Forum 2000, 16: 5-12. 43. A. McIntyre Guilty Bystanders? On the Legitimacy of Duty to Rescue Statutes. Philosophy and Public Affairs 1994, 23: 157-91. 44. N. Daniels. Four unsolved rationing problems. A challenge. Hastings Cent Rep 1994, 24: 27-9.

45. L. Belsky & H.S. Richardson. Medical researchers' ancillary clinical care responsibilities. BMJ 2004, 328: 1494-6. 46. L.B. Murphy. 2000. Moral Demands in Nonideal Theory. Oxford, New York: Oxford University Press. 47. R. Macklin. Four forward-looking guidance points. Developing World Bioeth 2001, 1: 121-34. 48. D.W. Brock. 2002. Health Resource Allocation for Vulnerable Populations. In Ethical Dimensions of Health Policy. M. Danis, et al., eds. Oxford: Oxford University Press: 283-309.

Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies

at the University of Montreal, conducted the research and prepared the Best Practices. F. . .1

D. Avard, L Black, J. Samuel, G. Griener, and B.M. Knoppers With the collaboration of Institute for Human Development, Child and Youth Health (ciHR); and Ethics Office (CIHR)

Scope of Application E.. .1 These Best Practices are designed to assist in the application of the TCPS2 in the paediatric population. They are complementary to, and do not supersede, the TCPS2, and are intended as voluntary guidance for the Canadian health research community working with children and adolescents... [...1 The Development of the Best Practices F.. .1 Considering the need for more explicit discussion of paediatric health research, the National Council on Ethics in Human Research (NCEHR) proposed in 2008 to undertake a two-year project to develop a guidance document for paediatric health researchers, research ethics boards (REB), and institutions. This project was initiated and orchestrated through NCEHR's Emerging Issues Analysis Committee, and in collaboration with the Canadian Institutes of Health Research (CIHR) Institute of Human Development, Child and Youth Health (IHDCYH), the CIHR Ethics Office, Health Canada, and other key organizations such as the Maternal, Infant, Child, Youth, Research Network (MICYRN). The Centre of Genomics and Policy at McGill University, formerly part of the Centre de recherche en droit public (CRDP)

Guiding Principles

The Best Practices use the three ethical principles introduced in the TCPS2—respect for persons, concern for welfare, and justice—as Guiding Principles because of their importance within the discourse of biomedical ethics. These are essentially the same principles adopted in The Belmont Report and similar to the principles used by Beauchamp and Childress in their widely cited Principles of Biomedical Ethics.' This does not mean that all ethical norms incorporate the definitions or terms used by the TCPS2 or that all commentators accept these principles, but the foundation of each principle is incorporated into the Best Practices to guide research decisions [. . .] Respect for Persons in the Paediatric Research Context

It is clear that the morally important capacity to make decisions and determine the course of one's own life does not instantaneously spring, fully formed, into existence. Rather, it develops over time and through exercise. Moreover, it may also be diminished or impaired to varying degrees by illness or other conditions. One must carefully consider how to demonstrate respect for the person with diminished capacity, which includes minors. Moreover, legal restrictions often exist that limit the decision-making capacities of minors. Most provinces promote an individual determination of capacity, but not all. In those instances where the capacity is developing, but not yet sufficiently developed, one must respect the nascent autonomy and foster its further development, while at the same time protecting the interests of the individual. Ethically robust respect for the developing autonomy of a child or adolescent

Source: Centre of Genomics and Policy (cGP), Maternal Infant Child and Youth Research Network (micYRN), 2012. Reprinted with permission.

uu I

ai

I I ill

11111d

9 Research with Humans

is shown by the procedures of informing the minor have different standards for when participation is perabout the proposed research; doing so in a manner missible, illustrating the practical difference between tailored to the child's abili:y to understand; solicit- adult and paediatric research. For instance, there ing his or her assent for participation; and respecting must be sufficient justification for the research to be dissent. In some instances, respecting the child's po- performed on minors instead of on adults. Additiontential for the development of autonomy may provide ally, the risk generally should be minimal.' (The difa very strong reason to delay recruitment until that ficulties in defining minimal risk will be discussed in autonomy blossoms and the mature child can decide greater detail in Chapter VI.) When the minor lacks whether to participate. the capacity to consent, the TCPS2 requires either that When initial recruitment cannot be delayed and the research pose no more than minimal risk or that the individual's participation continues over a long participation is in the minor's best interests—that is, period, the requirement that informed consent be research participation is the best option available to the maintained generates a need to revisit assent and, minor. Notably, this requirement is somewhat more subsequently, consent as autonomy advances. As the restrictive than that in the earlier version of the TCPS, capacity of a child or adolescent to understand infor- and it is also more restrictive than regulations in the mation matures, the sophistication of the informa- United States.° tion provided to him or to her should increase, and The concept of "best interest" deserves additional the ethical importance of assent increases. And at the discussion in the context of children. A competent adult point the individual achieves autonomy (e.g., legal can be expected to judge his or her own best interests— capacity), continued participation will require his or so long as the risk is within accepted bounds. A child or her informed consent. adolescent, though, trusts others to make this judgment. An additional and important consideration in Children and adolescents "are particularly vulnerable to paediatric research is that minors are rarely, if ever, in a conflicting interests and values. The historical record position to make decisions free from outside influence. demonstrates that those who make decisions on their Even as children and adolescents gain greater capabil- behalf do not always serve the children's best interity to understand, they stir rely on their parents or est. At times, even their parents' interests conflict with guardians for most decisions in their lives.' Moreover, theirs."' Researchers, parents, guardians, and research physicians (or researchers) a-e likely to be in a position ethics boards should be careful that other consideraof trust and, even without overt action on their part, tions do not impact a determination of what is in the minors may do what is wanted of them merely because best interest of a child or adolescent. . .1 it is suggested by a trusted Ferson.3 In sum, the use in the TCPS2 of "respect for persons"4—and the adoption Justice of this principle in the Best Practices—goes beyond the [. ..1 traditional concept of autonomy to include protection of those with diminished or no autonomy. Justice in the Paediatric Context In sum, the use in the TCPS2 of "respect for per- Applying the principle of justice creates unavoidable sons"—and the adoption of this principle in the Best tension in the context of research with children and Practices—goes beyond the traditional concept of au- adolescents. At the broadest level of social policy, justonomy to include protection of those with diminished tice calls for research aimed at issues affecting children. or no autonomy . . . In paeCiatric research, this exten- Such research will inevitably need to recruit minors sion serves to improve the ethical standing of minors. as participants and in a great many instances minors This principle is further strengthened by the second who are unable to make decisions for themselves. The Guiding Principle of "concern for welfare." . .1 vulnerability of certain participants who can't decide for themselves, such as children and adolescents, is Concern for Welfare an important difference that must be taken into account so that they may enjoy equitable treatment in [. . .1 research.8 Consequently, additional measures are reConcern for Welfare in the Paediatric Context quired to protect those who are not in a position to Application of the principle of concern for welfare to protect themselves. the context of paediatric research is similar to research Justice also requires researchers not to deliberinvolving adults: researchers still have an obligation ately avoid including children and adolescents in an to minimize harms and to promote the benefit of the effort to reduce the perceived difficulty involved in participant. Research involving minors does, however, conducting research ethically. It also requires those

Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents

who perform ethics review to carefully consider the necessity of such inclusion and to work with researchers to avoid exploitation. The TCPS2 recognizes that it would be unfair to exclude minors from research that might benefit them, especially as there are diseases that are unique to children and adolescents or that are expressed differently in children.9 This could particularly concern specific cancers that affect the paediatric population. Furthermore, inclusion of minors in areas such as pharmaceutical research could provide essential information of efficacy and dosage of medication when physicians are otherwise left without guidance.1° However, particular care must be taken to ensure that minors in research are not shouldered with disproportionate or unreasonable risk, especially in light of their vulnerability and general inability to consent to research on their own behalf. Application of the Guiding Principles to the Best Practices

These Guiding Principles serve as the backdrop to the Best Practices. However, as they are generally applied in the context of adult research, their applicability to paediatric research is imperfect. Yet, perfection is not necessary. Indeed, "principles are neither the start nor the end of the process of ethical reflection."11 No single principle has primacy when considering potential outcomes for research participants. The principles are interdependent and complementary. They represent questions that should be asked and considerations that should be taken into account and perhaps will lead to a deeper discussion of the ethical issues facing researchers, REBs, and families working with the paediatric population. E. . .1 Chapter I: Inclusion of Minors in Research Inclusion of Minors in Research: International and Canadian Contexts

The Nuremberg Code of Ethics of 1947 did not address the inclusion of minors in research. This is unfortunate, as some research that took place during the Nazi period in Germany used children. Following the Nuremberg Code, the twentieth century was further marked by a number of scandals involving paediatric research, for example the Willowbrook case in the 1950s.12 While conducting a hepatitis study on healthy institutionalized children, researchers intentionally infected them with hepatitis in order to understand the disease and to develop a vaccine.13 Even without guidance that directly addressed research involving minors, it was clear that deliberately harming them for the sake of paediatric research ran contrary to

existing international and national ethics. In response, several guidelines were established to ensure observance specifically of the rights of minors participating in research.14 However, in an attempt to protect them, minors were effectively excluded from research and became "therapeutic orphans."13 The unintended consequence of this exclusion was a lack of knowledge of paediatric development and of appropriate medical treatments for minors in general, thereby jeopardizing their health and well-being in the long term. It then became necessary for the international community to re-evaluate their normative guidance documents in order to promote a balance between the duty to protect vulnerable persons, such as children and adolescents, and the need to include them in research. Though the need to conduct research involving children and provide guidance for their inclusion is evident, it was not until the 1964 Declaration of Helsinki that inclusion criteria for minors were clearly laid out.16 Today, there is consensus in international and Canadian ethical norms regarding the need to include children and adolescents in research while offering appropriate protection. Guidelines from the Council of International Organizations of Medical Sciences (ciomS) mention that the participation of minors is indispensable for research concerning diseases affecting infants, and the TCPS2 states that inequity is created "when particular groups fail to receive fair benefits of research or when groups, or their data or their biological materials, are excluded from research arbitrarily or for reasons unrelated to the research question."17 This reflects an important shift in norms related to research involving minors. The inclusion of minors in research is subject to specific conditions to ensure their protection. Unanimity exists among international and national norms on the following conditions: minors should only be involved when the research cannot be carried out on adults; consent of the parents as well as the assent of the minor, when feasible, are required; research should involve no more than minimal risk if there is no prospect of direct benefit to the minor, and research should be approved by an REB and satisfy the legal requirements of the jurisdiction. Additional elements may also be considered, such as the importance of the research in validating adult data, its direct relation to a condition occurring in minors, or its legality. With respect to the criterion of minimal risk, the TCPS2 specifies that "REBs have special ethical obligations to individuals or groups whose circumstances make them vulnerable" (e.g., children) and that their inclusion should not exacerbate their vulnerability.18

1111. II

II .11

11

IIILJJ

9 Research with Humans

The TCPS2 also allows the inclusion of minors in research that involves more than minimal risk (defined in Chapter 2, section B of the TCPS2) if it has "the prospect of direct benefits for them."19 Thus, it would be possible for minors suffering from life-threatening diseases to participate in research that involves more than minimal risk if they can benefit f.:-om it2° (for additional discussion of minimal risk, see Chapter VI—Evaluation of Risks and Benefits). Only a few ethical norms have addressed research involving very vulnerable minors (e.g., impaired or institutionalized minors). However, such inclusion is limited to specific projects. For example, the International Conference on Harmonization (ICH) Clinical Investigation of Medicinal Products in the Pediatric Population El 1 provides that such research should be limited to "diseases or conditions found principally or exclusively in these groups."21 In addition, impaired or institutionalized minors may b otherwise healthy, which creates additional complexity when considering their participation in research: should protections for these populations be even greater than for healthy minors who are not impaired or institutionalized? The inclusion of healthy minors in research is generally limited to studies on prevention or vaccine trials.22 Finally, international and Canadian ethical norms do not generally provide an order of preference in the selection of different groups of children or adolescents for inclusion. However, CIOMS, ICH, and the European Commission address the issue of involving older minors first in research, if possible. The rationale for this agebased stratification of child participants is based on the decreasing vulnerability of minors as they mature. General Statement on the Inclusion of Minors in Research

The inclusion of minors M research promotes their safety and well-being. Children and adolescents differ significantly from adults, physiologically and psychologically as well as developmentally. Their developmental stages influence the limitations and potential benefits of research. Some diseases are found only in tie paediatric population. However, given that minors are a vulnerable population, research should be subject to a rigorous governance framework complying with national ethical and professional norms and legislation and informed by international ones. This framework respects the fundamental principles of research involving human subjects. [...I

Lack of Pharmaceutical Data

There is currently a lack of pharmaceutical data on the paediatric population23 Most paediatric drugs are prescribed off-label due to the fact that they either have never been formally tested on children or adolescents or data from paediatric clinical trials in other countries has not yet been submitted in Canada. Physicians may also extrapolate paediatric dosage from adult data24 even though "children are not small adults."25 Moreover, there may be variations in maturation within the same age group of minors. The absence of age-appropriate formulation of drugs and the lack of data on the efficacy and toxicity of these drugs may expose minors to serious harm. CIOMS underscores that, in the past, drugs that had not been tested on children were nevertheless administered,26 exposing them to serious harm in the absence of sufficient knowledge on the safety and efficacy of such drugs. Even today, although drug labels may clearly state that the drug has not been approved for use in children and adolescents, clinicians might still prescribe the drug because there is nothing else available.27 This is problematic because developmental stages may influence the efficacy or toxicity of the drug28—prescribed drugs are metabolized, extracted, or absorbed differently.29 For example, Health Canada decided in 2008 that some over-the-counter cough and cold medicines (e.g., antitussives and expectorants) should not be labelled for use in children under 6 years old.3° Over-the-counter cough and cold medicines have a long history of use in children; however, there is limited evidence available to support the efficacy of these products in this population. In addition, reports of misuse, overdose, and rare side-effects have raised concerns about the use of these medicines in children under six. Some countries and organizations have begun initiatives to encourage clinical research involving the paediatric population.31 For example, the United States adopted legislative provisions to increase the numbers of drug trials involving minors by both offering incentives to manufacturers and by publishing requests for proposals to third parties in case of lack of interest by manufacturers.32 The European Commission also adopted legislative provisions to facilitate and harmonize the conduct of paediatric clinical trials.33 In Canada, two important initiatives were undertaken to encourage paediatric research. First, Health Canada adopted a guidance document called Guidance for Industry: Clinical Investigation of Medicinal Products in the Pediatric Population, ICH Topic El 1 and prepared a document called Health Canada Addendum to ICH Guidance Document El 1: Clinical Investigation of Medicinal Products in

Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents

Conditions for Inclusion of Minors in Research • •

participation of minors in research is justifiable when the research cannot be carried out with adults; when research requires the participation of minors, least vulnerable minors (e.g., older or more developed children or adolescents) should be included first in the project, if possible and scientifically appropriate; • minors should derive a direct or indirect benefit from their participation in research; • minors should not be exposed to more than minimal risk when research does not hold the prospect of direct benefit; • a minor with a life-threatening disease may be included in research with hope of direct benefit only if: 1. the risks of participation are commensurate with the benefits; and 2. there is no treatment from which the minor can hope to benefit; • a minor with a life-threatening disease may be included in research which does not offer hope of direct benefit to her/him only if: 1. the research does not expose him/her to more than minimal risk; and 2. the research may benefit minors with the same life-threatening disease; • impaired, abused or institutionalized minors should only be included in research if: 1. the research relates directly to their disease o r condition; and 2. the research offers hope of potential benefit to the minor concerned or the research represents no more than minimal risk; some research might require the participation of healthy minors in order to determine, for example, the effect of diet or environmental factors on a genetic predisposition, or the efficacy of a paediatric vaccine, but the research should not expose healthy minors to more than minimal risk.

the Pediatric Population to assist the industry and the researchers conducting research on medicinal products for paediatric use.34 Second, the Food and Drug Regulations were amended in 2006 to extend the data protection period by six months for certain paediatric drugs. To obtain this extension, manufacturers must, within the first five years of the protection period, submit the results of paediatric clinical trials, designed and conducted for the purpose of increasing knowledge of the use of the drug in paediatric populations. Extending the term of data protection in this manner is intended to encourage the submission of paediatric research results to provide health benefits to children and adolescents.35 [. . .] Inclusion of Minors in Research at the End of Life

The inclusion in research of minors who are dying and are receiving palliative care raises many ethical questions because of their double vulnerability: 1) they are minors; and 2) they are dying.38 As mentioned in Section 1.1, the need to include minors in research is widely recognized. There are not always specifications with regard to the categories of children or adolescents who should be included in research (e.g., healthy minors or minors with life-threatening illnesses). Each research project involving minors should thus be evaluated on

a case-by-case basis. However, some organisations have adopted specific guidelines on the participation of dying minors in research.37 Thus, they seem to create a special category for children and adolescents who present with a life-limiting illness and for whom there is no scientifically proven curative therapy available. For example, the American Academy of Pediatrics recognized specifically that minors at the end of life can participate in research when they suffer from "a life-threatening condition that does not respond to all standard therapies, and the [child's] illness is such that death is imminent."38 The following conditions apply to such research: 1. The question being addressed is extremely important. 2. The therapy [experimental treatment] being proposed is well founded in animal and clinical research and/or there is a good expectation that the therapy may be beneficial. 3. Physicians who are not involved in the research must document that the clinical condition of the patient is such that death appears inevitable and standard therapy has not improved the patient's prognosis. 4. The potential benefits outweigh the potential risks.39

9 Research with Humans

An ethical argument often raised against the inclusion of dying minors in research is the potential burden of participation in research_ 4° Some authors assert that these children or adolescents might be asked to take on a bigger burden than other minors because of their "compromised health status and limited remaining time."'" Moreover, they may prefer to spend their remaining time in other ways than participating in research.'" However, the data collected to date on the potential burden on children and adolescents arising from their participation does not support this finding.43 Therefore, it would be unjust to exclude them from research on this basis. There may however be a need to conduct research to determine the perceptions of the minors and parents regarding their participation in research. E. . .1 Chapter II: Consent to Research

Free and Informed Consent Requirements

The informed consent of the competent adolescent, or that of the parents if the minor is not competent to consent,44 is subject to a number of requirements according to international and Canadian norms on ethical research involving human participants. These requirements—which are generally applicable to all research participants and not just minors—mainly focus on the quality of the consent (e.g., free and informed), the elements to include in the consent form, the capacity to understand the information provided, and the actual process of obtaining informed consent. It should be noted that these Best Practices do not provide guidance on how to assess competency. International and Canadian ethical norms agree on the need to obtain the consent of the competent adolescent, or that of the parents if the minor is not competent to consent, before inclusion in research. They are also in agreement on the characteristics of that consent. Indeed, all of the norms analyzed provide that consent must be free (e.g., obtained without manipulation or undue influence) and informed (e.g., researchers must provide all pertinent information). There is agreement on a list of core elements to be discussed and included in the consent process, which attempts to ensure that all pertinent information is provided. These core elements include: the aims of the research, research procedures, potential risks and benefits, participant or third party access to the information collected, compensation of the participant and/or family, right of withdrawal and a description of alternative treatments. However, this list is not

exhaustive. Depending on the type of research conducted, additional elements may need to be included in the consent form. For example, in genetic research, it may be necessary to state the policy on the disclosure of results of genetic tests and their familial implications to the participant and family. In pharmaceutical research, the availability of the drug after the completion of a trial or the lack of information on the drug being studied should be included in the consent form. In the context of longitudinal studies using biobanks, the future research uses (and the requirement for REB approval for those studies), storage and destruction of data and/or samples, who can access the biobank, and who has governance and control over the biobank should also be stated. International norms as well as Canadian norms state that information should be given in understandable language. Researchers should therefore adapt the language used to the abilities of the person consenting. Consent should also be written except when the person consenting cannot read or write, if a written consent is contrary to local custom, or when another reason is judged acceptable by a REB. If this situation occurs, verbal consent is possible but should be documented. CIOMS and icPS2 address the possibility of an implied consent, meaning a consent implied by voluntary actions (e.g., return of a questionnaire by mail). F. . .1 Similarly, international and Canadian ethical norms strongly recommend that cultural background be considered (for additional information, refer to icPS2, Chapter 9, "Research Involving the First Nations, Inuit and Metis Peoples of Canada"). For example, in some communities, a handshake may constitute evidence of trust sufficient to express consent, while in other communities the giving and receiving of gifts constitutes consent.45 Parental Consent for an Early Phase Clinical Trial

Pharmaceutical research also raises the question of whether parents can consent to the participation of their child in an early phase clinical tria1.46 Usually, Phase I examines the "metabolism and pharmacologic actions of the drug in humans" while Phase II examines the "effectiveness of the drug for a particular indication [. . .1 in patients with the disease or condition under study [. . .]"47 Thus, there is a low chance that anyone participating in a Phase I trial would directly benefit since the aim is not to test the efficacy of the new drug but rather its toxicity. Furthermore, Phase I or II studies generally expose minors to more than minimal risk. For these reasons, it is unlikely that healthy minors would be permitted to participate in the early phases of clinical trials due to the increase in risk. Evidently, the

Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents

development of vaccines and other preventative treatments with healthy minors is legitimate prima facie.48 Nonetheless, the inclusion of terminally ill minors may be considered for early phase studies.49 REBs should consider that minors may benefit from the research even if the probability of success is low. In any case, the best interest of the child should prevail. [. . .] Chapter III: Assent of the Minor Assent of the Minor: International and Canadian Contexts [. . .] Minors are not presumed, as adults are, to have

the required competency to provide a free and informed consent to their participation in research. For minors who are not considered to be legally competent or are not legally emancipated, assent, rather than consent, should be sought, together with parental consent. Most international and Canadian ethical norms acknowledge the importance of including children and adolescents, who are capable of understanding, in the decisionmaking process and of obtaining their assent. Assent may be defined as the minor's willingness to participate in the proposed research." The 1989 United Nations' Convention on the Rights of the Child states that, when minors are able to express their own views, they have the right to express those views freely. Furthermore, "the views of the child [should be] given due weight in accordance with the age and maturity of the child."' International and Canadian ethical norms require that researchers obtain the assent of minors before involving them in research. However, assent might be impossible or impracticable to obtain in some circumstances. Only a few policy documents address this matter by stating that not all age groups can provide an assent (e.g., when the child is too immature) and not all situations or types of research can foster this requirement (e.g., emergency research, serious illness). However, when minors regain capacity (e.g., following an emergency), assent should be sought to continue their participation in research. Even if ethical norms do not frame such contexts, exceptions to obtaining assent should also be extended to minors who cannot assent because of developmental and cognitive disability and, obviously, to newborns and the very young; that is, to all children who do not have the capacity to understand. In most Canadian jurisdictions, when the adolescent has a level of comprehension of research similar to a competent adult, consent should be sought. The minor's assent alone is insufficient to be included in research. It should be obtained in addition to the consent given by the parents and should

include important information about the proposed research project. Yet, unlike with consent, neither international nor Canadian ethical norms detail the elements to include in obtaining assent, nor do they advise on who should discuss assent with the minor (e.g., the researcher or someone else on the research team). Some norms specify that information about the project (e.g., nature and purpose of the research), the right to decline, the right to withdraw, and information on potential risks and benefits should be provided to the minor. No further guidance is provided for researchers. Therefore, it may be useful to refer to the elements needed for an informed consent and to adapt these to the particular context of assent. This was suggested by the European Commission in its 2008 Ethical Considerations for Clinical Trials on Medicinal Products Conducted with the Paediatric Population.'2 Although these elements of consent should be considered, the assent process should not be as legalistic as the consent process, considering that minors have a limited comprehension of the research. However, since comprehension will evolve with age, an adolescent should receive more information than a 7-year-old, for instance. Indeed, when seeking assent from a child or adolescent not competent to consent, researchers need to take into consideration age, degree of maturity, developmental stage, and intellectual capacities (e.g., special needs or learning difficulties). Yet, minors of the same age do not necessarily have the same degree of maturity and may not be at the same developmental stage. Thus, competency for assent should be determined on a case-by-case basis. The European Commission suggests that assent be obtained in a manner appropriate to one of three different age groups: 1) children from birth to 3 years of age (where assent is impossible); 2) children from 3 years of age and up (where children between ages 3-5 can understand some expression of altruism; children from ages 6-7 who have an emerging capacity to agree and understand; and children from age 9 who can understand the risks and benefits); and 3) adolescents (with an emerging capacity for independent decision-making and the capacity to make adult decisions in many areas of their life). The European Commission does not draw a line as to when adolescence begins. According to much of the literature, adolescence starts at 14 years of age," although there is not complete agreement on this.54 Information provided in the assent process should be disclosed to the extent allowed by the minor's maturity and intelligence. Researchers should use a level of language and wording that is appropriate to the

SI

II

6111

IIIlia

544 9 Research with Humans

age and psychological and intellectual maturity of the child or adolescent concerned and should also take into consideration any developmental or cognitive challenges. The terms used should be understandable and honest. International and national norms do not necessarily require that assent be written. According to the ICH and the European Commission, it is preferable that assent be written if the minor can read and write.55 However, TCPS2 acknowledges that assent may be expressed verbally or physically.56 Since not all minors can read and/or write, the assent process should be documented to ensure that the rights of the minors concerned have been respected. When assent is written, should the assent form be distinct from the consent form? Most ethical norms are silent regarding this question ICH states that the assent form may be separate from the consent form but does not clearly insist on this point. In contrast, the European Commission requires at the two documents be separate in order to ensure the use of age-appropriate information. However, a common practice is to allow a space on the parental consent form for the assent of the minor. Thus, researchers can explain the research to the minor, verbally and in appropriate language, and document the assent on the consent form. When the minor has provided a written assent, either on the same form as the consent or on a separate form, a copy of the assent form can be given to him/her where appropriate (primarily taking into account age). Finally, like consent, assent is a continuing process that should be renewed throughout the research project as the minor's capacities or the nature of the research changes. For instance, in the context of longitudinal studies, there is a need to continually reassess and renew assent throughout the duration of the research project.57 Also, when adolescents develop the legal capacity to provide a fuly informed consent for themselves or reach the legal age of majority and are capable of making independent decisions during the research, their informed consent should be sought when feasible—as a condition to their continued participation in the research project.58 Chapter IV: Dissent of the Minor

the proposed research. Dissent requires of the minor the same level of capacity to understand as does assent. It may be expressed verbally or physically (e.g., crying, resistance). The European Commission recommends documenting the dissent of the minor. Most international and Canadian ethical norms do not provide detailed guidance on the dissent of minors, except to state that it should be respected. The European Commission states that an "effort should be made to understand and respect differences of opinion between the minor and his/her parents or legal representative."59 But if the minor expresses strong and definitive objections, the dissent should then prevaiL In Canada, the TCPS2 states that minor's "expression of dissent or signs suggesting they do not wish to participate must be respected"6° with no provisions for overriding considerations except when the minor does not have the ability to understand the significance of the research. This inability to understand could be caused by age or lack of maturity (e.g., newborns or very young children), or cognitive or mental disorders. E. . .1 Overriding Dissent It is worth noting that ethical norms used in some juris-

dictions suggest that overriding dissent is possible in particular circumstances, although there is no unanimity regarding this question. For example, both CIOMS and ICH specify that the dissent of the minor may be overridden when: 1) the minor is too immature or too young; 2) there is no reasonable alternative other than what is available in research, and there are reasonable grounds to believe that it will offer benefit; or 3) the minor is suffering from a serious or life-threatening disease and dissent would jeopardize his/her welfare.6' When overriding dissisnt is possible, it is not clear whether researchers need REB approval to do so. While CIOMS provides that such an approval is necessary to override the dissent of a minor who is "older and more nearly capable of independent informed consent,' ICH states that continued parental consent should be sufficient to maintain the participation of the minor in research.63 1. . .1 The dissent of the minor, who is capable of understanding, should be respected.

[. • .] Respecting Dissent

Just as refusal is the opposite of consent, dissent is the opposite of assent and may be defined as the opposition of the minor incompetent to consent to participate in

Dissent may be verbal or behavioural (e.g., body movements) and may be expressed at any time during the research. It should be respected if the minor is capable of understanding, even if the parents consented to their minor's participation in the research project. F. . .1

Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents

"(a) the research involves no more than minimal risk to the participants; Departures from Consent: International and (b) the lack of the participant's consent is unlikely to Canadian Contexts adversely affect the welfare of the participant; Consent of the competent adolescent, or that of the (c) it is impossible or impracticable to carry out the parents if the minor is not competent to consent, is a research and to answer the research question propfundamental requirement for participation in research. erly, given the research design, if the prior consent of the participant is required; However, this obligation may be abrogated under very specific and limited conditions. [. . .1 (d) whenever possible and appropriate, after participation, or at a later time during the study, participants In all cases, to ensure appropriate protection of will be debriefed and provided with additional pertinthe population included in the proposed research, a ent information in accordance with Articles 3.2 and departure from consent is subject to REB approval at 3.4, at which point they will have the opportunity to the very least. In Canada, with the exception of secondary uses, the TCPS2 goes further by requiring that refuse consent in accordance with Article 3.1; and all of the 5 following elements be satisfied before any (e) the research does not involve a therapeutic intervention, or other clinical or diagnostic intervendeparture from the general process of consent may tions."64 [. . .1 be approved: Chapter V: Departures from Consent

Notes 1. Tom L. Beauchamp & James Childress, Principles of Biomedical Ethics, 6th ed. (New York: Oxford University Press, 2008) at 90. 2. DF Merlo, LE Knudsen, K Matusiewicz, L NiebrOj & KH Valtkangas, "Ethics in Studies in Children and Environmental Health" (2007) 33 J Med Ethics 408. 3. Ibid. 4. Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2010) Art. 1.1 [CIHR, TriCouncil Policy Statement]. 5. Ibid at Art. 4.6; JG Hoop, AC Smyth, LW Roberts, "Ethical Issues in Psychiatric Research on Children" (2008) 17 Child Adolesc Psych Clin N Am. 127. 6. The 1998 edition of TCPS required only that the research have potential to benefit the participants or "the group that they represent" (Article 5.3). The US regulations allow children to be exposed to greater than minimal risk if the research is of greater social benefit and certain additional review procedures are followed. Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (1998); 45 C.F.R. § 46.102(i) (2011). 7. V H Sharav, "Children in Clinical Research: A Conflict of Moral Values" (2003) 3 Am J Bioeth 12. 8. Beauchamp and Childress, supra at 1 9. Ibid. Chapter 4B. 10. R Rhodes, "Rethinking Research Ethics" (2005) 5 Am J Bioeth 7.

11. J Harris, "In Praise of Unprincipled Ethics" (2003) 29(5) J Med Ethics 303. 12. HK Beecher, "Ethics and Clinical Research" (1966) 274(24) N Engl Journal of Med 1354 [Beecher, "Ethics"]. 13. L Friedman Ross, Children in Medical Research: Access versus Protection (New York: Clarendon Press, 2006) 155. 14. MA Grodin & LH Glantz, Children as Research Subjects: Science, Ethics, and Law (New York: Oxford University Press, 1994). 15. E Kodish, Ethics and Research with Children: A Case-Based Approach (New York: Oxford University Press, 2005). 16. World Medical Association (wMA), Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects (Finland: 1964). 17. Council for International Organization of Medical Sciences (CIOMS), International Ethical Guidelines for Biomedical Research Involving Human Subjects (Geneva: 2002) Guideline 14 looms, Guidelines 2002]; Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2010) at 10 ICIHR, Tri-Council Policy Statement]. 18. CIHR, Tri-Council Policy Statement, ibid. at Ch. 2B, p. 23. 19. Ibid. at s. 4.6(b). 20. Ibid. 21. International Conference o n Harmonisation o f Technical Requirements for Registration of Pharmaceuticals for Human Use (I0-1), Clinical Investigation of Medicinal Products in the Pediatric Population E 11 (20th July 2000), s. 2.6.3 [Ioi, Clinical Investigation E11]. 22. European Medicines Agency (EMEA), Ethical Considerations for Clinical Trials on Medicinal Products Conducted with the Paediatric Population

11111 I

II

II

.11

11111J

9 Research with Humans

23.

24.

25.

26. 27.

28. 29. 30.

31.

32.

33. 34.

35. 36.

37.

(Recommendations of the Ad Hoc Group for the Development of Implementing Guidelines for Directive 2001/20/EC Relating to Good Clinical Practice in the Conduct of Clinical Trials on Medicinal Products for Human Use) (2008), online s. 15 [EMEA, Ethical Considerations for Clinical Tr als1. TB Ernest, DP Elder, LG Man ni, M Roberts & JL Ford, "Developing Paediatric Medicines: Identifying the Needs and Recognizing the Challenges" (2007) 59 Journal of Pharmacy and Pharmacology 1043 [Ernest, "Paediatric Medicines"]; Saint A Raymond & D Brasseur, "Development of Medicines for Children in Europe: Ethical Implications" (2005) 6 Paediatric Respiratory Reviews 45; Marilyn J Field & Richard E Behrman eds., (Institute of Medicine of the National Academies, Committee on Clinical Research Involving ChildrenBoard on Health Sciences Policy), Ethical Conduct of Clinical Research Involving Children (Washington The National Academies Press., 2004). R Truog, "Increasing the Participation of Children in Clinical Research" (2005) 31 Intensive Care Med 760. 156. TP Klassen, L Harding, JC Craig & M Orrfinga, "Children Are Not Just Small Adults: The Urgent Need for High Quality Trial Evidence in Children" (2008) 5 (8) PLOS 0001. aoms, Guidelines 2002, supra note 17. DG Waller, "Off-Label and Unlicensed Prescribing for Children: Have We Made Any Progress?' (2007) 64 Brit J Clin Pharmacology 1. Ernest, "Paediatric Medicines,' supra note 23. Ibid. Health Canada, Health Canada's Decision on Cough and Cold Medicines (Ottawa: 2008), online: . Canada, "A Canada Fit for Children: Canada's Plan of Action in Response to the May 2002 United Nations Special Session on CF ildren," (2004), online . Food and Drug Administratio a Modernization Act of 1997, Pub. L. No. 105-115, 2296 Stat. 111; Best Pharmaceuticals for Children Act of 2002, Pub. L. No. 107-109, 115 Stat. 1408. EMEA, Ethical Considerations for Clinical Trials, supra note 11. ICH, Clinical Investigation Ell, supra note 21. Regulations Amending the Food and Drug Regulations (Data Protection), C. Gaz. 1996. I. D Davies, "Palliative Care Research Involving Children and Adolescents," in D Avard, J Samuel & BM Knoppers, eds., Paediatric Research in Canada (Montreal: Les Editions Themis, 2009) at 191 [Davies, "Palliative Care"]. American Academy of Pediatrics, "Guidelines for the Ethical Conduct of Studies to Evaluate Drugs in Pediatric Populations" (1995) 95(2) Pediatrics 286 [American Academy of Pediatrics, "Guidelines"]; Children's Project on Palliative/Hospice Services (CaiPPs), "A Call for Change: Recommendation to Improve the Care of Children Living

38. 39. 40.

41.

42.

43.

44. 45. 46. 47. 48.

49. 50.

51.

52. 53.

54.

55.

with Life Threatening Conditions" (2001), online National Hospice and Palliative Care Organization. American Academy of Pediatrics, "Guidelines," ibid. at 292. Ibid. D Tomlinson et al., Challenges to Participation in Paediatric Palliative Care Research: A Review of the Literature (2007) 21 Palliative Medicine 435. CM Ulrich et al., "Respondent Burden in Clinical Research: When Are We Asking Too Much of Subjects?" (2005) 27 1RB 17. DJ Casarett & JH Karlawish, Are Special Ethical Guidelines Needed for Palliative Care Research? (2000) 20 J Pain Symptom Manage 130; Davies, "Palliative Care," supra note 36 at 191. D Tomlinson et al., "Challenges to Participation in Paediatric Palliative Care Research: A Review of the Literature," (2007) 21 Palliative Medicine 435 [Tomlinson "Challenges"]. To ease the reading of this document, the term "parents" also refers to legal representative or legal guardian of the child. C1HR, Tri-Council Policy Statement, supra note 4, art. 3.12. RC Barfield & C Church, Informed Consent in Pediatric Clinical Trials (2005) 17(1) Current Opinion in Pediatrics 20. 21 C.F.R § 312.21. Health Canada, Guidance for Industry: Health Canada Addendum to ICH Guidance Document Ell: Clinical Investigation of Medicinal Products in the Pediatric Population, Ottawa (2003), art 6.1. Barfield, "Informed Consent," supra note 46. European Medicines Agency (EMEA), Ethical Considerations for Clinical Trials on Medicinal Products Conducted with the Paediatric Population (Recommendations of the Ad Hoc Group for the Development of Implementing Guidelines for Directive 2001/20/EC Relating to Good Clinical Practice in the Conduct of Clinical Trials on Medicinal Products for Human Use), (2008), s. 5.7, online: . United Nations, Convention on the Rights of the Child, GA Res. 44/25, UN GAOR, 44th Sess., UN Doc. A/ RES/44/25 (1989) s. 12. EMEA, Ethical Considerations for Clinical Trials, ss. 7, 27. K Senecal, J Samuel & D Avard, Research and the Assent of the Child: Towards Harmonization? in D Avard, J Samuel & BM Knoppers, eds., Paediatric Research in Canada (Montreal: Les Editions Themis, 2009) at 33; D Wendler, Assent in Paediatric Research: Theoretical and Practical Considerations, (2006) 32 J Med Ethics 230; K Toner & R Schwartz, Why A Teenager over Age 14 Should Be Able to Consent, Rather Than Merely Assent, to Participation as a Human Subject of Research" (2003) 3(4) The American Journal of Bioethics 39. Also see Canadian Paediatric Society, Ethical Issues in Health Research in Children (2008) 13(8) Paediatr Child Health 707. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (tcH), Clinical Investigation of Medicinal Products in the Pediatric Population E 11 (20th July 2000). International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for

Avard/Black/Samuel/Griener/Knoppers: Best Practices for Health Research Involving Children and Adolescents

56.

57.

58.

59.

Human Use (10-1), Good Clinical Practice: Consolidated Guidelines E6(R1), (10th June 1996), s. 4.8.12 [ICH, Guidelines E61; EMEA, Ethical Considerations for Clinical Trials, supra note 1, s. 7.1.2. Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2010), art. 3.10 [am, TriCouncil Policy Statement]. Council for International Organization of Medical Sciences (cioms), International Ethical Guidelines for Biomedical Research Involving Human Subjects (Geneva: 2002), Guideline 14 [ooms, Guidelines 2002]; Council for International Organization of Medical Sciences (CIOMS), International Ethical Guidelines for Epidemiological Studies (Geneva: 2008), Guideline 14 looms, Guidelines 20081; EMEA, Ethical Considerations for Clinical Trials, supra note 1, ss. 6.4, 7. CIHR, Tri-Council Policy Statement, supra note 4, art. 3.9(e). European Medicines Agency (EMEA), Ethical Considerations for Clinical Trials on Medicinal Products Conducted with the Paediatric Population (Recommendations of the Ad Hoc Group for the Development of Implementing Guidelines for Directive

60.

61.

62. 63. 64.

2001/20/EC Relating To Good Clinical Practice in the Conduct of Clinical Trials on Medicinal Products for Human Use) (2008), online: s. 7.2 [EMEA, Ethical Considerations for Clinical Trials]. Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada, Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (2010), art. 3.10 [CIHR, TriCouncil Policy Statement]. Council for International Organization of Medical Sciences (CIOMS), International Ethical Guidelines for Biomedical Research Involving Human Subjects (Geneva: 2002), Guideline 14 [ooms, Guidelines 2002]; Council for International Organization of Medical Sciences (CIOMS), International Ethical Guidelines for Epidemiological Studies (Geneva: 2008), Guideline 14 [ooms, Guidelines 2008]; International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (10-1), Clinical Investigation of Medicinal Products in the Pediatric Population E 11 (20th July 2000) s. 2.6.3 licit Clinical Investigation E 11] . Ibid. ICH, Clinical Investigation Ell, supra note 61. CIHR, Tri-Council Policy Statement, supra note 4, art. 3.7.

9.5 Cases

Case 1

Dr Roger Poisson: Fraud in Breast Cancer Study In 1994, University of Montreal professor of surgery, Dr Roger Poisson, was found guilty by the US Office of Research Integrity of enrolling ineligible participants and falsifying and fabricating data between 1977 and 1990 in what was then the largest randomized clinical study of the safety and efficacy of surgical interventions in breast cancer patients. The study compared lumpectomy (with and without radiation treatment) to mastectomy. It confirmed what is now generally accepted, but was contrary to "common sense" and the expectations of many at the time, that lumpectomy-plus-radiation was preferable to mastectomy for some types of breast cancer. Dr Poisson recruited more participants into this study than any other researcher, about 16 per cent of all participants in the study. He accomplished this level of recruitment in part by falsifying participants' eligibility data. He kept a double set of records. One set contained the correct data, which he kept for himself,

and one contained false data which he sent to be used in trial results.' Dr Poisson defended his actions on the following grounds: (1) fraud in enrolling ineligible participants in clinical trials is justified to permit participants access to beneficial treatments that might otherwise be unavailable to them and because the close follow-up they receive is often better than clinical care. (2) Eligibility criteria for randomized clinical trials are too restrictive. (3) Randomization of research participants to different study arms would prevent bias from enrolling ineligible participants.' It is doubtful that these arguments are sufficient, but they are worth thinking about carefully. Enrolling ineligible participants may sometimes give more access to better treatments and follow-up care, but this may be at the expense of slowing scientific progress, adversely affecting many more patients than those enrolled in the study. We also saw in the introduction to this chapter that it is statistically unusual to receive access to unavailable beneficial experimental treatments. Similarly, Dr Poisson's claims that randomization to different study arms would prevent biased results from enrolling

II

IL

9 Research with Humans

ineligible participants cannot be taken seriously. His suggestion that eligibility for trials is too restrictive raises a more interesting question. Entry criteria for studies must not be arbitrary, and there is some evidence that study trials contain unnecessary exclusion criteria. However, if studies are opened up too broadly to give more patients a chance to access potentially beneficial therapies, therapeutic misconception is arguably motivating research at the potential expense of science and future patients. Dr Poisson's behaviour arguably reflects more general problems in combining the role of researcher with caregiver, which were discussed in the introduction

to this chapter. It can be a psychological challenge for physician-researchers to run randomized trials when they strongly believe one arm to be superior to another because of the duty of care physicians owe patients. The physician-researcher conflict raises the further question as to whether there is adequate scientific and ethical oversight of the conduct of clinical trials. Today, many years after this scandal, there remains no direct, on-site independent ethical oversight of trials after they receive initial review by a research ethics board. Scientific monitoring is also limited, typically to on-site review of collected data, which can be falsified.

Notes 1.

Arthur Schafer, "The Moral Anatomy and Ethical Pathology of the Randomized Clinical Trial," in P.P.D. Deyn, ed., The Ethics of Animal and Human Experimentation (John Libby & Company, 1994).

Case 2

Mr Halushka: Human Research and Harm to Participants In August 1961, Walter Halushka participated in a clinical trial for a new drug at the University of Saskatchewan where he was a student. During his initial visit, Dr Wyant told Mr Halushka that the trial for this drug had been conducted many times before and that the drug was "perfectly safe." Dr Wyant explained that electrodes would be put into Mr Halushka's head and limbs and a catheter would be inserted through an incision made in his arm; Mr Haluska would be paid $50 for participating in the study. Mr Halushka agreed and signed a consent form, which waived liability for the researchers and the University of Saskatchewan if there were any "untoward effects or accidents." Two days later, while undergoing the procedure, Mr Halushka was given the anaesthetic drug fluoromar, which had not been previously tested or used, and the catheter was inserted into his heart chambers. However the anaesthetic was too strong and caused cardiac arrest. Resuscitation had to be done by making a large incision in Mr Halushka's chest and manually massaging the heart. Mr Halushka was Note 1.

Halushka v University o f Saskatchewan et al., [1965] 53 436 (Sask. C.A.).

DLR (2d)

2. Charles Weijer, "The Breast Cancer Research Scandal," Canadian Medical Association Journal 152 (1995) no. 8, 1195-7.

unconscious for four days and was discharged 10 days later, having suffered brain damage and reduction of his mental abilities. Mr Halushka brought a lawsuit against the researchers and the university for trespass to the person and negligence. A jury found in Mr Halushka's favour, but the university and researchers appealed. The Saskatchewan Court of Appeal found that inadequate disclosure of information had occurred when the doctor stated the test was safe and failed to state both that fluoromar had not been used before and that the catheter would be inserted directly into the heart and not just into his arm vein. This negated Mr Halushka's consent. In his statement, which has since been used as the Canadian standard for information disclosure to ensure adequate consent to human research, Justice Hall made clear: There can be no exceptions to the ordinary requirements of disclosure in the case of research as there may well be in ordinary medical practice. . . . The subject of medical experimentation is entitled to full and frank disclosure of all the facts, probabilities, and opinions which a reasonable man might be expected to consider before giving his consent.'

Cases 1 549 Case 3

Luka Magnotta and the Legal Recognition of Research Confidentiality In May 2012, Luka Magnotta was arrested and charged with the gruesome, videotaped murder of a Concordia University student in Montreal. Five years earlier, Magnotta had been a research participant in a University of Ottawa study of sex workers (Magnotta had been a male escort and pornography actor). A research assistant who had interviewed him told police that Magnotta had been interviewed for the study. Police then arranged to seize the transcript and audiotape of the interview to use it in evidence at Magnotta's murder trial, but they held it in a sealed package while the lead researchers on the study, professors Chris Bruckert and Colette Parent, applied to the court to rule that the interview was inadmissible. Bruckert and Parent argued that without the court's recognition that the confidentiality of such interviews was protected, it would be impossible to conduct research on sex workers or other individuals living at the margins of the law or beyond. The hearing into the admissibility of the interview was held in April 2013 by Quebec Superior Court Justice Sophie Bourque. In January 2014, she ruled that the interview could not be disclosed to the court on the grounds that the public interest in protecting research confidentiality outweighed the value of the interview to the trial. Justice Bourque's decision followed the "Wigmore criteria" to assess whether communications that form part of research are admissible in criminal proceedings. These criteria establish a framework for

assessing, on a case-by-case basis, whether research confidentiality is protected from breaches when a research participant's communications may be relevant to a criminal trial. The Wigmore criteria are: 1. The communications must originate in a confidence that they will not be disclosed. 2. This element of confidentiality must be essential to the full and satisfactory maintenance of the relation between the parties. 3. The relation must be one which in the opinion of the community ought to be sedulously fostered. 4. The injury that would inure to the relation by the disclosure of the communications must be greater than the benefit thereby gained for the correct disposal of litigation.' The court held that these criteria were met and the interview should not be disclosed, observing that its relevance was "minimal and at most marginal" to the trial. The decision recognizes researcher—participant confidentiality as a legally protected relationship in Canadian law However, it remains an interesting question whether the case-by-case framework using the Wigmore principles will sufficiently protect researcher— participant relationships in ways that will not threaten or undermine some types of socially important research affecting human health—for example, research into use of illegal recreational drugs, child and elder abuse, and risky or illegal sexual behaviour.

Note 1.

Parent & Bruchert v the Queen re Magnotta, Quebec Superior Court, 21 January 2014, http://citoyens.soquij .qc.ca/ID=7E4DBC4D25FC81DCECFB65F7FCA1BFE

Case 4

Jesse Gelsinger: Research Conflicts and Ethical Review At the age of two, Jesse Gelsinger was diagnosed with an X-chromosome-linked genetic disease affecting his liver's ability to metabolize ammonia, a by-product of protein breakdown. Jesse was able to survive the normally fatal disease on a restricted diet and special medications. In September of 1999

at the age of 18, Jesse participated in a clinical trial at the University of Pennsylvania aimed at developing treatment for infants born with a severe form of the disease. To test the safety of the procedure, on 13 September Jesse was injected with an adenoviral vector carrying a corrected version of the gene. He died four days later on 17 September, apparently of a massive immune response triggered by the viral vector leading to multiple organ failure and brain death. Jesse was the first person to have died in a clinical trial for gene therapy.

II

III

III JJ

9 Research with Humans

550

A subsequent FDA investigation concluded that the scientists involved in the study broke several rules of conduct, including:

•

•

Paul Gelsinger, Jesses father, had harsh words also for the ethicist involved, Arthur Caplan. Dr Caplan worked in the main researcher's department "effectively making the researcher his boss, another serious conflict."' The senior Gelsinger was disturbed by Dr Caplan's claim that "all the controversy created by Jesse's death was good for the ethics train and that, `we (bioethicists) thrive on scandal,'" claiming that this statement demonstrated a lack of good judgment.'

•

•

failure to disclose the deaths of monkeys given a similar treatment; failure to disclose that two previous patients had experienced serious side effects from the gene therapy, inclusion of Gelsinger as a substitute for a volunteer who opted out in spite of Gelsinger having elevated ammonia levels, which would typically have excluded his participation; and

failure to disclose a possible conflict of interest insofar as both the university and the primary researcher had financial stakes in the research.'

Note 1.

Paul Gelsinger, "Jesse's Intent," wwwjesse-gelsinger.com, 20 September 2011.

9.6 Study Questions 1. What is therapeutic misconception, and how can it be reflected in research participants' and researchers' attitudes toward health research? Why is it ethically problematic? What measures should be taken to avoid it? 2. In what ways is obtaining informed consent often more challenging in health research as compared to the provision of clinical care? You might want to divide your answers by considering how consent in health research puts burdens on each of the main components of a valid informed consent: decision-making capacity (competence), information disclosure, and voluntariness. 3. What conditions must be satisfied for there to be clinical equipoise? Must all researchers with relevant expertise be indifferent among the study arms or only the majority of them? Or is a significant minority enough; if so, what would that be? 4. Does the notion of clinical equipoise properly describe when a multi-arm study is ethically acceptable? As part of answering this question, consider whether researchers owe research participants the same duty of care owed to patients. 5. Identify as many examples as possible of conflicts of interest that appear to be commonly present in experimental clinical research. What can be done to eliminate or manage those conflicts? 6. In what circumstances should community consent be sought for research? What are some ethical pitfalls associated with community consent? 7. What are the ethical issues raised by doing research on vulnerable populations?

9.7 Suggested Further Reading Research Ethics Codes and Regulations Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, and Social Sciences and Humanities Research Council. 2014. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. http://www.pre.ethics .gc.ca/eng/policy-politique/initiatives/tcps2-eptc2/Default. • Note: The TCPS2 Core Tutorial is at http://www.pre.ethics.gc.ca/eng/education/ tutorial-didacticiel.

Suggested Further Reading

Council for the International Organizations of Medical Services. 2002. International Ethical Guidelines for Biomedical Research Involving Human Subjects. http://www.cioms.chi publications/layout_guide2002.pdf. International Conference on Harmonization. 1996. Guideline for Good Clinical Practice. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E6/ E6_Rl_Guideline.pdf. Nuremberg Military Tribunal. 1947. The Nuremberg Code, from U.S. v Karl Brandt et al. https://history.nih.gov/research/downloads/nuremberg.pdf. US Department of Health and Human Services, National Institutes of Health, and Office of Human Research Protections. 2009 (revised). The Common Rule, Title 45 (Public Welfare), Code of Federal Regulations, Part 46 (Protection of Human Subjects). http:// www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46. World Medical Association. 2013. Declaration of Helsinki. http://www.wma.net/en/30 publications/lOpolicies/b3. Research Ethics Textbooks Coleman, Carl H., Jerry A. Menikoff, Jess A. Goldner, and Nancy N. Dubler. 2015. The Ethics and Regulation of Research with Human Subjects. 2nd edn. New Providence, NJ: LexisNexis. Emanuel, Ezekiel J., Christine C. Grady, Robert A. Crouch, Reidar K. Lie, Franklin G. Miller, and David D. Wendler, eds. 2008. The Oxford Textbook of Clinical Research Ethics. New York: Oxford University Press. Emanuel, Ezekiel J., Robert A. Crouch, John D. Arras, Jonathan D. Moreno, and Christine Grady. 2003. Ethical and Regulatory Aspects of Clinical Research. Baltimore: Johns Hopkins University Press. Lavery, James V., Christine Grady, Elizabeth Wahl, and Ezekiel Emanuel, eds. 2007. Ethical Issues in International Biomedical Research: A Casebook. Oxford University Press. Past Abuses in Human Research Beecher, Henry K. 1966. "Ethics and Clinical Research." New England Journal of Medicine 274: 1354-60. Faden, Ruth R., ed. 1996. The Human Radiation Experiments: Final Report of the Advisory Committee. New York: Oxford University Press. Goodman, Jordan, Anthony McElligott, and Lara Marks, eds. 2008. Useful Bodies: Humans in the Service of Medical Science in the Twentieth Century. Baltimore: Johns Hopkins University Press. Harris, Sheldon H. 1994. Factories of Death: Japanese Biological Warfare 1932-45 and the American Cover Up. New York: Routledge. Hodge, Felicia Schanche. 2012. "No Meaningful Apology for American Indian Unethical Research Abuses." Ethics & Behaviour 22 (6): 431-44. Jones, David S., Christine Grady, and Susan E. Lederer. 2016. "Ethics and Clinical Research—The 50th Anniversary of Beecher's Bombshell." New England Journal of Medicine 374: 2393-8. Mosbey, Ian. 2013. "Administering Colonial Science: Nutrition Research and Human Biomedical Experimentation in Aboriginal Communities and Residential Schools, 1942-1952. Social History 46 (91): 145-72. O'Mathnna, DOnal P 2006. "Human Dignity in the Nazi Era: Implications for Contemporary Bioethics." BMC Medical Ethics 7 (2). Pence, Gregory E. 1995. "The Tuskegee Study," in Classic Cases in Medical Ethics, 2nd edn. New York: McGraw Hill. Shushman, Miriam. 2005. The Drug Trial: Nancy Olivieri and the Science Scandal that Rocked the Hospital for Sick Children. Toronto: Random House Canada.

II Ail

111.thl

9 Research with Humans

Weijer, Charles. 1995. "The Breast Cancer Research Scandal." Canadian Medical Association Journal 152 (8): 1195-7. Welsome, Eileen. 1999. The Plutonium Files: America's Secret Medical Experiments in the Cold War. New York: Dial Press. Topics in Human Research Ethics Appelbaum, Paul S., Loren H. Roth, and Charles Lidz. 1982. "The Therapeutic Misconception: Informed Consent in Psychiatric Research." International Journal of Law and Psychiatry 5 (3-4): 319-29. Friedman, Benjamin. 1987. "Equipoise and the Ethics of Clinical Research." New England Journal of Medicine 317(3): 141-5. Henderson, Gail, et al. 2007. "Clinical Trials and Medical Care: Defining the Therapeutic Misconception." PLoS Med 4 (11). http://journals.plos.org/plosmedicine/article?id=10.1371/ journal.pmed.0040324. Heyd, David. 1996. "Experimentation on Trial: Why Should One Take Part in Medical Research?" Jahrbuk fur Recht and Ethik 4: 189-204. Kopelman, Loretta M. 2004. "Minimal Risk as an International Ethical Standard in Research." Journal of Medicine and Philosophy 29 (3): 351-78. Lewis, Steven, Patricia Baird, Robert G. Evans, William A. Ghali, Charles J. Wright, Elaine Gibson, and Francoise Baylis. 2001. "Dancing with the Porcupine: Rules for Governing University-Industry Relationship." Canadian Medical Association Journal 165 (5): 783-5. McGuire, Amy, Timothy Caulfield, and Mildred K. Cho. 2008. "Research Ethics and the Challenge of Whole Genome Sequencing." Nature Reviews Genetics 9: 152-6. Miller, Franklin D., and Howard Brody. 2002. "What Makes Placebo-Controlled Trials Unethical? American Journal of Bioethics 2 (2): 3-9. Palys, Ted, and David MacAlister. 2016. "Protecting Research Confidentiality via the Wigmore Criteria: Some Implications of Parent and Bruckert v The Queen and Luca Rocco Magnotta." Canadian Journal of Law and Society 31 (3): 473-93. Palys, Ted, and John Lowman 2014. Protecting Research Confidentiality: What Happens When Law and Ethics Collide. Toronto: James Lorimer and Company. Ross, Lainie Freidman. 2006. Children in Medical Research: Access versus Protection. Oxford University Press. Shalowitz, David I., and Franklin G. Miller. 2008. "Communicating the Results of Clinical Research to Participants: Attitudes, Practices, and Future Directions. PLoS Medicine. http://journals.plos.org/plosmedicine/article?id=10.1371/journal. pmed.0050091. Weijer, Charles. 1999. "Placebo-Controlled Trials in Schizophrenia: Are They Ethical? Are They Necessary? Schizophrenia Research 35 (3): 211-18. Wendler, David. 2006. "Assent in Paediatric Research: Theoretical and Practical Considerations." Journal of Medical Ethics 32: 229-34. , Leah Belsky, Kimberly M. Thompson, and Ezekiel J. Emanuel. 2005. "Quantifying the Federal Minimal Risk Standard Implications for Pediatric Research without a Prospect of a Benefit." Journal of the American Medical Association 294 (7): 826-31. Wendler, David, and Franklin Miller. 2004. "Deception in the Pursuit of Science. Archives of Internal Medicine 164 (6): 597-600. Animal Research DeGrazia, David. 2002. Animal Rights: A Very Short Introduction. New York: Oxford University Press. Holms, Helen Bequaert. 1989. "Can Clinical Research Be Both Ethical and Scientific?" Hypatia 4 (Summer): 154-65.

GDossary abortion termination of pregnancy. active euthanasia administering a lethal agent to another person in order to kill the person. advance directive a legal document made by persons with decision-making capacity to guide medical decisions when they lack decision-making capacity. age-based rationing rationing based on age and age alone. As used in the literature, it means that if an elderly person and a non-elderly person each has essentially the same medical problem, needs life-preserving treatment, and has the same prognosis, the treatment can be denied the elderly person on the basis of age. applied ethics the application of moral theories, principles, and ideas to specific moral problems. assent a right that incompetent persons can have to accept or refuse treatment or participation in research. It is to be contrasted with a competent person's right to consent. assisted suicide providing a person with the means of death. beneficence prevent harm, reduce harm, and do good. best-interest judgment the judgment that the reasonable person in the patient's position would make. care ethics morality is fundamentally about caring, fostering positive relationships, and related concepts. categorical imperative or the supreme principle of the categorical imperative the central concept in Kant's ethics; given several formulations.

clinical equipoise the state of genuine uncertainty within the expert community about which study arm is therapeutically superior in a multi-arm study of an experimental or existing therapy. club good a good that is not rival (in consumption) but is excludable. collective action problem any failure of a group of individuals to achieve a result (for example, the production of a good) that everyone values at least as much as the result of everyone acting in their own self-interest. commercial surrogacy payment to a woman beyond reimbursement of expenses for agreeing to carry a pregnancy for another person or couple who will become the child's parent(s). common-pool resource good a good that is rival (in consumption) but is not excludable. conflicts of interest situations in which a person has an obligation to a person or an institution that is in opposition to another interest or obligation, which could corrupt the decision-making of that person. conscientious objection/refusal when health care providers refuse on moral grounds to provide a treatment that is permitted by the standards of their profession. consequentialism rightness is a function of promoting good consequences. Conservative Theory the fetus has a right to life from the point of conception onwards. cultural relativism what is morally right is determined by whatever moral standards a culture endorses.

■m I

11

.111

111111

Glossary

decision-making capacity individuals can understand the nature and consequences of their decisions and are not subject to distorting mental illness or mental states. Individuals who have these abilities are also referred to as capable or competent. deontology rightness is not exclusively, and perhaps not at all, a function of promoting good consequences. determinism every event is brought about and necessitated by prior causes that occur according to the laws of nature. distributive justice the fair allocation of scarce resources. direct or proximate cause the cause that is closest to the event to be explained. doctrine of double effect (DDE) the principle that a proposed action that will have benefits but will also cause some harm is permissible if the action itself is morally appropriate and the foreseen harm is not intended. egalitarian theory of justice every person is equally valuable, and society has an obligation to provide them with the basic resources for a good life. euthanasia bringing about the death of a person in that person's interest. excludable good a good that people can be prevented from benefitting from without great cost. A non-excludable good is one where people can be stopped from consuming it only at great cost. experimental (or interventional) research attempts to advance knowledge by studying the effects of research interventions on research participants. false negative A test result that is negative but incorrect. A test result is negative when the test says a patient does not have the illness being tested for or a predictor for it. false positive a test result that is positive but incorrect. A test result is positive when the test says a patient has the illness being tested for or a predictor for it.

fetus (as used in biology) a pre-natal organism in which all organ systems are in place and functioning (approx. 12-36 weeks old). free rider a person who benefits from or consumes a good without contributing sufficiently to its production or continued existence. gatekeeping the role played by health care providers in determining who has access to health care services. genetic exceptionalism the view that genetic information is a special type of medical information and must therefore be treated with greater care—for example, with special laws—than other health care information. harm principle interference in the conduct of competent persons is only justified to prevent harm to others. Hawthorne Effect the alteration of research participants' behaviour as a result of knowing that they are being observed by researchers, typically by improving their behaviour. health region a geographical area defined for administrative purposes by a provincial or territorial ministry of health. Hippocratic Oath this ancient Greek text requires a new physician to swear to the healing gods to follow specific ethical standards. indirect or remote cause a cause that causes the proximate cause. infanticide killing an infant. informed decision a decision made when a person knows everything he or she would want to know before making the decision. instructional advance directive an advance directive in which persons designate what medical decisions they would or would not want made when they lack decisionmaking capacity.

feticide killing the fetus.

Liberal Theory the fetus lacks a right to life at all stages of fetal development.

fetus (as generally used in philosophic discussions of abortion) a human organism at any stage of fetal development from zygote to (but not including) infant.

liberal theory of justice a social view that combines egalitarian and libertarian elements. In the context of resource allocation, it means two-tier health care.

Glossary libertarian theory of justice every person has a right to private property and to enter into any agreements they want to. Government can interfere only to prevent force or fraud. luck egalitarianism individuals are responsible for consequences if and only if they are the result of their voluntary choices.

normative ethics the inquiry into the standards that determine how to act morally and lead a moral life. observational research attempts to advance knowledge by collecting data from research participants without making any change to their care, treatment, or lives in relation to what is being studied.

macroallocation the fair allocation of a society's resources for health care under conditions of scarcity.

overdiagnosis a correct diagnosis of illness (for example, cancer) where the illness will do no harm to the person during that person's lifetime.

MAID assisted suicide or active euthanasia performed by a physician or a nurse practitioner, as per amendments to the Criminal Code of Canada in 2016.

overtreatment medical treatment for an illness given when the illness will do no harm to the person during that person's lifetime.

maxim Kant's name for a personal rule for acting in a certain way. On some ideas about what Kant means, this must include a statement of the goal of acting and the kind of action that will achieve the goal.

PAD physician-assisted suicide or active euthanasia performed by a physician.

mesoallocation the fair allocation of scarce health care resources within a specific health care region or institution. metaethics the inquiry into the nature and status of morality. microallocation the fair allocation of scarce resources to specific individuals. Middle Theory the fetus lacks a right to life up to some point or threshold stage in fetal development and has a right to life after that time. moral realism moral facts exist independently of the evidence for them and about which we can have at least approximate knowledge. natural rights rights that all human beings have just in virtue of being human and which are not subject to political control. negative right A right not to be stopped by some institution or person or group from getting something (a good or benefit). It corresponds to a negative duty on the part of that institution or person or group not to interfere with those with the right from getting the good or benefit.

passive euthanasia withholding or withdrawing lifesustaining treatment (WEST). paternalism interfering with someone for their own good. physician-assisted suicide a physician provides a patient with the means of death. placebo a treatment all of whose effects are placebo effects; placebo effects are a good or bad effect of a treatment that occurs because the recipient believes the treatment will have the effect. positive right a right to be provided with something (a good or benefit) by some institution or person or group. It corresponds to a positive duty on the part of that institution or person or group to provide the good or benefit to those with the right. primary goods general all-purpose means, such as liberty, opportunities, education, and health, necessary for the pursuit of life goals. principles of biomedical ethics non-maleficence, beneficence, autonomy, and justice.

non-maleficence do not cause harm.

professional autonomy roughly, health care providers' right to say "no" when asked to act in ways that conflict with their professional obligations.

non-moral facts states of affairs that can be wholly or adequately described separately from moral values. For example, biological, psychological, social, physical facts.

proxy directive an advance directive in which persons designate who is to make decisions for them when they lack decision-making capacity.

II

556

.111

III 1.6.1

Glossary randomized, double-blind, placebo-controlled trial a study in which research participants are randomly assigned to either an experimental treatment or a placebo and neither they nor the researchers know what they are receiving. rationing addressing scarcity of health care resources by eliminating services or treatments offered, or excluding classes of individuals from health services or treatments, or prioritizing access. rights moral claims with special normative, apparently non-consequentialist, force. rival (or rivalrous) good a good where one person's benefit reduces or eliminates a benefit to others. In a non-rival good, one persons benefit or consumption does not affect another person's. screening test a medical test that looks for predictors or the early stages of illness when a person shows no symptoms.

substitute decision-maker a person who makes decisions for a patient when the patient lacks decision-making capacity. A substitute decision-maker is sometimes also called a surrogate decision-maker or (when the person has been legally designated as the decision-maker by the patient when capable) a proxy decision-maker. substituted judgment the judgment that a patient would make for himself or herself. terminal sedation the use of high doses of pain medication to treat a patients suffering at the end of life, with the awareness that such high dosages will hasten the patient's death. therapeutic misconception the failure to understand that health research is aimed primarily at producing knowledge and is not conducted specifically for the sake of research participants' health. therapeutic privilege the idea that a medical professional may deceive or withhold the truth from a patient when providing the truth will produce harm.

sentience the ability to have experiences. slippery slope a situation that is likely to lead to an undesirable consequence, whether or not one tries to prevent that. slippery slope argument an argument against allowing a situation to occur because it is likely to lead to an undesirable situation. social determinants of health the social and economic conditions in which people are born, grow, live, work, and age that have effects on their health. strong paternalism it is permissible to interfere with a competent or incompetent person's decision about treatment if it is in their best interest. study arm participants in clinical trials are randomly assigned to different groups (arms) receiving different interventions to compare the effectiveness and safety of those interventions. subjectivism morality is fundamentally about the expression of feeling or emotion.

tragedy of the commons when each individual seeks to maximize personal benefit from a scarce resource so that demand overwhelms supply and the resource is no longer available to some or all. triage the assignment of degrees of urgency to wounds, injuries, or illnesses to decide the order of treatment of a large number of patients under conditions of scarcity of medical resources. utilitarianism right actions are those that maximize overall happiness or well-being. virtue ethics morality is principally a matter of realizing in action character traits (virtues) that express being a good person. voluntary decision a decision made when a person is not subject to undue influences. weak paternalism it not permissible to interfere with a competent person's informed decision to accept or refuse an offered treatment.

Index A Third Way (Sumner), 262 Abortion of a Fetus Due to Diagnosis of Down Syndrome (case), 284 Abortion of Suspected Female Fetus (case), 283 Abortion through a Feminist Ethics Lens (Sherwin), 268 Active and Passive Euthanasia (Rachels), 309 Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures (Canadian Medical Association), 76 Allocation of Exotic Medical Life-Saving Therapy (Rescher), 391 American Medical Association, Deciding to Forego LifeSustaining Treatment, 82 Arndt v Smith (case), 154 Avard, D., Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies, 537 BC Physicians "Cherry-Picking" Patients for Ease of Care, Refusing Those in Need (case), 207 Beauchamp, T.L., The Concept of Informed Consent, 47 Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies (Avard et al.), 537 Beyerstein, D., The Functions and Limitations of Professional Codes of Ethics, 201 Bioethics Committee, Position Statement: Treatment Decisions Regarding Infants, Children, and Adolescents, 80 Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis—Learning from the Cases of Nancy Olivieri and David Healy (Schafer), 489 Black, L., Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies, 537 Boyle, J.M., Medical Ethics and Double Effect: The Case of Terminal Sedation, 316 Breaching Confidentiality (Unger), 128 Breast Cancer Screening (case), 473 Brett, A.S., Problem with Futility, 176 British Columbia Supreme Court, Controlling the Risks of PAD, 321

Brock, D.W., Standards of Competence, 32; Voluntary Active Euthanasia, 297 Brody, H., Transparency: Informed Consent in Primary Care, 51; Critique of Clinical Equipoise: Therapeutic Misconception in the Ethics of Clinical Trials, 509 Browne, A., Defining Death, 339 Browne, K., Measles and Free Riders: California's Mandatory Vaccination Law, 427 Buchanan, A.E., Standards of Competence, 32; Right to a Decent Minimum of Health Care, 367 Buying a Kidney in India but Requesting Canadian After-Care (case), 413 Callahan, D., When Self-Determination Runs Amok, 312 Canada's Highest Court Unchains Injection Drug Users; Implications for Harm Reduction as Standard of Healthcare (Small), 443 Canadian Medical Association, Advance Directives for Resuscitation and Other Life-Saving or Life-Sustaining Measures, 76 Canadian Medical Association, Caring in a Crisis: The Ethical Obligations of Physicians and Society during a Pandemic, 188 Canadian Paediatric Society, Position Statement Treatment Decisions Regarding Infants, Children, and Adolescents, 80 Caring in a Crisis: The Ethical Obligations of Physicians and Society during a Pandemic (Canadian Medical Association), 188 Caulfield, T., Direct-to-Consumer Genetics and Health Policy: A Worst-Case Scenario?, 434 Chadwick, R., Ethics of Screening: Is "Screeningitis" an Incurable Disease?, 436 Christie, T., Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate, 451 Concept of Informed Consent (Faden & Beauchamp), 47 Consent and Public Health (O'Neill), 60 Controlling the Risks of PAD (British Columbia Supreme Court), 321 Crash of Germanwings Flight 9525 (case), 155 Critique of Clinical Equipoise: Therapeutic Misconception in the Ethics of Clinical Trials (Miller & Brody), 509

II III

1111 l

Index Culture, Power, and Informed Consent: The Impact of Aboriginal Health Interpreters on Decision-Making (Kaufert & O'Neil), 56 Dalhousie University, Withholding and Withdrawal of Potentially Life-Sustaining Treatment, 295 Daniels, N., When Are Health Inequalities Unjust?—The Social Determinants of Health, 458 Deciding to Forego Life-Sustaining Treatment (Judicial Council & American Medical Association), 82 Defence of Abortion (Thomson), 253 Defense of Unqualified Medical Confidentiality (Kipnis), 117 Defining Death (Browne), 339 Department of Justice, Legislative Background: Medical Assistance in Dying (Bill C-14), 323 Direct-to-Consumer Genetics and Health Policy: A WorstCase Scenario? (Caulfield), 434 Dissent on Judicial Intervention in Pregnancy and Birth (Royal Commission on New Reproductive Technologies) (Scorsone), 281 Do Everything for Mom: Advance Directives and a Surrogate's Right to Demand Treatment (case), 93 Does Clifford Olson Deserve Medical Treatment? (case), 414 Dr Ewen Cameron (Ross), 485 Dr Nancy Morrison: Nonvoluntary Active Euthanasia of an Adult (case), 342 Dr Roger Poisson: Fraud in Breast Cancer Study (case), 547 Dworkin, R., Justice and the High Cost of Health, 372 Elizabeth and Eric MacDonald: Assisted Suicide (case,) 345 Emanuel, E.J., An Ethical Framework for Biomedical Research, 495; Protecting Communities in Biomedical Research, 518 Endangering Behaviour in a Pregnant Woman (case), 284 Enough: The Failure of the Living Will (Fagerlin & Schneider), 63 Ethical Framework for Biomedical Research (Emanuel, Wendler, & Grady), 495 Ethical Relativism in a Multicultural Society (Macklin), 83 Ethics of Screening: Is "Screeningitis" an Incurable Disease? (Shickle & Chadwick), 436 Faden, R.R., The Concept of Informed Consent, 47 Fagerlin, A., Enough: The Failure of the Living Will, 63 Fleck, L.M., Just Caring: In Defense of Limited Age-Based Healthcare Rationing, 399 Frader, J., Problem with Futility, 176 Free Riding and Organ Donation (Glannon), 431 Functions and Limitations of Professional Codes of Ethics (Beyerstein), 201 Genetic Exceptionalism and "Future Diaries": Is Genetic Information Different from Other Medical Information? (Murray), 147 George VIs Cancer (case), 153 Glannon, W, Free Riding and Organ Donation, 431 Grady, C., An Ethical Framework for Biomedical Research, 495

Griener, G., Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies, 537 Groarke, L., Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate, 451 Harrison, C., Involving Children in Medical Decisions, 76; Position Statement: Treatment Decisions Regarding Infants, Children, and Adolescents, 80 Hassan Rasouli and Medical Futility (case), 205 Having an Estate Sale of One's Organs (case), 413 Health Care Workers and Flu Shots (case), 471 Health Law Institute, Withholding and Withdrawal of Potentially Life-Sustaining Treatment, 295 Helping Residents to Live at Risk (case), 208 Hope, T., Why Undervaluing "Statistical" People Costs Lives, 377 How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales (Kaplan), 134 Human Organs, Scarcities, and Sale: Morality Revisited (Kishore), 407 Hurst, S.A., Vulnerability in Research and Health Care; Describing the Elephant in the Room?, 526 Involving Children in Medical Decisions (Harrison et al.), 76 Is It Ever Right to Violate Rules of Rationing? (case), 207 Jesse Gelsinger: Research Conflicts and Ethical Review (case), 549 Judicial Council, Deciding to Forego Life-Sustaining Treatment, 82 Judicial Intervention in Pregnancy and Birth (Royal Commission on New Reproductive Technologies), 277 Just Caring: In Defense of Limited Age-Based Healthcare Rationing (Fleck), 399 Justice and the High Cost of Health (Dworkin), 372 Kant, I., On the Supposed Right to Lie from Benevolent Motives, 106 Kaplan, B., How Should Health Data Be Used? Privacy, Secondary Use, and Big Data Sales, 134 Kaufert, J., Culture, Power, and Informed Consent: The Impact of Aboriginal Health Interpreters on Decision-Making, 56 Kenny, N.E. Involving Children in Medical Decisions, 76 Kipnis, K., A Defense of Unqualified Medical Confidentiality, 117 Kishore, R. R., Human Organs, Scarcities, and Sale: Morality Revisited, 407 Knoppers, B.M., Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies, 537 Legislative Background: Medical Assistance in Dying (Bill C-14) (Department of Justice), 323 Luka Magnotta and the Legal Recognition of Research Confidentiality (case), 549

Index 559

Macklin, R., Ethical Relativism in a Multicultural Society, 83 Markayla Sault: Caring for an Aboriginal Patient (case), 95 Marquis, D., Why Abortion Is Immora,1238 Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection (Provincial-Territorial Expert Advisory Group on Physician-Assisted Dying and the Special Joint Committee on Physician-Assisted Dying), 328 Measles and Free Riders: California's Mandatory Vaccination Law (Browne), 427 Medical Ethics and Double Effect: The Case of Terminal Sedation (Boyle), 316 Medical Secrecy: Patients' Right of Access to Medical Records (Schafer), 113 Miller, F.G., A Critique of Clinical Equipoise: Therapeutic Misconception in the Ethics of Clinical Trials, 509 Mr Halushka: Human Research and Harm to Participants (case), 548 Mr McCullough: Recommending Voluntary Passive Euthanasia (case), 344 Ms G and Refusal of Treatment while Pregnant (case), 283 Murray, T.H., Genetic Exceptionalism and "Future Diaries": Is Genetic Information Different from Other Medical Information?, 147 Nancy B.: Withdrawing Life-Sustaining Treatment (case), 341 No Chemotherapy for Anael: Surrogate Refusal of Treatment for a Minor Child (case), 92 Nuts and Bolts of Obtaining Consent to Treatment (Rozovsky & Rozovsky ), 45 O'Neil, J., Culture, Power, and Informed Consent: The Impact of Aboriginal Health Interpreters on DecisionMaking, 56 O'Neill, 0., Consent and Public Health, 60 On the Moral and Legal Status of Abortion (Warren), 248 On the Supposed Right to Lie from Benevolent Motives (Kant), 106 Optimal Care versus Cost Containment: What Is a Doctor to Do? (case), 412 Overall, C., Reproductive Freedom, Autonomy, and Reproductive Rights, 218 Patient and Physician Autonomy: Conflicting Rights and Obligations in the Physician—Patient Relationship (Pellegrino), 165 Pellegrino, E.D., Patient and Physician Autonomy: Conflicting Rights and Obligations in the Physician— Patient Relationship, 165 Personal Genome Mapping (case), 156 Position Statement: Treatment Decisions Regarding Infants, Children, and Adolescents (Harrison et al.), 80 Preconception Arrangements (Royal Commission on New Reproductive Technologies), 233 Preconception Gender Selection (Robertson), 226

Prenatal Diagnosis and Abortion or Infanticide through Declining Treatment (case), 282 Problem with Futility (Truog, Brett, & Frader), 176 Protecting Communities in Biomedical Research (Weijer & Emanuel), 518 Provincial-Territorial Expert Advisory Group on PhysicianAssisted Dying and the Special Joint Committee on Physician-Assisted Dying, Mature Minors, Mental Illness, Advance Directives, and Conscientious Objection, 328 QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference Do They Make for Health Care Priority Setting? (Robberstad), 382 Rachels, J., Active and Passive Euthanasia, 309 Rationing Services to an Elder Who Is Responsible for His Medical Condition (case), 413 Relational Approach to Autonomy in Health Care (Sherwin), 35 Repairing Research Integrity (Titus, Wells, & Rhoades), 523 Reproductive Freedom, Autonomy, and Reproductive Rights (Overall), 218 Rescher, N.P., Allocation of Exotic Medical Life-Saving Therapy, 391 Rhoades, 14, Repairing Research Integrity, 523 Right to a Decent Minimum of Health Care (Buchanan), 367 Robberstad, B., QALYs vs DALYs vs LYs Gained: What Are the Differences, and What Difference Do They Make for Health Care Priority Setting?, 382 Robertson, J.A., Preconception Gender Selection, 226 Romanow, R., Sustaining Medicare: The Commission on the Future of Health Care in Canada, 357 Ross, C.A., Dr Ewen Cameron, 485 Rowell, M., Involving Children in Medical Decisions, 76 Royal Commission on New Reproductive Technologies, Judicial Intervention in Pregnancy and Birth, 277 Royal College of Paediatrics and Child Health, Withholding or Withdrawing Life-Sustaining Treatment in Children, 81 Royal Commission on New Reproductive Technologies, Preconception Arrangements, 233 Rozovsky, F.A., The Nuts and Bolts of Obtaining Consent to Treatment, 45 Rozovsky, L.E., The Nuts and Bolts of Obtaining Consent to Treatment, 45 Samuel, J., Best Practices for Health Research Involving Children and Adolescents: Genetic, Pharmaceutical, and Longitudinal Studies, 537 SARS in Toronto: A Duty to Care (case), 206 Schafer, A., Medical Secrecy: Patients' Right of Access to Medical Records, 113; Biomedical Conflicts of Interest: A Defence of the Sequestration Thesis— Learning from the Cases of Nancy Olivieri and David Healy, 489 Schneider, C.E., Enough: The Failure of the Living Will, 63

II

III

III

Index Schuklnek, U., Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies, 191 Scorsone, S.R., Dissent on Judicial Intervention in Pregnancy and Birth (Royal Commission on New Reproductive Technologies), 281 Scott Starson: Refusing Treatment while Incompetent (case), 91 Sex Selection in Canada (case), 282 Sherwin, S., A Relational Approach to Autonomy in Health Care 35, Abortion through a Feminist Ethics Lens, 268 Shickle, D., Ethics of Screening: Is "Screeningitis" an Incurable Disease?, 436 Should Physicians Be Gatekeepers of Medical Resources? (Weinstein), 181 Sidarous, M., Involving Children in Medical Decisions, 76 Small, D., Canada's Highest Court Unchains Injection Drug Users; Implications for Harm Reduction as Standard of Healthcare, 443 Smalling, R., Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies, 191 Standards of Competence (Buchanan & Brock), 32 Stewart, R.S., Telling Patients the Truth, 110 Students and Study-Enhancing Drugs (case), 472 Sumner, L.W., A Third Way, 262 Sustaining Medicare: The Commission on the Future of Health Care in Canada (Romanow), 357 Sweet. W., Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate, 451 Telling Patients the Truth (Stewart), 110 Telling the Truth to Patients: A Clinical Ethics Exploration (Thomasma), 107 Thomasma, D.C., Telling the Truth to Patients: A Clinical Ethics Exploration, 107 Thomson, J.J., A Defence of Abortion, 253 Titus, S.L., Repairing Research Integrity, 523 Tracy Latimer: Nonvoluntary Active Euthanasia of a Minor (case), 343

Transparency: Informed Consent in Primary Care (Brody), 51 Treatment for Jehovah's Witnesses: Adults and Children (case), 94 Truog, R.D., Problem with Futility, 176 Unger, D., Breaching Confidentiality, 128 Virtue Ethics as an Alternative to Deontological and Consequential Reasoning in the Harm Reduction Debate (Christie, Groarke, & Sweet), 451 Voluntary Active Euthanasia (Brock), 297 Vulnerability in Research and Health Care; Describing the Elephant in the Room? (Hurst), 526 Warren, M.A., On the Moral and Legal Status of Abortion, 248 Water in First Nations Communities (case), 474 Weijer, C., Protecting Communities in Biomedical Research, 518 Weinstein, M.C., Should Physicians Be Gatekeepers of Medical Resources?, 181 Wells, J.A., Repairing Research Integrity, 523 Wendler, D., An Ethical Framework for Biomedical Research, 495 When Are Health Inequalities Unjust?—The Social Determinants of Health (Daniels), 458 When Is Treatment Futile? (case), 209 When Self-Determination Runs Amok (Callahan), 312 Why Abortion Is Immoral (Marquis), 238 Why Medical Professionals Have No Moral Claim to Conscientious Objection Accommodation in Liberal Democracies (Schuklnek & Smalling), 191 Why Undervaluing "Statistical" People Costs Lives (Hope), 377 Withholding and Withdrawal of Potentially Life-Sustaining Treatment (Health Law Institute & Dalhousie University), 295 Withholding or Withdrawing Life-Sustaining Treatment in Children (Royal College of Paediatrics and Child Health), 81

"A very useful collection of important classic and contemporary readings on a variety of important biomedical ethics issues that are relevant to Canadian students." —Jeff McLaughlin, Thompson Rivers University "The theoretical foundation this text provides is essential to teaching a good course in medical ethics."

—Melany Banks, Wilfrid Laurier University

HIGHLIGHTS •

Completely revised chapter introductions give students more robust, updated discussions of the concepts and issues arising in the readings.

•

Fully updated, this third edition provides students with current information and coverage of recent developments in the field, including a discussion of the new Canadian legislation on physician-assisted dying.

•

New chapters on public health and professional ethics introduce students to these two important areas of biomedical ethics study.

•

An ideal mix of classic and contemporary readings—with forty-four readings new to this edition—introduces students to foundational readings alongside the most recent developments in biomedical ethics.

•

Engaging case studies, including nineteen new to this edition, start students thinking about the possibilities and ramifications of bioethical decision-making.

About the Editors The late JOHNNA FISHER was an instructor in the Department of Philosophy at the University of British Columbia and Langara College. J.S. RUSSELL is a member and past chair of the Department of Philosophy at Langara College. ALISTER BROW NE is a faculty emeritus instructor and former chair of the Philosophy Department at Langara College. LESLIE BURKHOLDER is a senior instructor at the University of British Columbia.

www.oupcanada.com

SBN-13: 978-0- 9-902228-

OXFORD UNIVERSITY PRESS

MI 0

9 780199'022281